Part 1. Bitvise SSH Server Users' Guide
Installing Bitvise SSH Server
To install Bitvise SSH Server, execute the installer that you downloaded from our website and follow the process. As soon as the installer completes, you will have a working SSH server installation on your machine. No changes to the default configuration are necessary, you only need to start it. However, see also Configuring groups and accounts - this will help you restrict your users' access to those features that they actually need, which will improve security.
It is also possible to install Bitvise SSH Server in unattended mode, using command-line parameters to the installer. If you wish to make use of this feature, execute the following for a list of supported command-line parameters:
For example, if you wished to install Bitvise SSH Server on a fresh machine and start it immediately afterwards, you might execute the following:
BvSshServer-Inst.exe -defaultInstance -activationCode=0123...6789
net start BvSshServer
To apply a ready-made configuration file as part of the installation process, add the '-settings=...' parameter and specify the file from which Bitvise SSH Server should load its configuration. This file can be created by exporting the settings of an existing installation from its Bitvise SSH Server Control Panel.
To uninstall from the command line, copy the uninst.exe program in the Bitvise SSH Server installation directory to another temporary location, and execute it from there as follows:
uninst.exe "Bitvise SSH Server" -unat
If you are installing a single Bitvise SSH Server installation, you should of course simply execute the installer (without parameters) and follow the instructions provided by the user interface.
Upgrading from a previous version
Upgrading. If you have an older version of Bitvise SSH Server (WinSSHD) and wish to upgrade to the latest one, download the new installer from our website and execute it on the machine where your previous SSH server version is installed. The installer will detect an existing installation and will automatically remove it before installing the new one. During this process, your server keypair and settings will be preserved.
In case you must downgrade. During a downgrade, your SSH server settings will be reset. However, recent Bitvise SSH Server versions automatically create backups of your settings when they are modified, and you can revert to one of these backups if you decide to downgrade. The backups are located in the Config subdirectory of your SSH server installation directory.
Older WinSSHD versions (e.g. 4.xx) do not maintain backups. Before upgrading from such a version, you should use the WinSSHD Control Panel to export your settings to a file, in case you later decide to downgrade.
Activation. Any existing activation code you have will work for the new version only if the new version was released prior to the 'upgrade expiry' date embedded in the activation code. If your existing activation code is not valid for the upgraded-to version, the new version will install without a hitch, but will drop into time-limited evaluation. If your upgrade access has expired, log into your License Overview to purchase an upgrade extension, and acquire a new activation code.
If the purchasing process in your organization is slow, we do recommend that you initiate the upgrade extension process well before you plan to upgrade to a version not covered by your current upgrade access.
It is possible to upgrade Bitvise SSH Server in unattended mode, without having to explicitly remove the previous version. This can be done by executing the Bitvise SSH Server installer with command line parameters:
BvSshServer-Inst.exe -installDir="C:\Program Files\Bitvise SSH Server"
BvSshServer-Inst.exe -site="Bitvise SSH Server"
You will also need to supply the -acceptEula parameter to indicate acceptance of the Bitvise SSH Server End User License Agreement.
It is possible to use the installer for unattended installation to a named site. In this case, use the '-site' parameter (instead of '-installDir') and specify the name of the site desired.
In the latest Bitvise SSH Server versions, the installer will automatically stop the SSH service and Control Panel if they were running prior to the upgrade. In older WinSSHD versions, it may be necessary to execute a "net stop WinSSHD" command before the installer (and close the WinSSHD Control Panel, if open).
After a successful upgrade, the command "net start BvSshServer" can be executed to start the SSH server.
Upgrading Bitvise SSH Server when it provides exclusive access
Sometimes, Bitvise SSH Server is installed on machines where it provides exclusive access to those machines, and no other ports are open. In such situations, bringing the SSH server down for maintenance or an upgrade can render the machine inaccessible if something goes wrong.
In such situations, we recommend installing an additional SSH server installation as a named site. The additional SSH server installation should accept connections on a different port than the primary installation. This port should be accessible through any routers and firewalls. When maintenance or upgrade is needed on the primary installation, access the server through the alternate installation, and vice versa.
Note that multiple Bitvise SSH Server installations running directly on the same OS installation do not constitute an additional machine, and are covered by the same license. Therefore, no additional purchase is necessary for the maintenance installation.
Starting Bitvise SSH Server and monitoring activity
Bitvise SSH Server can be started and stopped in the following ways:
- using the Bitvise SSH Server Control Panel;
- from Administrative Tools > Services;
- from the Windows command prompt using commands 'net start BvSshServer' and 'net stop BvSshServer';
- the service can also be stopped remotely, using the Bitvise SSH Server Remote Control Panel, in Bitvise SSH Client.
The Bitvise SSH Server Control Panel features a Session tab, which shows SSH sessions currently active on the server.
Since version 5.06, the Bitvise SSH Server Control Panel also features an Activity tab, which shows a history of recent events on the SSH server, such as logins, disconnects, or file transfers. When the SSH Server Control Panel is open or minimized, it can also be configured to show pop-up notifications for events that show up in the Activity tab.
The Session and Activity tabs are intended to provide a casual overview of SSH server activity, but not a thorough overview. For a thorough overview or diagnostics, consult Bitvise SSH Server log files.
When Bitvise SSH Server is running, its default logging behavior in versions 4 and 5 is as follows:
Errors and warnings are logged to the Application section of the Windows Event Log. Many applications write their log messages to this log; to reduce clutter, Bitvise SSH Server records no info messages here by default.
You can investigate the Event log either through Administrative Tools > Event Viewer, or using the Bitvise SSH Server Control Panel, by clicking the 'View Windows Event Log...' button.
If your Application log becomes full, Windows may stop logging new events until the log is cleared or its size restrictions are loosened. You can investigate the maximum size settings of your Application log by selecting it in the Windows Event Viewer and clicking Action > Properties. The log can be cleared either from the Event Viewer or through the Bitvise SSH Server Control Panel.
- Errors, warnings and informational messages are logged into managed textual log files, created by default in the Logs subdirectory of the SSH server installation directory. These log files are stamped with the date and time of their creation and can be viewed with any text editing software, e.g. Notepad.
The logging level for each of the two destinations (log files or Windows Event Log) can be changed in Bitvise SSH Server Advanced settings. For security reasons, we recommend that you log errors and warnings at least. You should inspect the log periodically to make sure that everything is running as expected. For storage reasons, we recommend not setting the log level higher than info, except temporarily for troubleshooting.
Connecting for the first time
If you are new to Bitvise SSH Server, we highly recommend that you first make sure that you can establish a working SSH connection before you change any settings on the server. If you cannot connect to the SSH server using its default configuration, this is most likely due to a network or firewall problem that you will need to resolve before you are able to connect.
In its default configuration, Bitvise SSH Server accepts connections on the well-known port number for SSH servers, 22. To avoid drive-by password guessing, we recommend that you change this to a random port number between 1024 and 65535. This port number - either 22, or the port number you choose - is the only port you need to open in your firewall in order to connect to the SSH server. If you use port forwarding to tunnel other applications through SSH, you should not open any additional ports for the tunneled connections. All tunnelled connections are forwarded through the SSH session, established through the main SSH Server's listening port.
When connecting to Bitvise SSH Server with an SSH client for the first time, log in with the username and password of a Windows account that exists on the machine where the SSH server is running. To log into a Windows domain account, specify it in the 'domain\account' format.
You can use any SSH client to log into Bitvise SSH Server, as long as it supports SSH protocol version 2. Some Unix installations and old routers still have archaic SSH implementations which only support SSH version 1. Such installations must be upgraded, as SSH1 contains security flaws. In general, security software, including SSH software, should be kept up-to-date to minimize exposure to security flaws.
Clients that are known to work well with Bitvise SSH Server include Bitvise SSH Client (which works best), as well as CuteFTP Pro, Tectia, F-Secure / WRQ / Reflection, VanDyke (SecureCRT, SecureFX), OpenSSH, PuTTY, FileZilla, and many others. WinSCP works well in SFTP mode.
Opening Bitvise SSH Server to the internet
Bitvise SSH Server is intended to run with minimal configuration after initial installation. However, when installed in a LAN environment, it will not immediately receive connections from the internet by default.
In order to open Bitvise SSH Server to the internet, other network components must first be configured. The most prominent such components are the firewall on the machine where the SSH server is running, and the router on the LAN to which this machine is attached.
Before you open Bitvise SSH Server to the internet, perform the following important steps:
- Follow instructions in Connecting for the first time to make sure you are able to connect to the SSH server from 'localhost'; that is, that you can connect from a client running on the same machine where the SSH server is installed.
- Follow instructions in Securing Bitvise SSH Server to lock down your settings to a degree where you are comfortable with them. After locking down your settings, use a client installed on the same machine to verify that everything is behaving the way you want.
Only when you are satisfied with the security of your settings, and when your settings work when connecting from 'localhost', open your SSH server to the internet by:
- opening the Bitvise SSH Server listening port(s) in your firewall, and
- configuring your router to forward internet connections to the machine where the SSH server runs.
You can configure Bitvise SSH Server to perform the above tasks automatically:
- Open Bitvise SSH Server Easy settings. In the first tab (Server), change the setting "Open Windows Firewall" to "Open port(s) to any computer". This will automatically open your listening ports in your firewall when the SSH server is running.
- On the same tab, enable the setting "Automatically configure router (requires UPnP)". This will automatically configure your router to forward internet connections to Bitvise SSH Server when it is running.
If you have other software or hardware firewalls in addition to the Windows firewall, you will have to configure those firewalls manually.
In order for UPnP NAT forwarding to work, your router must support the Universal Plug and Play standard. Most recent routers do. If yours does not, you will have to configure it manually.
Configuring groups and accounts in Bitvise SSH Server
Immediately after initial installation, when started under original default settings, Bitvise SSH Server will accept password, NTLM or Kerberos-based login to any Windows account that has Windows permissions log into the machine where the SSH server is running.
When a Windows account user logs in, Bitvise SSH Server will impersonate the security context of that Windows account throughout the user's SSH session. Under default settings, the server will allow any successfully logged on user to take any action that the user is permitted by Windows and file system permissions. Such actions include accessing the terminal shell, running a program, uploading and downloading files, or connecting to another machine using SSH port forwarding.
Most administrators will find it desirable to configure Bitvise SSH Server in a way that restricts users' access further. The groups and accounts sections of Bitvise SSH Server's Advanced settings provide the means for this configurability. The groups and accounts in SSH server settings are an additional layer of security, imposed by the SSH server on top of the Windows permission system. Bitvise SSH Server settings do not replace Windows permissions, but provide complementary settings which Windows does not provide on its own.
Additionally, virtual groups and virtual account settings provide the means to add users in Bitvise SSH Server without having to create separate Windows groups, or having to create and maintain a Windows account for every user.
Windows groups and accounts
By default, the Bitvise SSH Server configuration for Windows groups and accounts is very straightforward. It consists of a single 'Everyone' group. In a default configuration, the SSH server settings for the Everyone group apply to all Windows accounts that log in via SSH.
When a user tries to log into Bitvise SSH Server with a Windows account, the server determines the settings for that account in the following manner:
Account settings. The SSH server searches the entries in 'Windows accounts' to find a match for the account that's logging in. If a match is found, the settings in the account entry are superimposed on the settings found for the account's group. If the 'Specify group' option is enabled, it is used to choose the account's group settings entry.
Group settings. If the SSH server was able to find a match for a Windows account settings entry in SSH server settings, and if this entry uses the 'Specify group' setting, then Bitvise SSH Server will use the configured group settings entry, under the following conditions: that the specified Windows group settings entry exists in the first place; and that the Windows account is actually a member of that group.
If any of these conditions is not met, Bitvise SSH Server looks up the local and domain groups of which the account is a member:
If none of the user's groups have an SSH server group settings entry, the Everyone group settings entry will be used.
If only one of the user's groups has an SSH server group settings entry, that group settings entry will be used, as long as it appears above the Everyone group.
If more than one of the user's groups have an SSH server group settings entry, then the user's "primary group" setting in Active Directory will control which group settings entry is used.
If more than one of the user's groups have an SSH server group settings entry, but the user is not a domain user, or the "primary group" setting does not resolve the group, then the SSH server will choose the group settings entry that appears first in SSH server settings.
This means that:
Bitvise SSH Server account settings can be configured individually by adding individual account entries in 'Windows accounts'.
SSH server account settings can be configured en masse, without having to add or maintain individual account entries, by configuring Bitvise SSH Server settings for a number of Windows groups. When there is no individual account settings entry, the SSH server will use appropriate group settings according to the rules described above.
When configuring settings for multiple Windows accounts through groups, automatic expansion of environment variables in string configuration fields may be helpful. Bitvise SSH Server will substitute environment variables in string fields such as 'Terminal shell', 'Initial directory' and 'SFTP root directory'. For Windows accounts, at least the following environment variables will be defined: USERNAME, USERDOMAIN, SSHWINGROUP (the name of the group settings entry selected, or 'EVERYONE'), and SSHWINGROUPDOMAIN (defined if the value of SSHWINGROUP references a domain group).
Virtual groups and accounts
For administrators who want to avoid setting up a separate Windows account for every SSH user, Bitvise SSH Server provides the means to create virtual accounts. Virtual accounts behave like Windows accounts, except for the following differences:
Windows accounts. A Windows account is created in Windows, and can be used to log into Bitvise SSH Server whether or not there is a corresponding Windows account entry. A Windows account exists outside of the SSH server as a Windows security principal.
Virtual accounts. A virtual account is created by adding an entry to 'Virtual accounts' in Bitvise SSH Server settings. A virtual account exists only inside the SSH server, but there is no awareness of virtual accounts in applications that an SSH session launches. Instead, those external applications are aware of a Windows account that is configured to back the virtual account. The backing Windows account provides an impersonation context on the level of the operating system.
Windows groups. The mapping between Windows account entries and Windows group entries in Bitvise SSH Server settings can be complex. It depends on the Windows account's actual Windows group memberships, Active Directory primary group settings, the 'Specify group' setting in the SSH account settings entry, etc.
Virtual groups.The mapping between a virtual account and its corresponding virtual group is straightforward. The virtual account entry always directly specifies a single corresponding virtual group.
Windows accounts. The password of a Windows account is maintained by Windows. It is possible to change it either using the Windows Control Panel or Computer Management, or through Bitvise SSH Server during SSH user authentication, or using the included bvPwd command line utility.
Virtual accounts. The password of a virtual account is maintained in Bitvise SSH Server Settings. It is configured by the administrator and cannot be changed by the user of the virtual account.
Backing Windows account. A virtual account still requires configuring a backing Windows account to provide the operating system-level impersonation context. Bitvise SSH Server will impersonate this backing Windows account when the virtual account is logged into. A single backing account can be used for any number of virtual users, and the backing account can be defined either for individual virtual accounts or for whole virtual groups.
Since version 5.50, it is no longer necessary to store the backing Windows account's password in the SSH server's password cache. However, a logon session created this way will need to use explicit authentication to access network resources.
When installed on machines that are not domain controllers, Bitvise SSH Server creates and manages a local Windows account. This account is used if no 'Windows account name' is configured for a virtual user. The Windows account is named 'BvSsh_VirtualUsers' on default (unnamed) SSH server installations, but can be named differently if you installed Bitvise SSH Server as a named site. Run 'net user' from a Command Prompt to discover the name of this account. Domain controllers do not have local accounts, so this feature is not available on domain controllers.
In other respects, a virtual account is just like a Windows account. Virtual account settings are superimposed on the corresponding virtual group settings just like with Windows group and account settings entries. All the SSH server settings for virtual accounts that look the same as for Windows accounts behave the same way.
Security architecture in Bitvise SSH Server
Bitvise SSH Server acts as an extension of the Windows operating system to support SSH login. As such, it needs to act in ways the OS would act to organize security contexts for logon sessions. This means the main SSH Server process is designed to run as Local System; however, the logon sessions themselves run in security contexts of users who are actually logging in.
When a user logs into the SSH Server, for example to use SFTP or open a terminal session:
- The SSH session will run as Local System until the user is authenticated.
- When the user is authenticated, the thread with the SSH session will switch to that user's security context. If the code implementing the SSH session were to misbehave, it's possible to switch back to the Local System security context from within the thread.
- The SFTP server or terminal session is launched as a separate process entirely in the user's security context. If the code implementing the SFTP server or the terminal shell were to misbehave, it cannot switch to the Local System security context.
Security context for virtual accounts
Virtual accounts configured in SSH Server settings do not exist as Windows accounts, but must still run in a Windows security context. On computers that are not domain controllers, the SSH Server will automatically create and manage a local Windows account named BvSsh_VirtualUsers. Unless a virtual account or virtual group is explicitly configured to use a different security context in Advanced SSH Server settings, the SSH Server will use BvSsh_VirtualUsers by default. If a virtual account running in this default context needs to access filesystem resources, Windows filesystem permissions need to be granted to the BvSsh_VirtualUsers account.
The default account that provides a security context for virtual accounts is named differently on a named SSH Server installation. For example, if you installed an SSH Server instance named "MyInstance", the account might be named BvSsh_MyInstance. You can use Windows local user management tools to discover the account name in this case. You can also find the name of the account in the SSH Server's textual log file entries corresponding to a logon session for a virtual account.
When the SSH Server is installed on a domain controller, it cannot create a default account to provide a security context for virtual accounts because Windows domain controllers do not support local Windows accounts. In this case, if the domain administrator wishes to use virtual accounts in the SSH Server, they must first create a domain account to provide a security context for virtual account logon sessions. This domain account must be configured as the security context for virtual accounts in Advanced SSH Server settings.
If virtual accounts should be able to access network resources using the credentials of an account that has been configured as a custom security context for virtual users, the password for this account should be added to the SSH Server's password cache using the Manage password cache interface, accessible via the Server tab in the SSH Server Control Panel. This will allow the SSH Server to create a logon session with credentials to access network resources in the name of this account.
Running the SSH Server under a different service account
Bitvise SSH Server is designed to run as Local System, but can run as a different account if you grant it privileges that effectively make it as powerful as Local System. These privileges are:
- Act as part of operating system (SeTcbPrivilege)
- Replace a process level token (SeAssignPrimaryTokenPrivilege)
- Adjust memory quotas for a process (SeIncreaseQuotaPrivilege)
Changing the SSH Server to run in a domain account security context will not grant your SSH users access to domain resources, because the users run in their own security context. If you need users to access network shares, see the page Accessing network shares to configure this in the SSH Server.
Using Bitvise SSH Server in a domain
Bitvise SSH Server fully supports environments with domain, domain forest, and Unix realm authentication. Since version 5.50, changes to Active Directory settings are no longer necessary to authenticate against the SSH server, except when using:
- domain accounts with public key authentication and without a password cache;
- virtual accounts with backing Windows domain accounts and without a password cache.
Active Directory Permissions for Password-less Logon
If you would like to use Windows domain accounts with public key authentication, or as backing accounts for virtual accounts; and if you do not wish to configure passwords for these domain accounts in the SSH server's password cache; then you will instead need to ensure that the SSH server has read permissions to user data in the Active Directory.
A default Active Directory installation may grant the necessary read permissions through the Active Directory group named "Pre-Windows 2000 Compatible Access". If these default settings have been changed, a permissions issue might arise when trying to use domain accounts with password-less logon.
If the SSH server's log files indicate permission-related issues when trying to use domain accounts with password-less logon, grant the necessary read permissions as follows:
- On the Domain Controller, open 'Active Directory Users and Computers' under Administrative Tools.
- In the View menu, enable 'Advanced Features'.
- Right click on the Users container in the tree view, and click Properties.
- In the Security tab of the new dialog, click Advanced.
- In the Permissions tab of the new dialog, add the computer running Bitvise SSH Server with ApplyTo=ThisObjectAndAllChildObjects and ReadAllProperties=Allow.
Windows domain order
Since version 5.50, Bitvise SSH Server no longer requires the Domain Order setting to enable login with non-fully-qualified usernames. Domain users are now able to log in, without providing a domain name as part of their username, using default SSH server settings.
The Windows domain order feature is still supported for administrators who wish to explicitly configure the order in which non-fully-qualified usernames should be looked up, to ensure predictable results.
Loading Windows Profiles
When configuring Bitvise SSH Server to provide SFTP and SCP access for domain users, users may have large Windows profiles that will be loaded before the user's file transfer session can start. Very large profiles may delay session startup. If you wish to prevent the SSH server from loading Windows profiles, please note that any of the following conditions will cause Bitvise SSH Server to load a user's Windows profile:
- "Map remote home directory" is enabled for the user in Advanced settings.
- "Map remembered shares" is enabled for the user in Advanced settings.
- There is an on-logon or on-logoff command configured to run in the user's context, and the "Do not load profile" option in the settings for the command is disabled.
- A terminal shell is opened by the client.
- An exec request is executed by the client.
- The client starts an SCP or SFTP session, and the "Load profile for SCP and SFTP" setting is enabled for the user in Advanced settings.
If you wish to prevent the SSH server from loading Windows profiles, you will need to make sure that none of the above conditions apply.
Network vs. interactive logon
Windows recognizes different types of logon with subtly different security implications. Bitvise SSH Server can be configured, on a per-account or per-group basis, to use either of the following two logon types:
Network logon. By default, this logon type is configured in SSH Server Advanced settings for both Windows and virtual group settings entries. This logon type is therefore used, by default, for all accounts.
Use of the "network" logon type allows the user to log in via SSH even if the underlying Windows account is not granted the Windows security privilege Log on locally. However, on Windows Server 2003, the default filesystem permissions normally block access to cmd.exe and other command line tools when a logon session does not have the "interactive" logon type.
Interactive logon. Use of the "interactive" logon type requires a Windows account to be granted the Windows security privilege to Log on locally. By default, this logon type is not used by the SSH Server because a significant number of Windows servers are configured in such a way that this privilege is not granted for non-Administrator accounts.
For some purposes, the "interactive" logon type is required. For example, it is not possible to access server-side printers in an SSH terminal session unless the SSH user is either an administrator, or has the "interactive" logon type. The "interactive" logon type is also required on Windows Server 2003 for terminal shell access.
Selecting the right logon type
We recommend that users who require terminal shell access use the "interactive" logon type. It will usually also make sense for them to log in via SSH as Windows users, not as virtual accounts.
On the other hand, we recommend that users who will use only file transfer and/or tunneling use the "network" logon type. If this use is outside of a domain environment, it may also make sense (less overhead) to create for these users virtual accounts, which are internal to Bitvise SSH Server, instead of creating a separate account in Windows for each user.
Technical clarification of logon type 8: NETWORK_CLEARTEXT
When the "network" logon type is used, the SSH Server actually uses Windows logon type 8, which is NETWORK_CLEARTEXT. This is not a security issue.
Windows logon types are described by Microsoft as part of documentation for the LogonUser API. At the time of this writing, authoritative descriptions of logon types NETWORK and NETWORK_CLEARTEXT are as follows:
LOGON32_LOGON_NETWORK: "This logon type is intended for high performance servers to authenticate plaintext passwords. The LogonUser function does not cache credentials for this logon type."
LOGON32_LOGON_NETWORK_CLEARTEXT: "This logon type preserves the name and password in the authentication package, which allows the server to make connections to other network servers while impersonating the client. A server can accept plaintext credentials from a client, call LogonUser, verify that the user can access the system across the network, and still communicate with other servers."
Since 2013, misunderstandings of this logon type have proliferated. This appears to be thanks to this paper by a forensics student, which butchered the description as follows:
- NetworkCleartext: "This is a kind of network logon where the password is sent over the network in the clear text. This is logged as logon type 8. Windows server doesn't allow connection to shared file or printers with clear text authentication. (...)"
This passage is a misunderstanding. Regardless, it is easier to find and read than LogonUser documentation. Thus, it has become cited in online answers, even though it directly contradicts the authoritative descriptions.
What "NETWORK_CLEARTEXT" means is: (1) the login is made over the network, and (2) the server where the login is performed has access to the original password, and can use it to impersonate the user to access further resources as part of the logon session. It does not mean that a password is sent, at any time, over the network unencrypted.
The use of NETWORK_CLEARTEXT - or another logon type that caches credentials, e.g. INTERACTIVE - is required if the session needs to have access to network resources. If NETWORK is used, Windows will not cache the logon credentials, and access to network shares on the user's behalf will not be possible without additional configuration.
The main distinction between logon types NETWORK_CLEARTEXT and INTERACTIVE is that the former requires the user to hold the Windows security privilege Access this computer from the network, whereas the latter requires Log on locally. Beyond this, the logon process is very similar. If you log into Windows via the local keyboard, this is much the same as when the SSH Server calls LogonUser with logon type INTERACTIVE. With NETWORK_CLEARTEXT, the logon process is also much the same; except that the user must hold a different Windows security privilege.
In no event does the SSH Server receive or send the password unencrypted over the network. The password is obtained either locally from the SSH Server's password cache, or is sent by the user over an encrypted SSH session. The SSH Server then passes the password to the Windows LogonUser function. For domain accounts, this will cause Windows to authenticate with the domain controller. As far as we know, Windows also does not send the password unencrypted over the network.
Setting up public key authentication
If your SSH client supports it, you can use public key authentication to log into Bitvise SSH Server. On Windows, we recommend Bitvise SSH Client, which has strong support for public key authentication, as well as password authentication, and Kerberos single sign-on in domain environments.
If you are new to public key authentication, we first suggest reading Public keys in SSH.
To set up public key authentication, you first need to generate a keypair on the client, or select one or more existing keypairs for use with client authentication. The procedure for generating the keypair depends on the client software being used:
- If you are using Bitvise SSH Client, click the link titled Client key manager in the Login tab. You can generate, edit, import and export keypairs in the dialog box that pops up.
- If you are using a different client, you need to follow its process for generating keypairs. For example, in OpenSSH, keypairs are generated using the ssh-keygen utility. Make sure to generate an SSH2 keypair (not SSH1). Use either the ECDSA, RSA, or DSA (DSS) algorithms. We recommend either ECDSA, or RSA with a size of 2048 bits or higher.
Once the keypair has been generated, you need to import the public key (not the whole keypair!) into the SSH Server.
SSH Public Key Subsystem
Recent Bitvise SSH Server and SSH Client versions support the SSH Public Key Subsystem. To import a public key into the SSH Server this way:
- Use the SSH Client to connect to the SSH Server using the account for which you want to configure the public key.
- Once connected, open Client key manager. Right-click on the keypair, and select Upload to server.
- You should now be able to authenticate using this keypair.
You can also use this feature to manage keys in the SSH Server from the command line, using the spksc client included with Bitvise SSH Client.
Import Public Key via File
With most clients, you can export the public key into a file, transfer the file to the SSH Server, and import it into SSH Server settings. In this case:
- Export the public key in either the standard SSH2 public key format, or in the OpenSSH format. If you are using OpenSSH, the public key file can be exported from an existing keypair using the ssh-keygen utility (consult 'man ssh-keygen').
- Once your public key file has been exported, transfer it to the machine where Bitvise SSH Server is installed, or a machine from which you manage the SSH Server remotely using Bitvise SSH Client.
- Use either Easy SSH Server settings (screenshot) or Advanced settings (screenshot) to import the public key into the SSH user's account settings entry.
Common mistakes: Make sure that you don't try to import the client's key into the server's host key management interface. The host key management interface is accessed directly from the "Server" tab of the Bitvise SSH Server Control Panel, and is intended to manage keypairs that authenticate the server. These keypairs are separate and unrelated to client authentication.
Synchronization with authorized_keys
For Windows accounts, the SSH Server also supports synchronization with ~/.ssh/authorized_keys. This feature must be enabled in Advanced SSH Server settings, under Access control. It allows Windows users to upload their SSH public key to a file named authorized_keys under a subdirectory named .ssh under the user's Windows profile directory. If this setting is enabled, the SSH Server will check for the existence of the authorized_keys file when the user logs out. If the file exists, the SSH Server will replace all of the public keys configured for the user with keys found in this file.
Use this feature with caution: if your system has unused authorized_keys files laying around, this may cause public keys to be unexpectedly deleted.
Fine-grained tunneling restrictions
'Tunneling' or 'port forwarding' refers to the ability of an SSH client (a) to have the SSH server initiate a TCP/IP connection to another server on the SSH client's behalf (called client-to-server tunneling), or (b) to have the SSH server accept incoming TCP/IP connections on a server's interface and port and forward those connections to the client (called server-to-client port forwarding). (You can learn more in our Short guide to SSH port forwarding.)
If your requirements are simple, Bitvise SSH Server provides two easy ways to control a user's or group's access to tunneling. In the Bitvise SSH Server settings entry for the account or group, there are fields Permit C2S port forwarding and Permit S2C port forwarding. Disable the first and the user will not be able to tell the SSH server to initiate outbound connections. Disable the second and the user will not be able to instruct the SSH server to listen for connections to forward to the SSH client.
Sometimes, such simple controls are not sufficient. For example, you may want to allow the user to use port forwarding to access a service provided by a particular machine on the server's local network; but you don't want to allow the user to use this capability to access any server on the internet, e.g. as a proxy for web browsing.
Such fine-grained control is provided by the Connect rules and Listening rules settings available in Bitvise SSH Server Advanced settings, separately for each group or account settings entry.
Connect rules control what destinations the SSH client will be able to connect to using client-2-server port forwarding. There are four types of connect rules: those that match IPv4 addresses, IPv4 addresses, DNS names, and a separate rule type that matches everything.
An IPv4 rule allows you to specify either a complete IP address (significant bits = 32) or a whole subnet (significant bits = 8 for 255.0.0.0, 16 for 255.255.0.0 or 24 for 255.255.255.0). The IP address 0.0.0.0 with significant bits = 0 will match any destination, and is equivalent to a match-all rule.
IPv6 rules work similarly to IPv4 rules, except that there are up to 128 significant bits.
A DNS name rule allows you to specify a destination either using a specific DNS name or a wildcard of the form *.com, *.bitvise.com or *.research.bitvise.com. A lone wildcard (just *) will match any destination, and is equivalent to a match-all rule.
Bitvise SSH Server processes Connect rules in the order they are configured. When the first match is found, the action configured for that rule is taken, and no further rules are searched.
If Bitvise SSH Server gets a client-to-server tunneling request for which there is no match in the account's Connect rules, the Connect rules of the corresponding group settings entry will be processed. If no match is found in the group Connect rules either, the connection is rejected.
By default, the Connect rule list for a group contains a single entry allowing access to all destinations if 'Permit C2S port forwarding' for the user is true. An account's Connect rule list is empty by default, passing all decisions to Connect rules defined for the user's group.
Listen rules control what server interfaces and ports the user will be able to bind in order to accept connections and forward them to the SSH client. There is a separate list of listening rules for IPv4 and IPv6 requests.
A listen rule identifies an IP address of one of the server's network interfaces, and a port range for which the SSH client is allowed or denied listening. The special address 0.0.0.0 for IPv4, or "::" for IPv6, matches any interface.
A listen rule may contain additional Accept rules which control the origin hosts from which connections to the interface and port range defined in the listen rule will be accepted. By default, the accept rule list contains a single entry allowing connections from all sources.
If Bitvise SSH Server gets a server-to-client tunneling request for which there is no match in the account's Listen rules, the Listen rules of the account's group settings entry will be processed. If no match is found in the group Listen rules either, the tunneling attempt is rejected.
By default, the Listen rule list for a group contains a single entry allowing all interfaces and ports to be bound if 'Permit S2C port forwarding' for the user is true. An account's Listen rule list is empty by default, passing all decisions to Listen rules defined for the user's group.
Example 1: Permit a client-to-server destination
Suppose your SSH server resides on machine 10.10.10.5 in your internal network, and you wish to allow the user to connect, via SSH tunneling, to a Remote Desktop service running on machine 10.10.10.16. You would first need to decide whether to configure this policy for the user's group or for the individual user. If for the individual user, you would need to add a Bitvise SSH Server account settings entry for the user if one does not yet exist. Then, in Advanced settings, you would open the group or account settings entry that you wish to configure this restriction for, and perform the following:
- Add 'allow' rule for client-2-server connections:
- open the 'Connect rules' list and click Add
- under Address type, select IPv4
- under IPV4 Address, input the IP address of the server to which the user is allowed to connect - in our example, 10.10.10.16
- under Significant bits, enter 32 to specify that the IP address in this case identifies an individual IP (10.10.10.16), not a subnet (e.g. 24 for 10.10.10.x)
- under Port range rule, set 'Port from' to the RDP port (3389), and 'Port to' to the same value
- under Instructions, enable the 'Allow connect' setting, and leave the rest at defaults
- click OK to confirm and add the configured rule
- Add a 'deny' rule for other client-2-server connections:
- click Add in 'Connect rules'
- the default will be a rule that covers all connections. Disable the 'Allow connect' setting, and click OK to add the rule.
In this example, if you wanted to prohibit the user from setting up any kind of server-to-client port forwarding whatsoever, you would simply set 'Permit S2C port forwarding' to false. Otherwise, if you wanted to configure a specific range of ports and interfaces where the SSH client may instruct the SSH server to listen, you would add appropriate Listen rules as in the Example 2 (below).
Example 2: Permit a server-to-client binding
Suppose your SSH server machine has two network interfaces: 10.10.10.5 is the private IP address in the local area network and 188.8.131.52 is the server's public IP address on the internet. You know that the user who will be logging into the SSH server will need to run a program on the server side which will initiate a TCP connection to the client, and the user will achieve this using server-to-client port forwarding. You want to allow the user to forward connections from the server's local network through the server's 10.10.10.5 private network interface, as well as from the server itself using the 127.0.0.1 loopback interface, but you do not wish to allow the user to listen for connections from the internet through interface 184.108.40.206. You also want to restrict the user to listening only on ports 1024-65535.
Again, you would first need to decide whether to configure this policy for the user's group or for the individual user. If for the individual user, you would need to add a Bitvise SSH Server account settings entry for the user if one does not yet exist. Then, in Advanced settings, you would open the group or account settings entry that you wish to configure this restriction for, and perform the following:
- Add 'allow' listening rule for 10.10.10.5:
- open 'Listening rules: IPv4' and click Add
- under IPv4 Address, input the IP address of the interface - in our example, 10.10.10.5
- under Port range, enter 1024 into 'Port from' and 65535 into 'Port to'
- enable the 'Allow listening' checkbox
- under Instructions, enable the 'Allow connect' setting and leave the rest at default values
- click OK to confirm and add the configured rule
- Add 'allow' listening rule for 127.0.0.1:
- repeat steps under 1, but using 127.0.0.1 instead of 10.10.10.5. You should now have 2 listening rules.
- Add 'deny' rule for other listening interfaces:
- click Add in 'Listening rules: IPv4'
- the default will be a rule that covers all IPv4 listening interfaces and all ports. Disable the 'Allow connect' setting, and click OK to add the rule.
- Add a 'deny' rule for IPv6:
- click Add in 'Listening rules: IPv6'
- the default will be a rule that covers all IPv6 listening interfaces and all ports. Disable the 'Allow connect' setting, and click OK to add the rule. Since you're not using any IPv6 interfaces, this will be the only IPv6 Listening rule.
In this example, if you wanted to prohibit the user from setting up any kind of client-to-server forwardings whatsoever, you would simply set 'Permit C2S port forwarding' to false. Otherwise, if you wanted to configure a specific range of destination servers and their ports to which the SSH client may connect, you would add appropriate Connect rules as in Example 1 (above).
Master/slave synchronization and clusters
In Bitvise SSH Server versions 6.xx and higher, the SSH server can be run in master/slave mode, which facilitates its use in a cluster or a large-scale deployment.
The scope of the master/slave feature is to automate synchronization of SSH server settings between SSH servers. It is intended for use in environments where administrators would like to apply settings changes on one server (the master), and have the changes automatically propagate to others (slaves). The master/slave feature does not interact with solutions for server monitoring or load balancing. If your deployment requires e.g. load balancing, you will need an external solution for that.
To cause some or all aspects of the SSH server's configuration to be automatically reproduced from a primary installation to one or more secondary installations, use the Instance type feature in Bitvise SSH Server Control Panel to configure the primary installation as the master. Then, configure secondary installations to run as slaves, and retrieve configuration changes from the master.
A typical cluster installation may wish a secondary server to appear identical to a primary server to its users. To achieve this, a slave would reproduce all aspects of the SSH server's configuration: settings, host keys, and password cache. The aspects of SSH server configuration that will be copied from the master are configured in the Instance Type dialog for each slave installation.
Configuring master/slave synchronizationMaster/slave synchronization is configured through the Instance type setting in the Bitvise SSH Server Control Panel (top right corner of the Server tab). The following steps are required:
- On the master server:
- Set instance type to Master, and configure a password which slave SSH servers will be required to present in order to synchronize settings from the master. We highly recommend configuring a long, secure, randomly generated password as described on this page.
- Use the Manage host keys interface to export the public keys of all host keys used by the SSH server. Alternately, write down the master's employed host key fingerprints so that you can enter them manually into slave configuration.
- On slave servers:
- Set instance type to Slave.
- Import the master's host keys through the Host and fingerprints setting. Alternately, use Add Fp to add a master's host key fingerprint, without importing the key.
- Enter the master's network address and port, and set the synchronization password to match the one configured on the master.
- In the remaining slave settings, configure which aspects of SSH server settings to synchronize from the master. Host keys can be synchronized from the master only if this is permitted in master settings.
- If you enable Auto-manage trusted host keys, the slave server will automatically add to its "Host keys and fingerprints" setting any new host keys generated on the master, assuming they haven't yet been employed. If the host key is already employed when it is first seen by the slave, the slave will not be able to connect regardless of this setting, because it has no previous knowledge of the key.
If a node fails...
If a slave goes down, then either the master and/or any other slaves will remain up. There will be nodes to handle connections, and it will remain possible to administer SSH Server settings for the cluster through the master. If the slave that failed is brought back online, it will re-synchronize.
If the master goes down, a slave will not automatically become a master. The master needs to be brought back online. Otherwise, an administrator needs to reconfigure the nodes in the cluster so that a different server will serve as master. While the master is down, changing SSH Server settings for the cluster through the master will not be possible, but slaves will continue to operate according to the last settings they received from the master. When the master is brought back online, slaves will re-synchronize.
Upgrading servers in a master/slave configuration
For a master/slave arrangement to function, servers that receive synchronization data must use an SSH Server version equal to, or greater than, the server from which they obtain the data.
Servers in a master/slave configuration should therefore be upgraded in the following order:
- Slave servers first.
- Secondary masters second.
- Master server last.
If a master server is upgraded to an SSH Server version that uses a settings format newer than some of the slaves, those slaves will no longer be able to synchronize with that master. However, newer versions of slaves will continue to recognize settings from an older master.
Unattended slave installation
If you would like to script several SSH Server slave installations, so that they can be performed unattended, the first preparatory step is to use the graphical SSH Server Control Panel on an exemplary slave installation to configure settings for a typical slave. This includes a step to import master host keys. Once the settings are configured and saved, use the same interface to export instance type settings into a file. For example, we assume the file is named BvSshServerSlave.wit.
On slaves you want to script, the next step is to perform a normal unattended SSH Server installation, which can be done independent of instance type. This is described on the page Installing Bitvise SSH Server.
Once the SSH Server is installed, you can use the utility BssCfg, which can be found in the SSH Server installation directory, to import slave settings from the command line, as follows:
BssCfg instanceType importBin C:\Path\BvSshServerSlave.wit
This command needs to be run in an elevated, administrative Command Prompt or PowerShell session.
Once this completes, the SSH Server is configured as a slave, and can be started.
Environment Variable Expansion
A number of string fields in Bitvise SSH Server's account and group settings entries support environment variable expansion. To see if a particular field supports this, check the help text associated with that field in Advanced SSH Server settings.
The basic rule of environment variable expansion is that if there is an environment variable named VAR, containing the value "value", then any occurrence of "%VAR%" will be replaced with "value":
C:\Dir\File-%VAR%.txt -> C:\Dir\File-value.txt
The SSH Server also supports advanced expansion rules. These mirror the expansion functionality available in the Windows Command Prompt:
|%VAR:~N%||Uses a substring of the value of VAR, starting from zero-based offset N.|
|%VAR:~N,L%||Uses a substring of the value of VAR, of length L, starting from zero-based offset N.|
|%VAR:S1=S2%||Before using the value of VAR, replaces each occurrence of S1 within the value with S2.|
|%VAR:*S1=S2%||Before using the value of VAR, replaces with S2 all characters from the beginning of the value and until the end of the first occurrence of S1.|
|%=LOWER:VAR%||Converts the value to lowercase. This can be combined with other advanced expansion rules. When combined, conversion to lowercase will be done first. For example: if USERNAME has value "JOHN", then %=LOWER:USERNAME:j=J% will produce "John". Requires SSH Server 6.41 or higher.|
|%=UPPER:VAR%||Like =LOWER, but converts to uppercase. Requires SSH Server 6.41 or higher.|
For example, a home directory structure such as M:\Home\a\Aaron, M:\Home\b\Benjamin, can be configured on a group-wide basis, without requiring account-specific settings entries, by setting the group home directory to:
When using environment variables in an On-upload command, we recommend that the script is given no parameters on the command line, but that it instead obtains information from its environment block. For example, in PowerShell:
This is to avoid pitfalls when parsing the command line, which may contain a path under the SSH client's control.
When environment variable expansion is performed in the context of a logged-on user's session, any variable in the user's environment block can be used.
When environment variable expansion is performed for on-logon, on-logoff, on-upload commands configured to execute in service context, any variable in the server's environment block can be used.
In addition, the SSH Server will define the following variables:
|HOME||If not already set by Windows, set by the SSH Server to a concatenation of HOMEDRIVE and HOMEPATH. The SSH Server does not set this for commands executed in service context.|
|HOMEDRIVE||If not already set by Windows, set by the SSH Server to the drive part of what the SSH Server thinks is the user's home path. It is possible for this home path to change as the session progresses. The SSH Server does not set this for commands executed in service context.|
|HOMEPATH||If not already set by Windows, set by the SSH Server to the directory path part of what the SSH Server thinks is the user's home path. It is possible for this home path to change as the session progresses. The SSH Server does not set this for commands executed in service context.|
|SSH_CLIENT||The SSH client's IP address, followed by space, followed by port number. The SSH Server does set this for commands executed in service context.|
|SSH_CONNECTION||The SSH client's IP address, followed by space, followed by the local interface address at which the SSH Server accepted the client's connection. The SSH Server does set this for commands executed in service context.|
|SSHSESSIONID||The numerical session ID assigned to the current client's session by the SSH Server. The first session receives ID 1001. The SSH Server does set this for commands executed in service context.|
|SSHUPLOADBYTES||Defined during execution of an on-upload command. Contains the number of bytes written to the uploaded file by the client.|
|SSHUPLOADFILE||Defined during execution of an on-upload command. Contains the full local drive and path to the file written to by the client. Warning: Part or all of this parameter is controlled by the SSH client, which can create file names using syntax not expected by the server administrator. If passing this environment variable to e.g. the Windows command interpreter, make sure to enclose it in double quotes: "%SSHUPLOADFILE%"|
|SSHUPLOADENDBY||Has the value "CLIENT" if the file was closed by the client, or "CLEANUP" if it was closed by session teardown. Files with the value "CLEANUP" are likely to be incomplete. Added in version 7.12.|
|SSHUPLOADNEW||Defined during execution of an on-upload command. Contains 1 if a new file was created, 0 otherwise. Added in version 6.31.|
|SSHUPLOADRESIZE||Defined during execution of an on-upload command. Contains 1 if at least one file resize request by the client was successfully completed, or if an existing file was truncated when opened. Contains 0 otherwise. Added in version 6.31.|
|SSHWINGROUP||Defined if the logged-on user is a Windows account (not virtual). Set to the name of the Windows group settings entry from which the Windows account inherits settings. Set to "EVERYONE" if the account inherits settings from the Everyone group. The SSH Server does set this for commands executed in service context.|
|SSHWINGROUPDOMAIN||Defined if the logged-on user is a Windows account (not virtual), and if the user inherits settings from a Windows domain group. Set to the domain name configured in the group settings entry. The SSH Server does set this for commands executed in service context.|
|SSHWINUSER||Defined if the logged-on user is a Windows account (not virtual). Set to the account name portion (without domain) of the user's full account name. The SSH Server does set this for commands executed in service context.|
|SSHWINUSERDOMAIN||Defined if the logged-on user is a Windows account (not virtual). Set to the domain name portion of the user's full account name, or the name of the local computer if the account is local. The SSH Server does set this for commands executed in service context.|
|USERDOMAIN||If not already set by Windows, the SSH Server sets this to the domain name portion of the user's full account name, or the name of the local computer if the account is local. The SSH Server does not set this for commands executed in service context.|
|USERNAME||If not already set by Windows, the SSH Server sets this to the account name portion (without domain) of the user's full account name. The SSH Server does not set this for commands executed in service context.|
|VIRTGROUP||Defined if the logged-on user is a virtual account defined in SSH Server settings. Set to the name of the virtual group settings entry from which the virtual account inherits settings. The SSH Server does set this for commands executed in service context.|
|VIRTUSER||Defined if the logged-on user is a virtual account defined in SSH Server settings. Set to the name of the virtual account. The SSH Server does set this for commands executed in service context.|
Scriptable configuration with BssCfg and PowerShell
Bitvise SSH Server comes with a textual configuration utility, BssCfg, which is useful for administering SSH servers in large-scale installations. It also comes with a configuration COM object, BssCfgManip, which can be used to configure the SSH Server from any language that supports COM, but is especially intended for use with PowerShell. Using BssCfg and PowerShell, it is possible to administer the SSH Server remotely from client machines where it is not possible to use Bitvise SSH Client and its Remote Control Panel feature.
Note: Everything that can be configured through BssCfg and PowerShell can also be configured through the graphical Bitvise SSH Server Control Panel. Users who don't need scriptable configuration do not need to learn BssCfg or PowerShell. Such users should simply use the graphical settings, accessed through the Bitvise SSH Server Control Panel.
The BssCfg utility resides in the SSH Server installation directory and can be run from a Windows Command Prompt or PowerShell. It can be run either locally or remotely using an SSH session. If run without parameters, it will show available commands and options.
BssCfg must be run by a user with administrative permissions, in an elevated Command Prompt or PowerShell window.
Configuring the SSH Server with PowerShell
To instantiate the SSH Server's configuration COM object in PowerShell, use:
$cfg = new-object -com "BssCfg726.BssCfg726"
- Like the BssCfg utility, the BssCfgManip COM object must be instantiated by a user with administrative permissions, in an elevated PowerShell window.
- The exact name of the BssCfgManip COM object is changed in SSH Server versions that contain changes to the configuration format. To find the name of the COM object for your installation, check the file BssCfgManip.idl in the SSH Server installation directory.
Once the object is instantiated, you can view its properties and methods:
$cfg | Get-Member
Locking and loading settings
When the BssCfgManip COM object is first instantiated, neither settings nor keypairs are locked or loaded. To modify settings, you must lock and load them as follows:
# Manipulate settings
Similarly, to modify keypairs - lock and load them as follows:
# Manipulate keypairs
Master/slave settings are stored separately, and must be locked and loaded as follows:
# Manipulate master/slave settings
If you are loading settings or keypairs only to view them; but not change them; it is not necessary to lock them.
To access server settings, use the property $cfg.settings:
# Load server settings and display the textual log file directory
Write-Host "Log file directory: $($cfg.settings.server.logging.logFileDir)"
To access master/slave instance type settings, use $cfg.instance:
# Load instance type settings and print the master server's address
Write-Host "Master server's address: $($cfg.instance.slave.host)"
Under $cfg.settings and $cfg.instance, every object in the hierarchy features integrated help. You can display it as follows:
You can find built-in help at all levels in the settings hierarchy. For example:
The same help text has also been compiled, and can be viewed as HTML:
The following example scripts have been renamed to .txt from their original extensions:
- PowerShell: Import a public key into a virtual account settings entry.
- PowerShell: Loop through virtual account entries, displaying the virtual filesystem mount points defined by each virtual account settings entry.
- PowerShell: Read a file with IP addresses and add them to Client address rules.
- PowerShell: Create a virtual account with server-configured port forwarding rules in the client-to-server and server-to-client direction.
- PowerShell: Locate any DSA keys, both among host keypairs used for server authentication, and public keys imported for client authentication in account and group settings entries.
- VBScript: Load, Update and Apply Settings.
- VBScript: Export Public Key in OpenSSH known_hosts Format.
- VBScript: Loop through Windows account entries, resetting their mount points so that each account has a single mount point with a real root path set to the value of the "Initial terminal shell directory" for the account.
- C#: List virtual accounts or remove a virtual account.
You can find the following scripts in your SSH Server installation directory:
- VirtAccountExporter.ps1: A PowerShell script that exports virtual account information into a CSV text file.
- VirtAccountImporter.ps1: A PowerShell script that imports new virtual accounts from a CSV text file.
These scripts can be used as-is, or modified according to your needs.
All enumeration values are accessible by name via the BssCfgManip COM object. For example, the following code can be used to set the setting pwCacheAutoSave:
$cfg.settings.access.pwCacheAutoSave = $cfg.PwCacheAutoSave.allAccounts
To list all available values for the PwCacheAutoSave enumeration, use:
Dual list properties
Every list is represented in its parent object by two properties through which it can be accessed. For example, the list of virtual accounts can be accessed through virtAccounts and virtAccountsEx. The extended list property (ending in Ex) allows manipulation of the list (add, erase, move items). The simple property (virtAccounts) provides easy, enumerable access to the items.
Simple list property - enumerate virtual accounts:
Foreach ($account in $cfg.settings.access.virtAccounts)
Extended list property - discard all virtual accounts:
Adding new accounts and groups
Accounts and groups are stored by the SSH Server in lists. Some of the fields in each list entry are the entry's sort key; when you use the help property, these fields are displayed with a ">" prefix. Other fields in each list entry may have a unique constraint placed on them; such fields will be marked with a "!" prefix in the result of help. For example, virtual accounts are sorted according to the value of their virtAccount field. On the other hand, values of the groupType, winDomain and group properties of Windows groups must be unique.
New account and group entries are added in two steps:
The entry is initialized with the new settings it will contain. In this step, it is accessed using the new property of the extended list. For example:
$cfg.settings.access.winAccountsEx.new.winDomain = "DomainName"
$cfg.settings.access.winAccountsEx.new.winAccount = "User"
$cfg.settings.access.winAccountsEx.new.loginAllowed = $cfg.DefaultYesNo.yes
$cfg.settings.access.winAccountsEx.new.xfer.mountPointsEx.new.realRootPath = "C:\Sftp\User"
When the initial data for the new list entry is configured, it is committed into the list:
This process applies not only to account and group settings entries, but to all settings items stored as lists. In the above example, there is already a second nested example, where a mount point is added to xfer.mountPointsEx.
Removing list entries
Accounts and groups can be removed from their lists as follows:
For ($i = 0; $i -lt $cfg.settings.access.winAccountsEx.count; )
If ($cfg.settings.access.winAccountsEx.GetItem($i).winAccount -ieq "User")
The above will remove any and all Windows account settings entries whose Windows account name matches "User". For example, it will match and remove a domain account "Domain\User", as well as a local account "User".
Blocking IP addresses
In order to block a large number of IP addresses, you can use PowerShell, or "BssCfg settings importText", to run settings instructions such as these:
$cfg.settings.access.clientAddressesEx.new.addressRule.addressType = $cfg.AddressVer6Type.ipv4
$cfg.settings.access.clientAddressesEx.new.addressRule.ipv4 = "10.2.3.4"
$cfg.settings.access.clientAddressesEx.new.addressRule.sigBitsIpv4 = 32
$cfg.settings.access.clientAddressesEx.new.instr.allowConnect = $false
$cfg.settings.access.clientAddressesEx.new.addressRule.addressType = $cfg.AddressVer6Type.anyIP
$cfg.settings.access.clientAddressesEx.new.instr.allowConnect = $true
These instructions will:
- Clear the list of rules in Advanced settings > Access control > Client address rules.
- Add a rule to block connections from the IPv4 address 10.2.3.4, identified with all 32 bits.
- Add a final rule to allow connections from other IP address.
Faster import using BssCfg
BssCfg implements a number of commands to access all aspects of the SSH Server's configuration. Among them are the commands BssCfg settings exportText and importText.
The command settings exportText is roughly equivalent to the following in PowerShell:
$cfg = new-object -com "BssCfg726.BssCfg726"
The result of settings exportText is executable directly as a PowerShell script, and will set SSH Server settings to what they were at the time of export. However, the settings in this format can also be imported using BssCfg settings importText.
The command settings importText imports SSH Server settings in the textual format exported by exportText. The syntax supported by importText is very limited, and does not support most PowerShell language features. The main feature of importText is that it is much faster than direct execution of the settings as a script in PowerShell.
If you run into any problems or need help with scripted configuration, feel free to contact us. Your questions will help us augment this guide so that answers to your questions, as well as solutions to any problems, will be readily available.
Virtual filesystem providers
Bitvise SSH Server supports pluggable filesystem providers. Third parties can implement custom providers to support any type of backing store for files accessed with SFTP or SCP via an SSH session. Possibilities include SFTP/SCP access to files contained in an encrypted database; or a bridge that permits SFTP/SCP access to files actually hosted on an FTP server. Third-party providers can be mounted concurrently with the SSH server's default provider, FlowSfsWin, allowing SSH users to access the Windows filesystem and a third-party provider's virtual filesystem concurrently.
Virtual filesystem providers need to be implemented as 32-bit DLLs with a C-style interface. No licensing or royalty fees are required. To get started, download our Virtual Filesystem Provider Development Kit.
SSH Server Control Panel protocol
Third party programs can interact with Bitvise SSH Server using the same protocol used by the SSH Server Control Panel, and can obtain the same information from the SSH Server that the SSH Server Control Panel displays. Look for BssStat.exe and BssStat.cpp in your SSH Server installation directory for an example command line program that implements the protocol, and its full source code.
Interpreting SSH Server Log Files using Microsoft Log Parser
Bitvise SSH Server's textual log files are recorded in a machine processable XML format. It is straightforward to process XML files using any .NET language, but another way to extract information from Bitvise SSH Server log files is using Microsoft Log Parser.
When using Microsoft Log Parser, it is important to use the following parameters:
With the -fNames:XPath parameter, field names will appear unambiguously, using the full XPath of the attribute to which they refer. The default setting, "compact", can assign unpredictable field names to attributes used in different types of log entries.
A basic command to find out who and when logged onto the server:
LogParser -i:XML -fNames:XPath -fMode:Tree "SELECT /log/event/@time, /log/event/session/@windowsAccount, /log/event/session/@virtualAccount FROM *.log WHERE /log/event/@name = 'I_LOGON_AUTH_SUCCEEDED'"
A optimization of the above command - we can shorten the paths using the -rootXPath parameter:
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time, /event/session/@windowsAccount, /event/session/@virtualAccount FROM *.log WHERE /event/@name = 'I_LOGON_AUTH_SUCCEEDED'"
We can use aliases to clarify output headers:
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time AS Time, /event/session/@windowsAccount AS WinAccount, /event/session/@virtualAccount AS VirtAccount FROM *.log WHERE /event/@name = 'I_LOGON_AUTH_SUCCEEDED'"
Show only logons to Windows accounts:
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time AS Time, /event/session/@windowsAccount AS WinAccount, FROM *.log WHERE /event/@name = 'I_LOGON_AUTH_SUCCEEDED' AND /event/session/@virtualAccount IS NULL"
Show only logons to virtual accounts - using the VirtAccount alias in WHERE clause to shorten the command:
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time AS Time, /event/session/@virtualAccount AS VirtAccount FROM *.log WHERE /event/@name = 'I_LOGON_AUTH_SUCCEEDED' AND VirtAccount IS NOT NULL"
Show all Windows and virtual logons on August 7, 2014 (local time):
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time AS Time, /event/session/@windowsAccount AS WinAccount, /event/session/@virtualAccount AS VirtAccount FROM *.log WHERE /event/@name = 'I_LOGON_AUTH_SUCCEEDED' AND Time BETWEEN '2014-08-07' AND '2014-08-08'"
Enumerate the times and remote addresses of logins for a virtual user named "Michele". This is not yet ideal - the matching in this example is case-sensitive:
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time AS Time, /event/session/@remoteAddress AS RemoteAddress FROM *.log WHERE /event/@name = 'I_LOGON_AUTH_SUCCEEDED' AND /event/session/@virtualAccount = 'Michele'"
An improvement on the above matches the username in case-insensitive fashion:
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time AS Time, /event/session/@remoteAddress AS RemoteAddress FROM *.log WHERE /event/@name = 'I_LOGON_AUTH_SUCCEEDED' AND /event/session/@virtualAccount LIKE 'Michele'"
To enumerate the times and remote addresses of all virtual users whose name starts with "M", case insensitive:
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time AS Time, /event/session/@remoteAddress AS RemoteAddress, /event/session/@virtualAccount AS VirtAccount FROM *.log WHERE /event/@name = 'I_LOGON_AUTH_SUCCEEDED' AND VirtAccount LIKE 'M%'"
To find out who and when transferred which files:
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time AS Time, /event/session/@windowsAccount AS WinAccount, /event/session/@virtualAccount AS VirtAccount, /event/sfs/parameters/@path AS Path, /event/sfs/parameters/@bytesWritten AS BytesWritten, /event/sfs/parameters/@bytesRead AS BytesRead FROM *.log WHERE /event/@name = 'I_SFS_TRANSFER_FILE'"
Limit the above query to files that have only been written to:
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/@time AS Time, /event/session/@windowsAccount AS WinAccount, /event/session/@virtualAccount AS VirtAccount, /event/sfs/parameters/@path AS Path, /event/sfs/parameters/@bytesWritten AS BytesWritten, /event/sfs/parameters/@bytesRead AS BytesRead FROM *.log WHERE /event/@name = 'I_SFS_TRANSFER_FILE' AND BytesWritten <> 0"
Find out who removed which '*.docx' or '*.doc' files (case-insensitive) on August 7, 2014, between 2 pm and 7:30 pm (local time):
LogParser -i:XML -fNames:XPath -fMode:Tree -rootXPath:/log/event "SELECT /event/session/@windowsAccount AS WinAccount, /event/session/@virtualAccount AS VirtAccount, /event/sfs/parameters/@path AS Path FROM *.log WHERE /event/@name = 'I_SFS_REMOVE_FILE' AND /event/@time BETWEEN '2014-08-07 14:00' AND '2014-08-07 19:30' AND (Path LIKE '%.docx' OR Path LIKE '%.doc')"
For more information on the Log Parser's SQL SELECT statement, execute:
LogParser -h GRAMMAR
We also suggest checking out the -o:DATAGRID output format.
Useful resources and utilities
Log analysis. Bitvise SSH Server records extensive textual log files in a machine-processable XML format. We recommend Microsoft Log Parser for analyzing the log files and extracting information you require. See also Interpreting SSH Server Log Files using Microsoft Log Parser.
Command-line utilities. The following is a list of command-line utilities which will likely be useful to Bitvise SSH Server users, along with short descriptions and links to documentation and/or source. If there is a utility you feel should be added to this list, let us know.
- BvRun. Included with Bitvise SSH Server. Executes a process in a custom manner. In particular, allows an SSH user to run a process so that it will not be terminated when the SSH session terminates using the -brj parameter. (This requires the 'allow job breakaway' feature to be enabled in SSH server settings.) Execute 'bvRun /?' from your SSH server directory for help.
- BvPwd. Included with Bitvise SSH Server. Allows a user to change their Windows password from a Command Prompt, in case they are using an SSH client that does not support password change. Execute 'bvPwd /?' from your SSH server directory for help.
- BssStat. Included with Bitvise SSH Server. Displays SSH sessions currently connected to the SSH server, as well as their tunnelled TCP/IP connections. Also included is C++ source code which demonstrates obtaining this information programmatically. Execute 'BssStat' from your SSH server directory for help.
- NetStat. Included with Windows. Run 'netstat -ano' for a list of active TCP/IP connections and listening ports, including PIDs of the processes using them. Valuable in diagnosing networking issues.
- SchTasks and At. Included with Windows. These commands allow Cron-like scheduling of tasks from the Windows command line. SchTasks is more capable and is a replacement for At in Windows XP. Documentation can be found online at Microsoft's site in the Windows XP Command-Line Reference A-Z.
There are plenty other useful utilities that can be found both already present in Windows or among the Windows Resource Kit Tools.
- Microsoft's Command-Line Reference documents the command-line utilities available in the latest versions of Windows.
- Instructions for writing Windows batch files (command line shell scripts executed by Cmd.exe), a useful skill for any Windows server administrator - particularly when managing machines remotely via SSH and SFTP.
Part 2. Tutorials
Frequently Asked Questions about Bitvise SSH Server
As an administrator of Bitvise SSH Server, you should first become comfortable with the SSH server's log files. Bitvise SSH Server writes warnings and errors into the Application section of the Windows Event Log, but it also writes more detailed information to textual log files. These are located by default in the 'Logs' subdirectory of the SSH server installation directory.
Whenever you have a problem, the SSH server log files are the first place you should look.
- Personal Edition
- Configuring and Running
- File Transfer
- Public Key Authentication
- Account Settings
- Usage Issues and Operation Concerns
- Upgrading and Moving
- Contacting Support
Q000. Where do I get an activation code for personal use?
No activation code is needed to use Bitvise SSH Server for personal use. If your Bitvise SSH Server Control Panel is saying that there is an evaluation period, this means that you installed the product as the Standard Edition. In this case, you need to uninstall Bitvise SSH Server, re-install it again, and choose the Personal Edition this time.
Note that Bitvise SSH Server may be installed in the Personal Edition only by genuine, non-commercial personal users who are not using the SSH server as part of a commercial endeavor, and are not using it in an organization, whether commercial or otherwise. All commercial or organizational use requires a purchased license.
Q020. What are the limitations of the Personal Edition?
The Bitvise SSH Server Personal Edition:
- On domain member computers, can use only local Windows accounts. On a domain controller, can use domain accounts of that domain controller.
- Can configure only one Windows group (Everyone).
- Can configure only one virtual group.
- Has a limit of 10 Windows account entries.
- Has a limit of 10 virtual account entries.
- GSSAPI authentication is disabled (Kerberos and NTLM).
If you are trying to make a decision about whether to use the Personal or Standard Edition, please note that in most cases, this is not a technical decision. All organizations, as well as personal users who do not qualify as non-commercial, must purchase a license for the Standard Edition. The Personal Edition is available only for users who are both personal and non-commercial, and are therefore likely to be unaffected by the above limitations.
Configuring and Running
Q100. After I install Bitvise SSH Server, what do I need to configure before I can start using it?
For a basic, open setup, just start Bitvise SSH Server and it will work. Use one of your existing Windows account names and passwords to log on. For a basic usage case, where you want to use the SSH server for remote administration, the default server settings do not need to be changed. The one exception is the Open Windows Firewall Setting, described in Q103.
After you have established a successful connection, consider locking down your settings to prevent SSH access to Windows accounts and features that you do not want to be accessible over SSH. See the page Securing Bitvise SSH Server for more information.
Q103. I can connect to Bitvise SSH Server from the local network, but not from the internet.
To help prevent inadvertently exposing your SSH server to the internet before it has been properly configured, Bitvise SSH Server will not open its ports to the internet by default. When you are ready to open your server to internet connections, go to Easy SSH server settings, and change the setting Open Windows Firewall to Open port(s) to any computer. If your Windows Firewall is disabled, or if you prefer to manage it manually, change this setting to Do not change Windows Firewall settings.
If you still cannot connect from the internet after making this change, make sure that your router is properly configured to forward SSH connections to the SSH server. You can configure the router directly through its administrative interface, or if the router can be managed using Universal Plug and Play, you can set Bitvise SSH Server to configure it. To let the SSH server manage the router, enable Automatically configure router (requires UPnP) in Easy SSH server settings.
Q110. How do I log in to a Windows domain account?
It is best practice to specify the username in either the domain - backslash - account format; for example, "COMPANY\John"; or with a fully qualified name; for example, "email@example.com".
It is possible to log into a domain account by providing only the account name, e.g. "John". In this case, authentication outcome may be undependable if Windows finds multiple matches. For example, there may be a local account named "John", a domain account named "DOMAIN\John", and another one named "OTHER\John". You can control the outcome in such circumstances by configuring the Windows domain order setting in Advanced SSH Server settings.
Users can also log in using Unix realm accounts if they are in a trust relationship with the server's domain. Such users must always provide the fully qualified Unix realm account name, because Windows cannot look up Unix realm usernames.
Q120. What client software can I use to connect to Bitvise SSH Server?
You can use any client program that supports SSH, as long as it implements SSH version 2 - the newer and secure version of the protocol. There are multiple types of SSH clients, including terminal session clients, file transfer clients, port forwarding clients, command execution clients, and they come in all sorts of combinations. If your client machine runs Windows, you can use Bitvise SSH Client for most purposes. Our SSH client offers an excellent terminal console, graphical file transfer, dynamic and manual port forwarding, as well as scriptable command-line clients and an FTP-to-SFTP bridge. Also available for Windows is PuTTY, which includes SSH file transfer programs 'pscp' and 'psftp'. On Unix platforms, the OpenSSH package is freely available and provides the 'ssh' program for terminal sessions and port forwarding, as well as 'scp' and 'sftp' for file transfers.
Q130. My Bitvise SSH Server logs show an error like 'Failed to bind listening socket', and I cannot connect to the server.
Such an error indicates that another application is already listening on the port you have configured for Bitvise SSH Server. The default port is 22, and this port is used as default by all SSH servers. It is likely that you already have another SSH server running on your machine, and that it is occupying port 22. You either need to shutdown the other SSH server, or configure Bitvise SSH Server to listen on a different port.
Q140. I can only log in with an administrator account - attempting to log in with a regular account fails.
There are two most common causes.
- You are trying to log in with an account configured in Bitvise SSH Server to use the 'interactive' logon type, but this account does not have the Windows permission to log on locally. On domain controllers, this permission is not granted to regular users by default and must be enabled in the Domain Controller Security Policy.
- You have successfully logged in with an account configured in Bitvise SSH Server to use the 'network' logon type, or you logged in using GSSAPI (Kerberos or NTLM) authentication, but starting the terminal shell failed with an Access Denied error. This is because default filesystem permissions on Windows 2003 servers grant access to cmd.exe and other command line tools only to 'interactive' users. Switch this user or group in Bitvise SSH Server to use the 'interactive' logon type, or modify filesystem permissions for cmd.exe and other command-line tools to allow execution by users logged in with the 'network' logon type.
For more information, please read the Network vs. interactive logon section in the Bitvise SSH Server Users' Guide.
Q150. I'm trying to get some SSH client to work with Bitvise SSH Server. However, the session gets terminated immediately after connecting, and the SSH server logs tell me: 'Unable to create child process: Access is denied.' What is going on?
In order to provide SFTP, SCP, terminal shell, or exec request functionality, Bitvise SSH Server must have permission from Windows to execute a child process in the name of the user. You have probably configured your machine in such a way that, when the user logs in and the SSH server starts impersonating that user, the server loses permission to execute the necessary child processes. In order to use Bitvise SSH Server, you must configure your machine so that the remote user will be able to run executables in the SSH server installation directory; plus, of course, whatever programs you want the user to be able to execute, such as the terminal shell - 'cmd.exe'. Read and execute access is also required to the dynamic load libraries that programs use - in particular, system libraries which reside in the \Windows and \Windows\System32 directories.
Q160. I'm trying to use PowerShell in a terminal session using the 'dumb' terminal type. It doesn't display a command prompt.
When PowerShell detects that its input or output streams have been redirected, it suppresses any prompts that it might normally display. If you want to host an interactive PowerShell prompt inside another application (such as Emacs), use "-" as the argument for the -File parameter. In many shells, this implies "taken from standard input."
powershell -File -
Thanks to Jim Snyder for discovering this solution.
Q170. An SSH client hangs for no apparent reason when connecting to Bitvise SSH Server, and then the session breaks due to an authentication timeout.
If the SSH client is set up to try Kerberos authentication, but Kerberos isn't configured properly between the client and the server, the client might hang when it tries to unsuccessfully figure out how to use Kerberos to authenticate with the server.
To rectify this behavior, you can disable GSSAPI (Kerberos) either in the SSH client, or in the SSH server.
To disable GSSAPI authentication in Bitvise SSH server, find the settings Kerberos 5 authentication and NTLM authentication on the Access control page of Advanced settings, and set them both to Disabled. Disabling both methods will disable GSSAPI authentication for all users. The server won't advertise the GSSAPI algorithms, and the client won't shoot itself in the foot trying to figure out how to authenticate using Kerberos.
Q180. How do I set up Bitvise SSH Server to use a different terminal shell, such as Windows PowerShell, or a Unix-style shell such as Bash?
The default terminal shell used by Bitvise SSH Server is the Windows Command Interpreter, cmd.exe, which is available on all Windows platforms. You can configure a different terminal shell in Advanced SSH Server settings, either individually for a particular user in their account settings entry, or for a group of users in their group settings entry. The two settings you must configure are as follows:
- Terminal shell: The command line to start your desired terminal shell interactively. For example:
PowerShell.exefor Windows PowerShell, or for Bash:
- Exec request prefix: The command line prefix to start a command with the shell non-interactively. For example:
PowerShell.exe -Command <SPACE>Note the space after "-Command" at the end! It needs to be there. Alternately, for Bash:
c:\path\to\bash.exe -c "Note the double quote mark at the end: that also needs to be there.
To use a terminal shell that doesn't come with Windows, such as Bash, you will need to install it. You can obtain Bash from a variety of sources, such as Cygwin.
For Windows PowerShell, see also Question 160.
Q185. How do I use the terminal shell or an exec request to run a process that should continue to run when the SSH session exits?
By default, all child processes of a terminal shell or exec request are terminated by the SSH Server automatically when the SSH session exits, unless you take extra steps to launch a process outside of the session's job object. You can do this using the following steps:
- Enable the setting Allow job breakaway in Advanced SSH Server settings. This setting can be configured either in an account settings entry for an individual account, or in a group settings entry as a default for multiple users. The location of this setting is as shown in this screenshot. If the user's shell access type is configured to Command Prompt or PowerShell, this setting is enabled by default.
- Within a terminal shell or exec request, use the BvRun utility in the SSH Server's installation directory using the -brj parameter to launch a child process outside of the session job. In addition, if you're launching a console program, use either -new or -det to run it in a new console or a detached console, otherwise the process will close when the console window closes. Example screenshot.
Q190. I would like to provide a user with access only to port forwarding, but the client disconnects unless I also allow terminal shell access.
Most clients can be configured to not require a terminal shell when using port forwarding:
- If using OpenSSH, use the -T parameter ("Disable pseudo-tty allocation").
- If using plink (part of PuTTY), use the -N parameter ("don't start a shell/command").
- If using the graphical PuTTY client, enable Connection > SSH > Don't start a shell or command at all.
If using Bitvise SSH Client, the session won't be disconnected if a terminal window can't be opened, in the first place. However, if terminal shell and/or SFTP access are not available, you may want to disable the automatic opening of a terminal shell and/or SFTP window. This is configured under Options > On Login.
Q210. When users log in, they can see all drives on my server. How do I limit them to a certain directory?
If using Easy settings, you will need to use the Virtual filesystem layout setting under a Windows or virtual account settings entry. Configure this setting to Limit to root directory, and then configure the Root directory. Alternately, select Advanced filesystem layout, and you can configure multiple directories.
If using Advanced settings, this feature is configurable either per-account or per-group. When editing account or group settings, click Virtual filesystem layout in the configuration tree on the left side of the account or group settings window. Edit Mount Points, and change the 'Real root path' setting for the default mount point (virtual mount path "/") to the directory you want the user to be able to access. The user or users will now be able to see only files and subdirectories in that folder.
If configuring mount points via Advanced settings for a specific account, note that the setting Inherit group mount points is enabled by default. Leaving this enabled will inherit a default mount point ("/") from group settings which allows full filesystem access unless it's redefined or undefined in the account settings entry.
If you want the user to be able to access multiple directories in independent locations, add additional mount points.
Note that the virtual filesystem limits will affect only users who log in using SFTP, SCP, or terminal shell using BvShell. Users who are allowed to use an external shell, such as PowerShell or the Command Prompt, will be able to use this shell to access the entire filesystem, limited by their Windows filesystem permissions.If you want users to only have file transfer access, you should configure their Shell access type to either BvShell or No shell access. See also Securing Bitvise SSH Server.
Q220. What is the difference between SCP and SFTP?
SCP and SFTP are different file transfer protocols. SFTP, despite its name, has no relation to FTP. It is a remote file access protocol which provides rich and fine-grained functionality for managing, accessing, and modifying files on an SSH server. SCP is an adaptation of the Unix utility 'rcp' to run over an SSH session, and provides simplistic file transfer operations only. SFTP is launched by the client opening a session channel and requesting the 'sftp' subsystem. SCP is launched by the client instructing the server to execute the SCP program via an SSH exec request.
Up to WinSSHD 4, the SCP subsystem was not supported as well as SFTP. In later versions, support for the two subsystems is integrated, and the same virtual filesystem can be accessed through SFTP or SCP. Since Bitvise SSH Server version 7, the BvShell shell access type now also provides access to the same virtual filesystem.
Q230. How do I get WinSCP to work with Bitvise SSH Server?
WinSCP works well in SFTP mode, without requiring additional configuration beyond any other SFTP client. In SCP mode, WinSCP requires a Unix-style terminal shell. WinSCP will work in SCP mode if you configure the user's Shell access type to BvShell.
Q240. The SFTP client I tested performs poorly with Bitvise SSH Server. How can I improve performance?
The first thing to check is whether the server is consuming at least one CPU core at 100% during most of the transfer.
If the server is consuming 100% of at least one CPU core, then you are running into hardware limitations of the server system.
If you are not seeing 100% consumption of a CPU core during most of the transfer, you're running into a limitation of the client or network. For high-performance transfers, the SFTP client must implement performance optimizations appropriate to available bandwidth and latency. These optimizations include:
- Read/write request sizing.
- Read/write request pipelining.
- SSH channel receive window sizing.
The only performance parameter the SFTP server has control over is its own SSH channel receive window size. However, this only affects the speed of uploads - not downloads - and Bitvise SSH Server is already aggressive in this regard; it's unlikely to bottleneck the client.
If your SFTP client doesn't reach transfer speeds that would cause the server to reach 100% of a CPU core, but network bandwidth is still available, try with a different client. Our Bitvise SSH Client performs aggressive pipelining, which might perform better than some other clients.
Please do not try to improve performance by disabling encryption. The performance impact of encryption is minimal in most cases, and disabling it defeats the principal goal of SSH, which is security.
Q250. How do I configure Bitvise SSH Server so that clients can upload files, but not see or modify existing files - a file drop scenario?
This behavior can be configured in Advanced SSH server settings. Under the account or group settings entry for which you want to configure this, open mount point settings under Virtual filesystem. To implement a file drop scenario, remove permissions to List, Read Existing, Write Existing, and Delete Existing, but enable the permission to Read/Write/Delete New. Also, enable Show empty directory if no access (enabled by default).
For more general information, see also Configuring Bitvise SSH Server for SFTP, SCP file transfer.
Q260. I don't want the SSH Server to load the users' Windows profiles during file transfer sessions. How do I prevent the Windows profile from being loaded?
Bitvise SSH Server will load the user's Windows profile if it's asked to provide functionality that requires the Windows profile. To avoid loading the Windows profile, turn off options which require it to be loaded. These options may be found in Advanced SSH server settings, either in a user or group settings entry. They are as follows:
- Session setup > Map remote home directory
- Session setup > Map remembered shares
- Session setup > On-logon or on-logoff command that does not disable profile loading
- File transfer > Load profile for SCP and SFTP
With all of the above options disabled, the SSH Server will not load the user's Windows profile for file transfer sessions.
In the case of virtual accounts used for file transfer, the most common culprit that's causing the Windows profile to be loaded is that Load profile for SCP and SFTP is enabled in the virtual group settings entry. This is disabled in new SSH Server installations by default, but the setting is inherited when upgrading older configurations.
Q265. Should I disable Windows profile loading?
The latest versions of Bitvise SSH Server are configured by default to avoid profile loading for virtual accounts, but profiles are loaded for Windows accounts by default. This is okay as long as the Windows accounts are not accessed very frequently.
Several versions of Windows contain a leak which will lead to resource exhaustion after a very large number of profiles have been loaded, requiring the system to be restarted. Unfortunately, our exploration of this issue indicates that this is not a problem we can fix in the SSH Server. You can work around it by following instructions in Q260 to disable profile loading.
Loading a Windows profile can also take varying amounts of time; sometimes up to a minute with large domain account profiles. Disabling profile loading when not needed will improve performance.
Q270. Does Bitvise SSH Server lock files being accessed by a client, to prevent them being accessed by other applications? How do we enable such blocking?
By default, files that clients are accessing via SFTP or SCP will be available for access by other processes. However, some SFTP clients that implement a sufficiently high SFTP version can request different locking.
For most clients that do not implement this, the administrator of the SSH Server can configure settings for a mount point so that files are locked by default. This is done in Advanced SSH server settings, under Virtual filesystem layout settings for the mount point in question, under Provider settings. The parameter that needs to be added is FileShare, with value Disable.
Q280. How can I arrange an email notification to be sent when a file is uploaded via Bitvise SSH Server?
The SSH Server can be configured to execute a command on successful upload. We provide the following example PowerShell script, which can send email notifications. The first few lines need to be modified according to your email setup:
Note that the file has been renamed to .txt from its original .ps1 extension.
This script can be used as an On-upload command in Advanced SSH Server settings. The setting can be configured either in an account settings entry for a single user, or in a group settings entry as a default for multiple users. When using the above script, the command would be simply as follows:
For more information about environment variables available to an On-upload script, see Environment variable expansion.
If a PowerShell script does not work, the Start-Transcript cmdlet can help with troubleshooting.
Public Key Authentication
If you are having problems related to public key authentication, you may also want to check our page about Public Keys in SSH.
Q300. Someone wants to use public key authentication to log into the Bitvise SSH Server I'm administering. They have already sent me their public key file. How do I tell the SSH server to use the public key file when that user logs in?
In the Bitvise SSH Server Control Panel, open Advanced settings and go to Access Control > Windows accounts (or Virtual accounts if this is a virtual user). If an entry for this user is not already present, you need to add one. For Windows accounts, the name of the entry must match the Windows username that will be used when logging in. Now, click Edit to open the account entry in a new window, and click the 'Public keys' link. A key management window will open which you can use to import the public key.
Screenshots for importing a client authentication public key:
We recommend also the Accounts and groups section of the Bitvise SSH Server Users' Guide for additional important information.
Q310. I am unable to import a user's public key into Bitvise SSH Server's user key management window. I keep getting a dialog box telling me that the public key could not be imported. What could be the problem?
It is most likely that the public key you are trying to import is not in the right format. It might be a private key instead of the public key, or it may be an SSH1 public key file instead of an SSH2 key. The formats supported by Bitvise SSH Server are the standard SSH2 public key format, and the OpenSSH SSH2 public key format. The OpenSSH SSH1 public key format is different and incompatible with SSH2.
Another possible reason you might have trouble importing a public key is if you try to import it into the SSH server's Manage host keys interface, instead of into an SSH account settings entry. The SSH server's host key management interface, which is accessible directly from the Bitvise SSH Server Control Panel, is intended to manage host keys that are used to authenticate the SSH server. The place to import a client authentication keypair is into an individual account settings entry, either in Easy or Advanced SSH server settings.
Q320. The client sent their public key, I imported it into SSH Server settings, but public key authentication doesn't work, and they're still being prompted for a password. Help!
Most likely, the user's client software is doing one or more of the following:
- Not attempting to authenticate using a public key because the keypair is not properly configured in the client.
- Attempting to authenticate with a keypair that corresponds to a different public key than the one you imported for this client.
- Attempting to log into an account that does not match the one for which you imported this public key.
To see which problem it is, check the Activity tab of the SSH Server Control Panel, and/or the SSH Server's textual log files. If the client is not attempting to use public key authentication, you will see this as an absence of any public key authentication messages in the logs. If the client is using a different key, log messages will show that the server does not recognize the key they're using. If the client is attempting to log into a different account, there will be discrepancies between the user name provided by the client, and the one for which the public key has been imported in SSH Server settings.
Q330. How do I set up public key authentication with Bitvise SSH Client?
If you are able to connect to the SSH Server using password authentication, and if the SSH Server administrator has not prohibited users from managing their public keys, the simplest process is:
- Use the graphical Bitvise SSH Client to connect to the server.
- Once connected, open the Client key manager interface.
- Use Generate New to create a new keypair.
- Right click on the keypair and select Upload to server. You should now be able to authenticate using this keypair.
Alternately, if you must configure public key authentication before connecting to the server, or the server does not allow you to manage your public keys:
- Open the graphical Bitvise SSH Client. You do not need to connect to the server.
- Open the Client key manager interface.
- Use Generate New to create a new keypair.
- Use the Export button to export the public key in standard SSH2 format.
- Send the public key file to the server administrator. If you're the administrator, transfer it to the server machine.
- Follow instructions in Q300 (above) to import the public key into Bitvise SSH Server.
Once the public key has been uploaded or imported into the Server, configure the Login > Authentication > Initial Method setting so that the Client will use your generated user keypair for authentication. You can also save your Bitvise SSH Client settings into a profile for convenience, and copy the keypair into the profile using the Client key manager.
If you wish to manage public keys configured for your account on the Server non-interactively, or via the command line, you can also use spksc - a command line public key management client that's included with Bitvise SSH Client. In this case, run spksc from a Command Prompt for help.
Q340. When I use password authentication, I can access EFS-encrypted files, and network shares on the server's Local Area Network. But when I use public key authentication, EFS-encrypted files and network shares are inaccessible. How can I access them when using public key authentication?
In order to access EFS-encrypted files, the server needs to provide Windows with your password. Similarly, to provide you with access to network shares on other computers in the server's network, the server needs to authenticate you with the computer providing the network share.
When you log in using password authentication, the SSH server conveys your password to Windows, and your login session is created in a way which allows Windows to access EFS-encrypted files, and pass your login credentials to other Windows computers in the network, providing you with access to network shares.
When you log in using public key authentication, Bitvise SSH Server versions 5.50 and higher are able to create your login session without the SSH server knowing your account password. However, a login session created this way does not have credentials necessary to access EFS-encrypted files and network shares.
One way to solve this is to add your Windows account's password to the SSH server's password cache. You can do this through the Manage password cache link on the Server tab of the Bitvise SSH Server Control Panel. The server will remember the password you enter indefinitely. When you log in using public key authentication, the server will use the cached password to create a logon session which will have credentials necessary to access network shares. This will work as long as the cached password remains synchronized with the account's actual password.
If you only need access to network shares (but not EFS-encrypted files), another way is to configure the SSH server, through per-group or per-account settings, to explicitly establish connections to one or more network shares, by providing network share access credentials in the SSH server's configuration. This can be done through the Windows file shares section of an account or group settings entry, in Advanced SSH server settings.
Q350. How can I let users manage their public keys themselves, without administrative intervention?
Bitvise SSH Server supports two ways for users to manage their client authentication public keys without requiring the administrator's manual intervention.
SSH clients that support the Secure Shell Public Key Subsystem (RFC 4819) can use this functionality to add or remove public keys associated with their account in Bitvise SSH Server settings. This feature is enabled for all accounts by default.
Windows accounts that have write access to their Windows profile directory can use the authorized_keys synchronization mechanism. To enable this, the administrator must enable the setting Synchronize with authorized_keys under Access control in Advanced SSH server settings. Windows users can then store a file named authorized_keys in a subdirectory named .ssh under their Windows profile directory. When the user's SSH session ends, Bitvise SSH Server will check for the presence of this file, and if it exists, the public keys encoded in this file will replace the public keys configured for the user in SSH server settings. This feature is disabled by default because some users have existing .ssh/authorized_keys files they are not aware of, which would conflict with intended SSH server settings.
Q360. When I examine the SSH server's log entries for a session, I see a logon attempt using the none method that fails, followed by a logon attempt using the publickey method that fails, followed by public key authentication that succeeds. How can I resolve the two failures?
These failures are a normal part of SSH authentication. First, the client may send a none authentication request, which is intended to fail, but provides the client with information about authentication methods supported by the server. Then, the client may attempt public key authentication without a signature, which is also intended to fail, but tells the client whether the server will accept the client's public key. Then, armed with this knowledge, the client sends the actual public key authentication request, which succeeds.
The client could avoid the preliminary requests if it were to assume outright that the server supports public key authentication, and that the server will accept the public key the client is trying to use. In this case, the client can just send the full public key request directly, as its first authentication request.
However, it is perfectly okay for the client to send the preliminary requests. This is a normal part of SSH authentication.
Q370. Public key authentication for Windows accounts does not work unless I enter the account's password in the SSH Server's password cache. Even though Windows has been restarted, the SSH Server Control Panel continues to display the message "Public key authentication, as well as virtual accounts that use a custom security context, will not be fully operational until Windows is restarted."
It is possible that Windows has been configured to not load LSA authentication packages installed by third party programs. This prevents the loading of the SSH Server's authentication package, which it needs to create logon sessions where a password is not available.
To allow the SSH Server's authentication package to load, follow the instructions under "To disable LSA protection" as described in the Microsoft TechNet article Configuring Additional LSA Protection.
Q400. How do Bitvise SSH Server account settings work?
We recommend the Accounts and groups section of the Bitvise SSH Server Users' Guide for this important explanation.
Q430. How can I limit a user so that they cannot access files outside of a certain directory?
The answer depends on what sort of access you have in mind. For shell access and remote execution, jailing a user is possible only through Windows file system permissions. On the other hand, if you are permitting the user only file transfer access (using SFTP and SCP), you can configure a limited-access virtual filesystem for the user by editing settings for their account or group in Bitvise SSH Server settings. In settings for the individual account or group, open Virtual filesystem layout > Mount points, and set the 'Real root path' setting for the default mount point ('/') to the directory you want them to access.
Q440. How do I use virtual accounts on a domain controller?
On computers that are not domain controllers, Bitvise SSH Server manages a local Windows account to provide a security context for virtual account login sessions. See Q530 for more information about this account.
On domain controllers, the SSH Server cannot create this account because there is no concept of local accounts on a domain controller. If you would like to use virtual accounts on a domain controller, you need to create or designate a domain account which will provide a security context for your virtual account login sessions. You then need to configure this backing account in Advanced SSH Server settings, either individually for each virtual account settings entry, or in a group settings entry used by one or more virtual accounts.
Q445. How do I configure virtual accounts to use a domain account as security context?
If you would like your virtual accounts to use a domain account as their security context, open Advanced SSH server settings, and edit your Virtual group settings entries as follows:
- Change Security context to A specific Windows domain account.
- Configure the Windows account domain and Windows account name settings to identify the domain account.
Virtual account settings entries will inherit their security context settings from their assigned virtual group by default - unless settings for a specific virtual account are configured differently.
We recommend also adding the domain account's password to the SSH Server's password cache. See Q450.
Q450. If I use an explicitly configured backing account to provide a custom security context for virtual accounts, do I need to enter the backing account's password into the SSH Server's password cache?
When you explicitly configure a backing account for virtual users, you can choose to save the password for this backing account in the SSH Server's password cache using the Manage password cache interface in the SSH Server Control Panel.
If you configure the password cache, the SSH Server will be able to create virtual account login sessions that will have implicit access to EFS-encrypted files and network resources (e.g. Windows shares) accessible to the backing account. If you do not configure the password cache, the virtual account sessions will still work, but without access to such resources. See also Q340, which describes the same issue when using Windows accounts with public key authentication.
Usage Issues and Operation Concerns
Q510. How can a user change their password remotely?
Bitvise SSH Server supports changing a Windows account password during SSH user authentication by using a client that supports this feature, such as Bitvise SSH Client.
Additionally, Bitvise SSH Server comes with a 'bvPwd' utility which allows any user to change their password if they know what it currently is. The utility can be found in the SSH server installation directory; run it with 'bvPwd -h' for help. Additionally, administrators can use the 'net user' command intrinsic to Windows to change any user's password - type 'net help user' in a Command Prompt for help.
Recent Bitvise SSH Server versions also allow virtual account users to change their passwords using an SSH client that supports this feature, such as Bitvise SSH Client. Virtual account passwords can also be changed by an administrator in Bitvise SSH Server settings or using bsscfg.
Q515. I'm using PuTTY or plink, and it logs me into a different account than I intend when I connect from a computer in the same domain. Why does this happen, and what can I do about it?
When used in a domain, PuTTY will automatically try to log you in using Kerberos + GSSAPI by default, using the Windows domain account with which you are logged into the client computer. PuTTY will do this even if you specify a different username for authentication, and have no intention of using Kerberos to authenticate.
In Advanced SSH Server settings, under Access control, there is the setting SSH username must match GSSAPI. You can set this to either Always or If the client is PuTTY (default) to cause the SSH Server to refuse PuTTY's login attempt unless it matches the username you specified.
A more drastic solution is to disable GSSAPI authentication either in PuTTY, or in the SSH Server. In PuTTY, it can be disabled under Connection > SSH > Auth > GSSAPI. In the SSH Server, it can be disabled in Advanced SSH server settings > Access control. If you disable both Kerberos 5 authentication and NTLM authentication, PuTTY will not be able to login with GSSAPI. But also, no other client will be able to log in using Kerberos or NTLM.
Q520. I'm seeing pop-ups from Bitvise SSH Server showing unknown IP addresses trying to guess usernames and passwords. This worries me. What can I do about this?
In order to refuse access to unauthorized users, while still allowing authorized users to log in, the SSH server must accept connection attempts coming from permitted sources, and must allow those connections to reach a point where the client can provide authentication credentials.
When installed with default settings, Bitvise SSH Server will already take several steps to thwart unauthorized attackers.
One way is by imposing a delay between login attempts. The default delay is 3 seconds. Without any other countermeasures, this 3 second delay would ensure that even an account with a weak password, e.g. 6 letters chosen randomly from an alphabet of 26, would on average take years of back-to-back attempts to guess. (Note however that passwords that short are still very weak and are not recommended.)
Another way Bitvise SSH Server tries to thwart attackers is through automatic blocking of IP addresses that have recently initiated multiple failed login attempts. In default settings, the SSH server will block for 1 hour any IP address that initiates more than 20 failed login attempts in 5 minutes.
If you wish to see fewer password guessing attempts, an effective mitigation is to configure your SSH server to accept connections on a port other than 22. This would not be very effective against a determined attacker, but will avoid random hackers looking for low-hanging fruit. Any random port number between 1024 and 65535 is suitable. The only issue is that any legitimate client that tries to connect to your server will then need to be configured with the port number in addition to the host name.
Despite these countermeasures, it is very important to make sure that your accounts are configured with complex passwords, and to lock down your settings so as not to grant access that you don't need to. For more information and for password complexity guidelines, see Securing Bitvise SSH Server.
Q530. When I install Bitvise SSH Server, it creates a local Windows account named BvSsh_VirtualUsers (or similar). What is the purpose of this account?
When you configure virtual users in Bitvise SSH Server settings, the SSH server needs to provide some kind of security context for actions taken by those users when they connect. Advanced SSH server settings allow you to configure a specific Windows account that should be used as security context for virtual users. If you don't take explicit steps to configure this, the SSH server will use as security context a default local Windows account, which it creates and manages for this purpose. In a default (unnamed) SSH server installation, this account is named BvSsh_VirtualUsers. (If your first installation was a version prior to 5.50, the account is named WinSSHD_VirtualUsers.)
You can use Windows Explorer, and other Windows administration tools, to apply Windows filesystem permissions to the BvSsh_VirtualUsers account. In this way, you can control what parts of your system a virtual user will be able to access. These Windows security permissions will apply to virtual users even if they are permitted to use terminal shell or exec requests, in which case, the virtual filesystem configured in SSH server settings does not apply.
You should not attempt to delete the BvSsh_VirtualUsers account, or change its password. Such changes will either be detrimental to your SSH server's operation, and/or will not be effective. Bitvise SSH Server will automatically enable this account when the SSH server is started, and disable it when the server is stopped. The SSH server will also periodically reset the password to this account, and set it to an extremely long, extremely complex random value. It will not be possible to log into this account, other than by allowing the SSH server to use it as security context for a virtual account.
Q540. Bitvise SSH Server's log files are very detailed. How do I extract just the information I'm looking for?
The SSH Server's log files are intended to be machine processable. Log files use the XML format, which can be handled by utilities such as Microsoft Log Parser, or custom applications. It's straightforward to process XML files using any .NET language.
For more information, see also Interpreting SSH Server log files using Microsoft Log Parser.
Q550. I enabled the Omit server version setting, but the server still sends the name of the product in the SSH version string. How can I completely remove product information from the version string sent to clients?
We support removing the exact server version number from the SSH version string, but we do not support completely removing the product name.
Any hacker who can exploit a server-specific vulnerability can also identify the server product based on the contents of the KEXINIT packet the server sends. KEXINIT packets are sent in plaintext and have specific patterns which are sufficient not just to identify the make of the server, but also a particular version subset.
Even if KEXINIT was not identifiable, it is always possible to narrow down the operating system make and version by the mere behavior of the TCP/IP stack on the server. The SSH Server cannot do anything about this.
As a result, removing the server software name from the version string only denies useful information to legitimate clients, while not significantly preventing attacks.
In general, we recommend the following measures to reduce the annoyance of drive-by hacking attempts:
- Instead of using the default SSH port (22), configure the SSH Server to accept connections on a random port between 1024 and 65535.
- If you only connect to the SSH Server using clients that support SSH protocol obfuscation (e.g. Bitvise SSH Client), enable obfuscation with a keyword.
Q560. How do I import a list of IP addresses I want to block from a file?
You can do this using the SSH Server's scriptable configuration language. For example:
After adding entries, edit the rule list to make sure that their order is correct - for example, that individual IP address block rules appear in the list before a blanket allow-all rule.
Q570. How many simultaneous client connections does the SSH Server support?
The answer depends on the type of sessions, their activity, the lowest performance you find acceptable, and hardware resources available to the server.
At the time of this writing, the SSH Server comes configured by default to allow a maximum of 60 concurrent sessions with processes. This is configured with the settings Limit total sessions and Maximal total sessions, found in Advanced SSH Server settings, under Session.
Sessions with processes include file transfer sessions, terminal sessions, and exec requests. Sessions that do not run child processes include sessions that only use port forwarding, or SSH Server Remote Control, or the SSH public key subsystem. There is no pre-configured default limit for sessions without processes.
The number of simultaneous sessions with processes is limited by default for two reasons. One is that these sessions consume more resources per session. The other is that older Windows versions, including Windows Server 2003, which our SSH Server still supports, will run out of a kernel resource called desktop heap space if more sessions with processes are created.
In more recent versions of Windows, desktop heap space is not statically limited, so you can raise or remove the default setting that limits the number of sessions with processes. In this case, the SSH Server's ability to support a greater number of sessions will depend on the activity of the sessions, and your hardware resources. If most of the sessions have low activity, hundreds can be supported. However, if all sessions attempt to perform file transfer at maximum available speed, they will start competing much more quickly for CPU and bandwidth.
Given unlimited hardware resources, the SSH Server can support several hundred simultaneous sessions on a 32-bit platform, and several thousand on a 64-bit platform. However, you are likely to run into CPU or bandwidth limitations before that, and may want to restrict the maximum number of sessions to a number that works for you.
Q580. While testing in a domain environment, I find that when using the gssapi-with-mic authentication method, the SSH Client allows me to log into the SSH Server with no password. Is this secure?
Yes, this is secure. In order for this type of logon to work, all of the following has to be true:
- Both the client and server computers have to be part of a Windows domain. The client can also be in a Unix realm.
- The client and server must be either in the same Windows domain, or in separate domains or realms that have an established trust relationship.
- The user has to be logged into the client computer as a domain account that can also log into Windows on the server.
- Bitvise SSH Server has to be configured to allow login with this domain account.
- GSSAPI authentication (Kerberos and/or NTLM) has to be enabled in the SSH Server. Because it only works in strict conditions, it is enabled by default. It can be disabled in Advanced SSH Server settings, under Access control.
If all of the above conditions are met, then the user is already logged into Windows on the client machine with an account that has permissions to log into the SSH Server. In this circumstance, the Windows domain infrastructure allows the user to use the gssapi-with-mic authentication method to perform single-sign-on authentication into the server, without having to authenticate again. This is a security advantage: it allows the user's password to be more complex, because it does not have to be entered as often.
If you would like to prevent this, make a change to any of the conditions enumerated above. The most common situation where this could be a problem is if the SSH Server is configured to accept login from domain accounts that you do not in fact want to log in. In this case, configure the SSH Server more strictly to accept login only using accounts you approve of.
Q585. File transfers sometimes get interrupted with an error like "MAC error", "data integrity failure", or "integrity check failed". What is the cause?
This type of error can be caused by:
An incorrect MTU setting. The Maximal Transmission Unit setting might be set to a too high value at one of the computers or network components involved in the connection.
This is a likely cause if the error:
- occurs with multiple types of client software over the same link;
- occurs early in the session, e.g. soon after starting the first upload or download, or even during authentication.
Accidental corruption of data in transmission that is not detected at a lower network layer. The TCP/IP protocol will detect errors in transmission with a probability of 65535 out of 65536, leaving a 1 in 2^16 chance that a transmission error will not be detected. If you are transferring a large file using a wireless link, or another type of connection that is prone to transmission errors, then any errors not detected by TCP/IP will cause an integrity check failure at the SSH layer. In this case, it is best to either switch to a more reliable link, or to use a client that is able to reconnect and resume transfer.
This is a likely cause if the error:
- occurs with multiple types of client software over the same link;
- occurs on the order of once per gigabyte of data transferred (depends on transmission error rate).
Implementation problem. The client that's interacting with the SSH Server might have a bug where it occasionally writes over the data it's sending or receiving, resulting in this error.
This is a likely cause if the error occurs with a single type of client software, but not other client software over the same link.
Intentional data corruption by an attacker with access to the network connection. This is usually least likely, but is the kind of attack that SSH is designed to detect. If an attacker tampers with data in transit, the SSH protocol cannot continue the connection, but it can detect and report that tampering may have taken place.
This is a possible cause if prolonged investigation of other causes does not yield results.
Q590. I keep getting the error: "The Bitvise SSH Server settings file is currently locked by another process." How do I resolve this?
This message indicates that some process on the system is keeping SSH Server settings open. This could be e.g. a settings window left open in an idle Remote Desktop session; it could be a BssCfgManip script that's intended to modify settings automatically, that has hung for some reason with locked settings; it could be the settings interface is open by an administrator through an SSH Client via the SSH Server Remote Control Panel.
If you cannot identify the program that is locking the settings, the easiest way to resolve the issue would be to restart the system. This will interrupt any SSH connections or Remote Desktop sessions that could be keeping the settings locked.
If the problem persists after restart, the most likely explanation is that the settings are being locked by an automated process. However, in most cases, the settings are just open in another logon session.
Q595. Users are getting the error: "Logon was successful, but the server has encountered logging problems. Only administrators can connect at this point." How do I resolve this?
You need to resolve the logging issue.
This error means that the SSH Server is unable to log to its textual log files. In Advanced SSH Server settings, under Logging, there is a setting that controls how the server should react when this happens. The default option is Allow connections from administrators only, and this option is being activated if you're seeing this message.
As server administrator, you need to investigate what's going on with the textual log files. Does the configured textual log file directory exist? We recommend that this directory is on a local drive; if it's configured as a network share, this may not be available to the SSH Server reliably.
Is there space on the drive where the log directory is located? If the drive is out of space, this issue would occur.
Is logging set to a higher level than needed? If yes, this can fill up disk space. The log level we recommend for textual log files is Errors, Warnings, Info. We do not recommend logging Trace or Debug events unless you are trying to debug a specific problem.
Upgrading and Moving
Q600. How do I upgrade my Bitvise SSH Server to the latest version?
There are two parts to the upgrade process:
Ensure that you have a license with upgrade access for the SSH server version to which you are upgrading. You can verify your upgrade access expiry date by logging into your License Overview.
If your upgrade access has expired, you will be able to add the desired number of license-years through the Place a New Order section under the license information. The expected new upgrade expiry date will be displayed on the page when you enter the desired number of license-years. The full cost will be displayed on the next page, at checkout.
- When you have a license with an activation code suitable for the latest version, download the installer for the latest version from our website. Run the installer on the computer where you want to upgrade your SSH server installation, and follow the process. Apply the new activation code in the Bitvise SSH Server Control Panel, once the upgrade is complete.
Q610. Will my settings and keys be preserved when I upgrade?
In general, yes. The upgrade process is intended to preserve your keypairs, your password cache, and your settings.
We do recommend making a backup of your settings before you upgrade. Settings can be exported using the Export function in the Bitvise SSH Server or WinSSHD Control Panel. A settings backup may be useful if the new version encounters a problem reading your settings, or if you decide to downgrade to the previous version again. The older version may not be able to read your settings once the new version has upgraded them.
Q620. I would like to move my SSH server to a different computer. How do I move my settings, password cache, and keypairs?
Settings. To move Bitvise SSH Server or WinSSHD settings, use the Export feature from the SSH server's Control Panel. On the new Bitvise SSH Server installation, use the Import feature to import the settings. Moving the settings will also move any client authentication public keys configured for user authentication.
Password Cache. In the latest Bitvise SSH Server versions, the password cache is necessary only if you use Windows domain accounts that need to log in using public key authentication, and also need to have implicit access as that account to network resources on other computers on the local network. In the latest Bitvise SSH Server versions, the password cache can be backed up into, and restored from a file, using Bitvise SSH Server Control Panel > Manage password cache > More > Backup items to file. If you want to move the password cache, but are using an older SSH server version which does not support backing up the password cache, you may want to upgrade the existing installation to a version that supports this, before moving.
Host Keypairs. It only makes sense to move keypairs used for SSH server authentication if the SSH clients accessing the server will continue to access the server using the same port, IP address, and/or DNS name. If there will be any change in what address or port the SSH client uses to access the SSH server, the client will need to re-verify the server's host key, so there's no benefit to transferring keypairs to the new installation.
If the clients will continue to access the server at the same address, the host authentication keypairs can be moved through the SSH Server Control Panel > Manage host keys.
Older WinSSHD versions (e.g. versions 4.xx) do not contain a user interface function to export a host authentication keypair. Depending on the version, you may be able to use the wcfg command line configuration utility to export the keypair. Alternately, you could upgrade to a more recent version in place, and export the keypair using the new version.
Client Public Keys. Public keys configured for client authentication are part of SSH server settings, and are moved implicitly when you export and import settings.
Q. I read the FAQ, but it didn't help me solve my problem. What do I do?
Contact us through our Contact page, and describe your problem in as detailed manner as possible. The more information you provide, the greater the chance of a swift and effective resolution.
Configuring Bitvise SSH Server for SFTP, SCP file transfer
Bitvise SSH Server provides multiple types of secure remote access to Windows. A frequent usage scenario is to configure the SSH Server specifically for file transfer, without exposing the machine to terminal shell, tunneling and other types of access. This tutorial explains step-by-step how to configure Bitvise SSH Server for a primary role as a file transfer server using SFTP and SCP.
Install Bitvise SSH Server. Do not start it yet.
When you install Bitvise SSH Server, the Easy settings wizard should appear. You can also access Easy settings at any later time by clicking Open easy settings.
If you have already performed any changes to SSH Server settings, click 'Restore', and then 'Reset settings to default values'.
The first tab of Easy settings is named Server settings. When you are ready for your server to accept connections over the internet, you will need to open this tab and enable the checkbox 'Automatically configure router (requires UPnP)'. You will also need to change the setting 'Open Windows Firewall' to 'Open port(s) to any computer'.
We recommend that you wait with the router and firewall settings until you have configured the server, and have tested your configuration by connecting to the server with an SSH or SFTP client installed on the same computer, or in your local network.
The next tab of Easy settings is named Windows accounts. This tutorial describes how to configure Bitvise SSH Server for file transfer using virtual accounts. Therefore, disable the checkbox 'Allow login to any Windows account'. This will prevent anyone from logging into your SSH Server using accounts not configured in SSH Server settings.
To use Bitvise SSH Server with virtual accounts only, do not add any Windows account entries under 'Windows accounts'.
The final tab of Easy settings is named Virtual accounts. Click the 'Add' button to add a virtual account, or use the 'Edit' button to edit an existing virtual account. Edit the virtual account settings as follows:
Virtual account name. This is the name that your user will use to log in.
Virtual account password. This is the password that your user will use to log in (unless you set up public key authentication).
Login allowed. Enable this if the account should be able to connect to your server. You can disable this to prevent access without deleting the account.
Allow file transfer. Enable this checkbox to allow SFTP and SCP access.
Shell access type. For virtual accounts, this is set by default to BvShell. This is a small shell provided by the SSH Server which respects the SSH Server's virtual filesystem settings. You can keep this setting configured to BvShell, or you can set this to No shell access. Important: Do not grant access to Command Prompt, PowerShell, or another shell except BvShell, if you want to restrict the user's filesystem access.
Allow port forwarding. Disable this checkbox to prevent the user from accessing other network services over SSH.
Virtual filesystem layout. Set this to Limit to root directory to limit the user's access to a single directory and its subdirectories. Set this to Advanced filesystem layout to configure a virtual filesystem for the user through which they can access multiple directory locations on the server.
To guarantee that your users can access the directories you configure for them, make sure that the Windows account BvSsh_VirtualUsers has Windows filesystem permissions to access those directories. This account is a member of the Users group, so if the Users group has sufficient access, the virtual account will have access as well.
- When you are done configuring virtual users, click 'Save changes' to exit Easy settings. You can now start Bitvise SSH Server and try connecting with an SCP or SFTP client. We also recommend trying to connect with an SSH terminal client to ensure that users cannot access terminal shell and port forwarding.
- Once you have tested your configuration and ensured that it works correctly, click 'Open easy settings' again and edit the router and firewall settings on the 'Server settings' tab to open your server to internet connections.
Having configured Bitvise SSH Server in this way, it will only accept connections from users who know one of the Virtual account usernames and passwords you have defined. The SSH Server will allow these users to only use SFTP or SCP, and none of the other SSH protocol features, and will restrict their file access to each user's root directory, or to their virtual filesystem mount points.
If you installed Bitvise SSH Server on a domain controller, the above steps will not be sufficient. Domain controllers do not have local accounts, so the SSH Server cannot manage a local account to provide the security context for virtual users. In this case, you will need to use the SSH Server's Advanced settings and configure a domain account to provide security context. Consult Configuring groups and accounts to learn more about how Bitvise SSH Server operates, so that you can configure it properly.
File transfer using Windows accounts
If you prefer your users to log in with Windows accounts, the process is nearly identical to the above instructions using virtual accounts. The main differences are:
- You need to configure Windows account settings entries instead of virtual account entries.
- You need to actually create the accounts in Windows.
- You need to grant the accounts Windows filesystem permissions to directories you want them to access.
- By default, Windows accounts are granted a more powerful Shell access type than virtual accounts. You need to change their Shell access type to BvShell or No shell access if you do not wish these accounts to access files and directories outside the mount points you configure for them in SSH Server settings.
On-upload scripts, actions, and commands
It is possible to configure the SSH Server to run a command or a script after a user completes an upload. To set this up, you need to configure the setting On-upload command, which can be found in Advanced SSH Server settings. The setting can be configured either in an account settings entry for an individual user, or in a group settings entry as a default for multiple users.
In most cases, we suggest you use the On-upload command to run a PowerShell script. The command can be simply:
We recommend that the script is given no parameters on the command line, but that it instead obtains information from the environment variables provided by the SSH Server. For example:
This is to avoid pitfalls when parsing the command line, which may contain a path under the SSH client's control.
If a PowerShell script does not work, the Start-Transcript cmdlet can help with troubleshooting.
On-upload email notifications
A common use case for the On-upload command is to set up email notifications for completed incoming transfers. We provide the following example PowerShell script, which can be used to send email notifications. The first few lines of the script need to be modified according to your email setup:
Please note that the file has been renamed to .txt from its original .ps1 extension.
Securing Bitvise SSH Server
At default settings, Bitvise SSH Server will permit any user who knows a valid Windows username and password to log in and use the following SSH services:
SFTP and SCP file transfer, allowing the user to access all files and folders that can be accessed by the Windows account they used to log in.
Access to a Command Prompt via terminal console, allowing the user to execute all programs that can be executed by the Windows account they used to log in.
- Routing TCP connections through the SSH server, either from the client to the internet, or from the internet and to the client.
Securing Bitvise SSH Server involves:
Configuring the SSH server to allow access only to a restricted subset of Windows accounts configured on the system, or only to virtual accounts configured in Bitvise SSH Server itself.
Identifying which of the above features you want to limit or disable, and doing so.
Making sure that strong authentication is in use for those accounts that can log in.
In order to avoid frustration, do not start locking down SSH server settings prematurely. Make sure that you can establish a connection first. Make sure that you and your users can use the SSH features that you want to use.
To avoid security issues, you can conduct such testing and preliminary setup with a closed firewall. Install an SSH client such as Bitvise SSH Client on the same machine where Bitvise SSH Server is installed, and use that client to connect to the SSH server to test the connection. After you are satisfied that the features you require work correctly, start securing SSH server settings. Once your settings are locked down to provide only the types of access you require, open the SSH port in your firewall and permit outside connections.
Restricting access to chosen accounts
If you are using Easy settings, disable the checkbox Allow login to any Windows account on the Windows accounts tab.
If you are using Advanced settings, go to the Everyone group under Windows groups, and disable the Login allowed checkbox. This prohibits SSH login to everyone except the users you configure.
In order to allow a Windows account to log in, you now need to merely add an SSH server account settings entry under Windows accounts, and configure the following fields:
Windows account domain. Leave this empty if you're adding a local Windows account.
Windows account name. The name of the Windows account you are adding. It must be an existing account already created in Windows.
Login allowed. Set this to Yes. This permits this particular user to log into the SSH server, overriding the fact that login is disabled in group settings.
For virtual users, similarly, all you need to do is add a virtual account entry, defining a username and password. For virtual users, you don't need to disable the group default for Login allowed, because there are no virtual users other than those that you configure, in the first place.
Disabling features you don't want
If you intend to use Bitvise SSH Server for file transfer, you will want to disable the other SSH features that you don't want your clients to use.
If you are using Easy settings, use them to configure access types for each account you add.
If you are using Advanced settings, the easiest way to disable features for all users is to do so at the group level. At this point, you may want to consult Configuring groups and accounts for an overview of how Bitvise SSH Server treats Windows- and virtual accounts and groups.
In the most common and straightforward case, you will have a single Windows group for Everyone in SSH Server settings. This Windows group controls settings defaults for all Windows accounts that might log in through the SSH server.
Open this group and configure the following settings:
Shell access type. Setting this to No shell access prevents users from executing arbitrary commands.
Permit C2S port forwarding. This prevents your users from accessing other network services over SSH.
Permit S2C port forwarding. This prevents your users from providing access to their own network services over SSH.
If you are using Bitvise SSH Server for more than only file transfer, you may want to leave some of these features enabled. In particular, if you are using SSH for tunneling, do not disable port forwarding. If you are using SSH for remote program execution or a remote console, do not disable shell access. Or disable these settings for the Everyone group, but enable them for the particular users that need these features.
If you will be using virtual accounts, apply the same restrictions to your virtual groups. (By default, there is a single virtual group named Virtual Users.)
Limiting directory access
By default, Bitvise SSH Server permits each user to access any and all parts of the filesystem that Windows filesystem permissions allow them access to. Frequently, you want to limit users to be able to access only a particular directory. Note, however, that it is only secure to impose such restrictions if you have also followed instructions above and disabled access to port forwarding and shell access.
Filesystem access is controlled:
- In Easy settings, under the Virtual filesystem layout section of each account settings entry.
- In Advanced settings, under the File transfer section of each group and account settings entry.
If you are using Easy settings, make sure the Virtual filesystem layout settings are configured securely for each user.
If you are using Advanced settings:
Open File transfer > Mount points for the Everyone group, edit the default mount point ("/"), and set the Real root path to point to an innocuous, empty directory. Or, if all your users should have access to the same folder, you can configure this to point to that directory.
In per-account settings, you can configure a different set of mount points for each user. Under Virtual filesystem layout settings for the user, disable the Use default layout checkbox. Then configure the Real root path for the default mount point ("/") to specify the directory which you want this user to access.
You can configure multiple mount points in this way, permitting the user to access a selected number of server directories. You can also use mount point permission settings to allow the user to only read, but not write to, files in a particular directory.
Ensuring strong authentication
Password authentication can be secure, but only if the passwords are complex. Unless you want the general public to log into a particular account, you need to ensure that all accounts for which you are permitting SSH login - be they Windows accounts or virtual - are configured with complex passwords. Bitvise SSH Server does impose delays and IP blocking to prevent aspiring attackers from successfully guessing a password, but this will not help if your passwords are as simple as "1234" or "password1".
A reasonably complex password would consist of at least 15 random characters from an alphabet of a-z, A-Z, and 0-9. If the chosen password is truly random, this provides the equivalent of about 90 bits of security. This is not as good as a 128-bit symmetric key, but is secure if the only way the attacker can guess a password is by trying to log in via SSH.
If you also want to protect against an attacker who has access to a cryptographic digest of your password, such as by having a copy of your authentication database, or by having physical read access to your system drive, then you need at least 22 characters from the same alphabet (a-z, A-Z, and 0-9) for security equivalent to a 128-bit symmetric key.
Expanding the password alphabet to include non-alphanumeric symbols may not be as great an idea as commonly supposed. Even if 28 symbols are included, the number of characters needed for 128-bit security is still 20. The 10% savings in password length are outweighed by the hassle of entering the symbols, and even more so by problems with programs that interpret such symbols incorrectly.
It is crucial, however, that you do not create your passwords by hand. If you do so, they will not be random. Use a password management utility to securely store your passwords in an encrypted database, and to randomly generate passwords of the desired length.
Public keys in SSH
This page attempts to explain public keys, as used in SSH, to readers unfamiliar with the concept.
The following concepts need to be understood by everyone, including beginner users:
- A private key is a very large, pseudo-randomly generated number, that contains your secret information in any operation involving public keys. You generate the private key on your computer, using one of a variety of programs, and store it securely. You need to keep your private key secure; you never send it to anyone.
- A public key is a very large number, mathematically derived from your private key. It is derived in such a way that the two numbers are linked, but so that the private key cannot be discovered by anyone who only knows the public key. The public key is what you send to other parties, to whom you want to authenticate yourself, or to whom you want to send encrypted messages. The public key is not sensitive, and does not need to be protected. It is public.
- A signature is a result of a cryptographic calculation that can be performed only by a person who holds a private key, and can be verified by anyone who knows the corresponding public key. If a person is able to produce a valid signature of random data on demand, this proves that they have access to the private key associated with a particular public key.
- A keypair is the combination of the private key and the public key. A keypair by necessity includes the private key, so a full keypair ought to be protected just like a private key.
- A fingerprint is a cryptographic digest that can be used to uniquely identify a public key. Fingerprints are used for practical purposes, to compare keys which would otherwise be too large and unwieldy to compare manually. The most common fingerprint formats used in SSH are the SHA-256 format (example - "C+VpXsf..."), the MD5 format (example - "43:71:be:ab:d3:..."), and the Bubble-Babble format (example - "xubem-kiloc-getad-ponyh-..."). Like a public key, the fingerprint is not sensitive information, and can be disclosed indiscriminately.
- A certificate is a public key, cryptographically signed by a certification authority in a PKI (Public Key Infrastructure). In the way SSH is most commonly used, it does not support certificates. Most common SSH implementations only use raw public keys and private keys.
SSH sessions use public keys for two main purposes: server authentication, and client authentication. Both processes work very similarly, but they involve separate sets of keys. When discussing a specific public key in the context of SSH, it is important to be aware whether the key is intended to authenticate the server, or a client.
In Bitvise SSH Server
In Bitvise SSH Server:
Keys that are used for server authentication are displayed in the Host keys section of the Bitvise SSH Server Control Panel, and can be configured through the link Manage host keys. This section manages private keys, which are stored by the SSH server so it can authenticate itself to clients that connect to it.
In Keypair management, you can use the Export feature to export the public key associated with each host authentication keypair. You can send such a public key to an SSH client, so it will be able to authenticate the SSH server when it connects to it. If you do not send the client the whole public key, you should at least provide the user with your server's host key fingerprint, so that the user can confirm the server's public key is correct when they see it. The server's host key fingerprints can be viewed and copied from the main screen of the Bitvise SSH Server Control Panel.
Keys that are used for client authentication are managed through individual account settings entries. To view or configure these keys, open Easy or Advanced SSH server settings, and open a specific Windows or virtual account entry. Within this entry, find the link named "Public keys". This link opens a dialog which can be used to import, export, or remove public keys that the client can use to log into the server.
The SSH server will allow a client to add or remove keys themselves if "Allow public key management" is enabled in the user's account or group settings entry in Advanced SSH server settings, or if "Synchronize with authorized_keys" is enabled in Advanced settings > Access control.
In Bitvise SSH Client
In Bitvise SSH Client:
Public keys of hosts you have connected to can be viewed in Host key manager. When you connect to an SSH server for the first time, and the client has no record of the server's host key, the client will display the server's host key fingerprint, and will ask you to verify the key. It is very important to verify the server's host key fingerprint at this point, which should be done against a fingerprint you received through other means from the server's administrator. If you verify that the key is correct, the client will save it, and trust it for further connections to the server.
If you do not verify the server's host key, then the client cannot verify that the server you are connecting to is in fact the server you think you are connecting to. Not verifying the server's host key enables a man-in-the-middle attack - a situation where you aren't actually connecting to the server you think you are connecting to, but rather to another server that impersonates you to the destination server, and the destination server to you. Such a man-in-the-middle is able to observe all data sent over the SSH session, as well as modify what's being sent in real time. The only way to defend against such an attack is to make sure to verify the server's host key.
Private keys used for client authentication can be generated, imported and exported in User keypair manager. This interface allows you to generate a keypair for client authentication; you should keep the private key that's part of this keypair private. You can use the Export function to export the public key portion of a keypair, and send it to a server administrator, who can import it to allow you to authenticate using that key. The only thing you need to send to the server administrator is the public key; you should never send the private key, or the whole keypair.
A short guide to SSH port forwarding
SSH port forwarding, or TCP/IP connection tunneling, is a process whereby a TCP/IP connection that would otherwise be insecure is tunneled through a secure SSH link, thus protecting the tunneled connection from network attacks. Port forwarding can be used to establish a form of a virtual private network (VPN).
To illustrate how port forwarding works, let us use an example. Suppose you are the network administrator in a company that has two buildings. In Building #1, there are numerous workstations residing in the subnet 10.1.1.*. In Building #2, there are multiple servers residing in the subnet 10.2.2.*. The two buildings are separated by a busy street with parking spaces on each side, and the subnets in the two buildings are linked wirelessly through an antenna on the roof of each building. The workstations in Building #1 are running a legacy client application that uses an unencrypted TCP/IP session to communicate sensitive data with the servers in Building #2.
One day, someone in your company notices that an unmarked black van has remained parked on the street between the two buildings for several days. As your CEO realizes that sensitive data is being transmitted unencrypted between the two buildings, he becomes worried that the van parked outside might be collecting the company's confidential information. He orders you to solve the problem ASAP.
What you do is this:
On each of the client workstations in Building #1 (in the above example, workstation 10.1.1.7 is shown), you install an SSH client. On the machine in Building #2 that runs the server for your legacy application, you install an SSH server. You configure the SSH client with the following client-to-server port forwarding rule: for each connection that comes on interface 127.0.0.1 and port 999, forward that connection to the SSH server, and request the SSH server to forward that connection to host 127.0.0.1 (relative to the server), port 123.
Now, your application client doesn't connect to the server directly anymore. Rather, it connects to the SSH client, which encrypts all data before transmission. The SSH client forwards the encrypted data to the SSH server, which decrypts it and forwards it to your application server. Data sent by the server application is similarly encrypted by the SSH server and forwarded back to the client.
Previously, the data that was being radioed between the two buildings was sent in plaintext and could be captured by anyone parking on the street below. Now, the data is encrypted using the SSH protocol, and is virtually impossible to decipher. The next day after installing SSH, you observe that the black unmarked van is gone.
Now, let us comment on the above example. It corresponds with the following C2S (client-to-server) port forwarding rule in the SSH client:
- Listen interface: 127.0.0.1 (this ensures that only connections from the local client machine, or loopback connections, are accepted for forwarding)
- Listen port: 999
- Destination host: 127.0.0.1 (important: the target address is relative to the server, not the client, so 127.0.0.1 will work fine if the target application server is listening on all interfaces - 0.0.0.0)
- Destination port: 123
Note that the listening interface configured on the SSH client is 127.0.0.1. By configuring the listening interface, you tell the SSH client what kinds of connections it will accept. If you configure the listening interface to 127.0.0.1, the SSH client will only accept connections originating on the same machine. If you configure the listening interface to equal the IP address of one of the network cards on the machine, the SSH client will accept only those connections that arrive through that network card. If you configure the listening address to 0.0.0.0, the SSH client will accept connections regardless of their origin.
Next, you will note that the listening port has been set to 999. The listening port could be set to any figure between 1 and 65535 that is not already occupied by another application listening for connections on the same machine. In this case, the SSH client listening port has been set to 999, but it could just as well have been set to 123, the same port at which the application server is listening.
Now comes the most confusing part: the address of the destination host. It is important to understand that, in a client-to-server port forwarding rule, the target host address is relative to the SSH server, not the client. This is the address that the SSH server will connect to when a connection needs to be forwarded. In this case, the target host address is set to 127.0.0.1 to have the SSH server connect to the application server which is running on the same machine.
Finally, the destination port specifies the port on which the target TCP/IP server is listening - in this case, 123.
The port forwarding configuration shown in the above example is strict: it minimizes the exposure of unencrypted data by constraining the SSH client to reside on the same machine as the application client, and the SSH server to reside on the same machine as the application server.
On the other hand, if you are only concerned about eavesdropping between the SSH client and the server, and do not mind unencrypted data in the local subnets, you might configure your SSH port forwarding rules like this:
This corresponds with the following C2S (client-to-server) port forwarding rule in the SSH client:
- Listen interface: 0.0.0.0 (this opens up the SSH client's forwarding socket to connections from other machines)
- Listen port: 123
- Destination host: 10.2.2.9 (the target application server is not on the same machine as the SSH server, so we need to enter its address as visible from the SSH server)
- Destination port: 123
With this setup, you only need one SSH client to forward the connections of multiple application clients; since the SSH client's listening address is configured to 0.0.0.0, the application clients do not need to reside on the same machine. With appropriately configured port forwarding rules, you can use the same SSH session to forward connections to multiple application servers, which can reside on machines different from the SSH server.
Even though our examples above only discuss client-to-server port forwarding rules, the concept of server-to-client port forwarding is entirely symmetric. Only the roles are reversed: with S2C forwarding, the listening address is relative to the SSH server, and the destination host address is relative to the SSH client.
It is a common mistake to define both a C2S as well as an S2C rule for the same forwarded connection. This is not necessary and will not work. S2C rules are required only if you are forwarding other connections which are established in the direction from the server to the client. Such connections are normally independent from, and unrelated to, those established from client to server. Only one type of rule is necessary for each connection.
Web browsing over SSH
It is possible to configure most browsers to use a SOCKS proxy for outgoing HTTP connections. This makes it possible to forward web browser traffic over an encrypted SSH connection.
The recommended browser for this purpose is Firefox, because it can be configured to resolve DNS names through the SOCKS proxy, so the names of the websites you're browsing don't leak out through DNS queries.
You will need an account at an SSH server which allows you to use port forwarding. Configure Bitvise SSH Client to connect to that SSH server, and enable the SOCKS proxy feature under the Services tab.
In Firefox, configure Bitvise SSH Client as the SOCKS proxy in Tools > Options > Advanced > Network > Connection > Settings. Use Manual proxy configuration, enter 127.0.0.1 under SOCKS proxy, and port 1080. (This is assuming you left SOCKS proxy settings in the SSH client at their defaults.)
Open a blank Firefox tab and navigate to "about:config". Find the setting:
Set this setting to true.
You are now done. Firefox will connect to websites through Bitvise SSH Client's SOCKS proxy feature, and your web traffic will be tunneled over the encrypted SSH connection between your SSH client and the SSH server.
Note that the part of the traffic between the SSH server and the web server(s) will remain unencrypted. By using SSH tunneling, you are shielding your web traffic from prying eyes in your local network or at your local Internet Service Provider. However, the plaintext of your web sessions will now be available to the SSH server administrator, as well as to the ISP through which the SSH server connects to your destination web servers.
X11 forwarding with Bitvise SSH Client
The X11 forwarding feature in Bitvise SSH Client provides one way for an SSH connection to access graphical applications running on the SSH server. X11 forwarding is an alternative to forwarding a Remote Desktop or VNC connection. It differs from Remote Desktop or VNC in that remote application windows appear seamlessly in the client's desktop, without forwarding a complete desktop. X11 forwarding is best used with Unix-style servers running applications intended to run under X11. For connections to Windows servers, Remote Desktop is the native option.
Installing an X11 server
In order to use X11 forwarding, an X11 server needs to be installed on the client. One such server is available as part of Cygwin.
To install the Cygwin X11 server without installing the entire (and large) Cygwin platform, perform the following steps in the Cygwin installer:
- Proceed to the Select Packages page. On this page, you should see a package tree with All - Default as the root.
- To prevent installing packages selected by default, cycle All - Default by clicking on the Default part until it is set to All - Uninstall. (screenshot)
- Find the package All / X11 / xorg-server: X.Org X servers and select it by changing Skip at the beginning of the line to the current stable version - that is, the first version after Skip. (screenshot)
- Proceed to the next page and complete the installation.
Starting the Cygwin X11 server
You can start the Cygwin X11 server by executing:
C:\cygwin\bin\XWin -listen tcp -multiwindow
If the X11 server starts successfully, a new X-resembling icon will appear in the task bar notification area (the system tray). To close the X11 server, right click on the icon and select Exit from the right-click menu.
If the X11 server fails to launch, an error message will be displayed. A common failure reason is that an X11 server is already running. It is possible to run multiple X11 servers on the same computer, but each will need to be associated with a unique display number. For example, to start the X11 server on display 3, you would execute:
C:\cygwin\bin\XWin :3 -listen tcp -multiwindow
Note: We do not recommend opening the firewall for the X11 server (the XWin.exe process).
Setting up Bitvise SSH Client
In the SSH client's Terminal tab, enable X11 forwarding. If your X11 server runs on a non-default display (a display other than 0), the setting X11 Forwarding - Display will need to be changed, as well. For example, if your X11 server runs on display 3, change the setting to: 127.0.0.1:3.0.
Use the Login button to establish an SSH connection. Open a terminal console, and in it, run an X11 program (e.g. xemacs). The program should appear in a new window on your screen.
Single-Click Remote Desktop Forwarding
Since Tunnelier (Bitvise SSH Client) 3.28 and later, this section is now largely obsolete. A Remote Desktop session can be launched by simply clicking on a button when the SSH session is established, and Bitvise SSH Client will setup all the settings and launch the Remote Desktop client for you.
Consult the below instructions if for some reason the automatic single-click solution fails, or if you must configure Remote Desktop to be tunnelled manually.
Securing Remote Desktop With SSH
Remote Desktop, previously known as Terminal Services, is a feature in Microsoft Windows that allows a user to interact with a Windows machine's desktop remotely from another Windows computer. The server must use a version of Windows that includes support for Remote Desktop as a server. This is included in all Server versions of Windows; and it is also included in desktop editions such as Professional, Ultimate, Enterprise, and Business; but it is not supported in Home editions. The client computer can use any version of Windows equipped with the Remote Desktop Connection client (mstsc.exe).
Although Remote Desktop takes measures to protect against passive attacks, it does not appear to provide much protection against an active attack. Also, opening port 3389 on the server means another Windows service open to remote vulnerability probing. Both issues can be avoided by routing the Remote Desktop session through SSH port forwarding. On the server machine, an SSH server, such as Bitvise SSH Server, must be installed. On the client machine, an SSH client, such as Bitvise SSH Client, must be configured so that connections to a specific local port will be forwarded to port 3389 on the Remote Desktop server. One must then direct the Remote Desktop client to connect to the SSH client instead of directly to the server, and the connection will be forwarded over the SSH-secured link.
Note that one must always be diligent in verifying the SSH server's fingerprint when establishing the SSH connection for the first time, otherwise SSH won't be better at defending against active attacks either.
Listening Port on XP vs. Win2K and Earlier
On Windows 2000 and earlier, the Terminal Services client does not support connecting to a custom listening port. For this reason, with the older Terminal Services clients, 3389 must be used as the port on which the SSH client will be listening. If this is not possible because a Terminal Services server is running on the same machine, a newer Remote Desktop client can be downloaded from Microsoft which supports connections to non-default ports.
On Windows XP, a port other than 3389 can be entered in the 'Computer' field of the initial RDC dialog box - for example, 'localhost:3390'. This is useful if you need to setup the SSH client to listen on a port other than 3389, for instance if port 3389 is already occupied by the local Remote Desktop server.
Connecting to Localhost on XP prior to SP2
Prior to Windows XP Service Pack 2, the Remote Desktop client on Windows XP explicitly prevented the user from connecting to localhost. For users who have not yet upgraded to SP2, there is a way around this limitation. Have the SSH client listen on 127.0.0.2, and connect the Remote Desktop client to 127.0.0.2 instead of localhost.
With the Windows XP SP2 version of the Remote Desktop client, it is possible to connect to localhost (127.0.0.1) as long as the port being used is other than the default (3389). Note however that connections through 127.0.0.2 do not work any more on Windows XP SP2. Because the 127.0.0.2 address is necessary prior to Windows XP SP2, the same forwarding setup will not work on SP2 as well as pre-SP2 machines.
Follow these steps if you wish to get quickly up and started with Remote Desktop over SSH. It is advised that you try to understand what is being done by each one of the steps presented. The difference between understanding and not understanding is frequently the difference between a security measure which works and one that only appears to.
- Install Bitvise SSH Server on the server (the machine you wish to access with Remote Desktop).
- No changes to the default Bitvise SSH Server configuration are required to use Remote Desktop over SSH. You may wish to make changes to the default SSH server configuration later on, to restrict what SSH features are accessible to remote users. However, for the time being, keep your SSH server settings at default until your Remote Desktop over SSH is up and running.
- Apart from installing Bitvise SSH Server, the only thing you need to do on the server is ensure that there is a Windows account which you can use to log on locally. This will normally be a Windows account which already exists and which you plan to be using to log into with Remote Desktop.
- Start the Bitvise SSH Server from the Bitvise SSH Server Control Panel.
- Install Bitvise SSH Client on the machine from which you wish to access the server.
- Configure the following settings on the Login tab in Bitvise SSH Client. Click also the 'Help' link on the Login tab for help with any of these settings.
- Host: The IP address or DNS name of the server that you are accessing.
- Port: You will normally use the default value, 22. This must match the port that Bitvise SSH Server is listening on. If you have made no changes to the default SSH server configuration to change the port it is listening on, use 22.
- Username: The Windows account name with which to log into the server. This must be a valid Windows account name with local logon permissions on the side of the server.
- Password: The password with which to log into the server, belonging to the account name specified by Username.
- Store encrypted password in profile: You may optionally wish to enable this setting so that you will not be asked to reenter the password each time when logging in after the SSH client has been restarted.
- In the C2S Forwarding
tab in Bitvise SSH Client, add a new entry and configure the following settings
for this entry. Click also the 'Help' link on the C2S Forwarding tab
for help with any of these settings.
- Status: This will be 'enabled' by default, leave it that way.
- Listen interface: The default value is 127.0.0.1. If the client machine is running Windows XP prior to Service Pack 2, change this to 127.0.0.2. If you are running Windows XP SP2, or if you are running Windows 2000 or earlier, leave this at the default value.
- List. Port: This is the local (client-side) port on which the SSH client will be listening for a connection from your Remote Desktop client. Set this to 3389 if running Windows 2000 or earlier. Otherwise, if using Windows XP, set this to 3390 or an arbitrary port number. The chosen port number needs to be reflected in your instructions to the Remote Desktop client (below). You can also execute 'netstat -an' from a command prompt and examine the output to ensure that your chosen port is not yet occupied. It is fine if there is not already a line like '0.0.0.0:(yourPortNr) ... LISTENING'.
- Destination Host: specifying localhost will work, assuming the Remote Desktop server is listening on all interfaces, which is normally the case. If it is listening on a particular interface, you can determine the interface by executing 'netstat -an' on the server and examining the output for a line like 'xxxxxx:3389 ... LISTENING'. If xxxxxx is 0.0.0.0, the Remote Desktop server is listening on all interfaces and 'localhost' will work here. Otherwise, the xxxxxx is the IP address that you need to enter in this field. Using 'localhost' will normally work though.
- Dest. Port: 3389.
- Click the Login button in Bitvise SSH Client and observe the log area for any errors. If the session is established without errors, the SSH setup is running, now you just need to connect through it with the Remote Desktop client.
- Run the Remote Desktop client. In Windows XP, you can find it through Start : All Programs : Accessories : Communications : Remote Desktop Connection. Alternately you can run it from a Windows command prompt (execute 'mstsc') or through Start : Run : 'mstsc'.
- In the Computer field, enter 127.0.0.1 if you configured the 'List. Port' setting in your C2S rule as 3389 (on Windows 2000 or earlier). On Windows XP prior to SP2, enter 127.0.0.2:xxxx, where xxxx is the port number you chose for the 'List. Port' field in your C2S rule. On Windows XP SP2 or higher, enter 127.0.0.1:xxxx, where xxxx is that same port number.
- Click Connect. The SSH session needs to be established with the C2S port forwarding rule active when you do this. If all is well, you should have a secure Remote Desktop connection to the server machine shortly.
- You can make sure that your Remote Desktop connection is going through SSH by checking the Bitvise SSH Client log area for a message saying 'Accepted client-to-server connection from ... to localhost:3389' corresponding to each connection attempt you make. Likewise, when your Remote Desktop session closes, the SSH client should output a log message stating 'Closing client-to-server forwarding channel from ... to localhost:3389'.
If you encounter problems establishing the SSH session, you will receive diagnostic information in the Bitvise SSH Client log area, as well as in the log entries recorded by the SSH server. Especially in the case of an authentication failure, the SSH server log entries will contain important diagnostic information.
Please see our contact and support page for more information and links to documents about how to go about resolving problems with Bitvise SSH Client and Server.
Securing WinVNC With SSH
VNC is a free client/server system which allows you to view a computing 'desktop' environment not only on the machine where it is running, but from anywhere on the Internet and from a wide variety of machine architectures.
You can combine WinVNC and an SSH port forwarding client/server pair, such as Bitvise SSH Client and Server, to form a secure solution for remote GUI login. Suppose you install a VNC server on machine A, and the SSH server on machine B. Machine A and machine B can be the same machine, and should generally be as close as possible, because only the connection between the VNC viewer and machine B will be secured; the connection between machine A and machine B will be unprotected. In order to securely access the VNC server from a client machine, you need to perform the following steps:
- Equip the client machine with an SSH port forwarding client. Configure this client to connect to the SSH server on machine B, and to forward connections that come to port 5900 on the client machine to port 5900 on machine A. In SSH terminology, this is client-to-server port forwarding, or 'local' forwarding. If you intend to use a VNC display other than 0, you should alter the port numbers appropriately: the VNC port number is 5900 plus the display number, e.g. 5905 for display 5.
- Configure the VNC viewer to connect to localhost. If you configured the SSH client to listen on a port other than 5900, you also need to specify the display number. E.g., specify 'localhost:5' if the SSH client is listening on port 5905.
- If the SSH and WinVNC servers both reside on the same computer, you will need to enable local loopback connections in WinVNC. See the WinVNC documentation - look for the AllowLoopback keyword under" Advanced Options.
What you need to know about the Internet
Our products can be used in ways that don't require much knowledge about the internet. You can just type in the address of the server you're connecting to, open an SFTP window and start transferring files. However, if you will be using the more advanced features of our products, such as tunneling, you will need to understand the basics of how the Internet is structured. This guide is an attempt at relaying some of that understanding.
This guide is composed of the following sections:
- IP addresses
- DNS names
- Types of IP addresses and subnets
- TCP and UDP
- Direction of TCP connections
- Connecting to the internet from office
- Connecting to the internet from home
- Dynamic IP address issues
- Virtual servers - port forwarding at the router
Every computer connected to the internet has an Internet Protocol or IP address which identifies the computer on the internet. In the currently most widely used version of the Internet Protocol - version 4 - IP addresses are 4 bytes long and are expressed in the form nn.nn.nn.nn. Each nn is a number between 0 and 255.
When you connect to a web server to browse a web page, the DNS name of the web server, e.g. www.bitvise.com, is automatically translated by the software in your machine to an IP address in the nn.nn.nn.nn form. This address is then used to connect to the actual web server.
For example, the IP address of the server hosting fogbugz.bitvise.com at the time of this writing is 220.127.116.11. Our primary website, on the other hand, is hosted on several servers, and their IP addresses are 18.104.22.168, 22.214.171.124, 126.96.36.199 and 188.8.131.52.
In a Windows Command Prompt session, you can discover the IP addresses associated with DNS names using the nslookup command: e.g. 'nslookup www.bitvise.com'.
IP addresses are difficult to remember, so the internet provides a translation service which translates memorable names into associated IP addresses. This facility is called the Domain Name System or DNS. You use DNS implicitly every time you type in an address such as 'www.bitvise.com' - your browser asks your operating system for translation into an IP address, and the operating system either returns a cached result, or inquires with a DNS server operated by your ISP. This server in turn either returns a cached result or inquires with another DNS server.
No computer is directly connected to every other computer on the internet. Instead, each computer is a member of one or more subnets. Subnets, in turn, are connected to each other by machines called routers or gateways, which belong to multiple subnets, forwarding internet traffic from one subnet to the other and reverse.
In order to successfully communicate with other computers throughout the internet, your computer must know what subnet it is part of, so that it knows what IP addresses are outside your local subnet and must be relayed through the gateway. In addition, your computer must of course also know the IP address of the gateway.
Typically, a subnet is a group of consecutive IP addresses, such as all IP addresses from 184.108.40.206 to 220.127.116.11. This is commonly expressed in either of two formats:
- The subnet mask format. Here, the subnet is expressed as 18.104.22.168 with subnet mask 255.255.255.0. The subnet mask indicates what bits of the subnet IP address indicate the actual subnet, and what bits are variable, indicating individual computers in the subnet. A byte consists of 8 bits, and 255 is 1111 1111 in binary. Therefore, 255.255.255.0 means that the first 3 bytes of the subnet IP address (11.22.33) indicate the actual subnet, and the last byte can be variable (and indicates computers in the subnet). If the subnet mask were 255.255.0.0, that would mean that the last two bytes are variable.
- The significant bits format. Here, the subnet is expressed as 22.214.171.124/24, which means subnet 126.96.36.199 with 24 significant bits. The 24 means that the first 24 bits of the subnet mask are 1, and all the following bits are 0. Thus, /24 is equivalent to a subnet mask of 255.255.255.0. /16 is equivalent to a subnet mask of 255.255.0.0. And because there are just 32 bits in an IP address, /32 indicates an IP address with no variable part: a fixed, constant IP address.
Our software expresses subnets using the significant bits format.
Types of IP addresses and subnets
There are three major types of IP addresses (or subnets) that you need to be aware of.
- Public IP addresses. Most IP addresses in the 32-bit address range have the purpose of uniquely identifying a computer on the internet. The IP address 188.8.131.52, for example, is a public IP address that uniquely identifies one of the servers hosting the www.bitvise.com website (and others). This is the type of IP address through which a server must be reachable in order to be accessible to computers throughout the internet.
- Private subnets.
Special ranges of the 32-bit IP address range have been set aside for
use in private networks, where the computers in such a network do not
need to be directly accessible from the internet as servers (but may
nevertheless access the internet through a gateway, as clients). These
- 10.0.0.0/8 (addresses from 10.0.0.0 to 10.255.255.255)
- 172.16.0.0/12 (addresses from 172.16.0.0 to 172.31.255.255)
- 192.168.0.0/16 (addresses from 192.168.0.0 to 192.168.255.255)
- Special IP ranges. There are several special purpose IP ranges, but the one you need to know about is 127.0.0.0/8 (addresses from 127.0.0.0 to 127.255.255.255). This is the local loopback range and is used to connect two programs running on the same machine. Any address in this range can be used for this kind of purpose, but the most commonly used are 127.0.0.1 and 127.0.0.2. The special DNS name 'localhost' translates to 127.0.0.1.
TCP and UDP
The Internet Protocol itself is a relatively rudimentary protocol which provides only the capability of delivering small chunks of data to other computers. The Internet Protocol does not provide reliability: chunks of data that are sent using the Internet Protocol may be lost. They also may arrive in an order different to the order in which the chunks were sent.
For some types of data transfer, the (un)reliability afforded by the Internet Protocol is fine. When streaming video, for example, it does not matter if chunks that make up intermediate frames of the video are lost. What matters is that most of the data arrives relatively quickly, allowing the video to be played with reasonable quality and on the fly. The User Datagram Protocol, or UDP, is a simple protocol layered on top of the Internet Protocol that provides this level of reliability. UDP is used for purposes such as relaying video and audio streams as well as for networked games; all environments where responsiveness and fast delivery are more important than perfect reliability.
For other types of data transfer, however, this level of reliability is not enough. When transferring a file, for example, you want to transfer all of its contents in perfect order and integrity; you don't want any chunks of it to accidentally be lost. When accessing a web page, likewise, you want all the text to be transferred without error. Data transfers that require this higher level of reliability use the Transmission Control Protocol, or TCP. Like UDP, TCP is a protocol layered on top of the Internet Protocol, but it is more complex than UDP: it contains mechanisms to ensure that data is received in order and that, if any chunks are lost, they are resent. The reliability provided by TCP has costs in terms of responsiveness. Before any data can be sent using TCP, the two computers must engage in a short back-to-forth to establish a TCP connection. If any data are lost during transmission, delivery of subsequent data awaits until the data that were lost are retransmitted and delivered. When there is a high rate of data loss on a connection, this may cause transmission to be jerky.
The majority of widely known protocols used on the internet are layered on top of TCP. These include:
- the Simple Mail Transfer Protocol (SMTP), used for email delivery;
- the Post Office Protocol (POP) and IMAP, used for email retrieval;
- the Hypertext Transfer Protocol (HTTP), used for accessing websites;
- as well as, of course, the Secure Shell protocol (SSH), which our products are about.
Direction of TCP connections
TCP connections are like phone calls: they are always initiated by one party and accepted (or not) by the other. The computer that originates the TCP connection is usually the client, and the computer that accepts it is usually the server. Sometimes, notably in the FTP protocol, a secondary TCP connection will be established in the reverse direction, from the server to the client. But, in protocols other than FTP, connections are almost always initiated by the client.
Regardless of the direction in which a TCP connection is established, data can always flow both ways. However, the direction of the TCP connection matters because it determines who the initiating party is, and is also used by network components to impose rules on whether a connection can be established.
In order to handle multiple simultaneous connections with the same computer, your computer must be able to distinguish them. To do so, each connection is assigned two port numbers, one at each end point of the connection. A connection is then uniquely identified with four pieces of information: (1) local address, (2) local port, (3) remote address, (4) remote port. Valid port numbers are between 1 and 65535. The party that originates a TCP connection usually selects a local port number at random. On the other hand, the port number of the party that accepts the connection must be known in advance by the party that originates the connection. You can confirm this by executing 'netstat -n' from a Windows Command Prompt just after loading a web page in your browser.
For example, this excerpt from 'netstat -n' output was taken just after opening www.bitvise.com in a browser.
TCP 10.10.10.123:21681 184.108.40.206:80 ESTABLISHED
The above output indicates an established TCP connection with local address 10.10.10.123, local port 21681, remote address 220.127.116.11 and remote port 80. The connection was initiated by the local machine, therefore the local port number 21681 was randomly selected, whereas the remote port number 80 is the well-known HTTP port. This is the port where the vast majority of web servers accept connection, so even when access to other ports is blocked, connections to port 80 will very likely be permitted.
Other well-known destination ports are:
- 21 - FTP (control connection)
- 22 - SSH
- 23 - Telnet
- 25 - SMTP
- 80 - HTTP
- 110 - POP3
- 143 - IMAP4
- 443 - HTTPS (HTTP over TLS or SSL)
- 1080 - SOCKS proxy
On Windows, a more exhaustive list of well-known ports can be found in the file \Windows\System32\Drivers\etc\services (open it with Notepad).
Connecting to the internet from office
In an office environment, your computer will most likely be connected to a subnet in one of the private address ranges. This means that your computer will have an IP address not unique throughout the internet, so it cannot communicate with other computers on the internet directly. However, the network administrators at your office have most likely applied one of the following solutions to allow you to access the internet.
Network address translation (NAT). In this setup, your computer directs all traffic destined to the internet through a gateway in your local subnet. This gateway has a public IP address which is unique and can be used for internet addressing. The gateway substitutes its own IP address and port in place of your computer's. When chunks of data arrive in reply, the gateway knows from the port number in the data that they must be forwarded to your computer and local port.
In this setup, your computer is led to believe that it is present on the internet with its private subnet IP address; but it isn't. The gateway is present on the internet and represents all computers in the subnet with its own public IP. All connections initiated to the internet by computers on the subnet appear to outside observers as coming from the gateway's public IP address.
- Proxy. In this setup, your computer cannot initiate connections to the internet directly. Instead, applications on your computer must contact one of several types of proxy servers residing on your local subnet, and ask the proxy server if it would kindly relay a connection to the outside. This is conceptually similar to NAT. However, whereas NAT works for all applications on your machine and requires from them no special awareness, the proxy setup works only with those applications which can connect to the internet through the proxy. The proxy setup also affords administrators more control: they can more easily restrict and monitor your traffic and permit or deny access selectively based not just on port numbers, but the content being accessed and protocols being used.
There is also a number of office environments where each computer has a separate, own public IP address. These are simple and involve no NAT or proxy servers as outlined above.
Connecting to the internet from home
From home, you usually connect to the internet through a modem - whether it is phone, cable, ISDN or DSL. In any case, you can either hook the modem directly to your computer; or, if you have multiple computers, you can buy a router, connect the router to your modem and your computers to the router.
- If you use a router, the machines connected to it are assigned addresses in a private subnet, and the router performs Network Address Translation to allow your machines to access the internet.
- If you connect the modem to your machine directly, the computer gets a public IP address directly accessible from the internet. If you then connect other machines to this machine (through a second network interface), those machines are joined to a private subnet. The directly connected machine then performs Network Address Translation to allow the other computers to access the internet.
In most cases, you will be provided a single public IP address by your internet provider. Sometimes this IP address will be fixed; this is called a static IP address. In other situations, the IP address will periodically change; this is called a dynamic IP addres. With dial-up modems, you will get a different public IP address every time you dial up. With DSL and cable modems, your IP address may change at a predefined time every day or night.
Dynamic IP address issues
The following issues correspond with a continuously changing IP address.
Whenever your public IP address changes, all ongoing TCP connections to and from your machine are terminated and must be reestablished using the new IP.
Since the IP address of your computer is unpredictable, it is difficult for others to connect to it. If you want to host any kind of network-accessible service on your machine, you need to either use a dynamic DNS service; this works by allocating you a DNS name which is regularly updated to reflect your changing IP address; or you need to implement a more pedestrian solution, such as configuring a program on your computer to periodically connect to another server and store your current IP address there, making it available for retrieval.
If you want to host a service on your home machine and find that your IP address changes periodically, the best way around this problem is to ask your ISP to grant you a static IP. They will frequently agree to do this free of charge. If this is unavailable, you can use a dynamic DNS service.
Virtual servers - port forwarding at the router
If you want to make a server accessible from the internet, but the computer on which the server will be based has only a private subnet IP addresses, there is a solution. Usually, the router which connects the private subnet to the internet can be configured to forward all incoming connections on a certain port to one of the computers inside the private network. This is called port forwarding (not the same thing as SSH port forwarding) or a 'virtual server' facility (although the server is quite real; it's just its IP address that is not).
This setup generally works just fine, but there is one thing to remember. The IP address by which the server is known to internet clients is not the IP address that the server machine actually has. This distinction between the public IP address at the router, and the private IP address of the actual server machine inside, frequently arises in SSH connection tunneling, leading to incorrect configuration if not properly understood.
Modern computers run a large number of local services (such as Windows file and printer sharing) which accept connections on various port numbers, but are meant to be accessible only from locally trusted subnets. Preventing the wider internet from accessing these services in possibly malicious ways is the purpose of ingress firewalls.
In organizations, gateways that connect the local subnet to the internet usually feature an ingress firewall. This firewall should normally be configured to allow no connections into the subnet, except connections to servers that must accept connections from the internet.
At home, your ISP will usually not protect your PC from malicious access from the internet. Instead, this task must be performed by a firewall installed on your home router, or if your computer is connected to the internet directly, a software firewall in your machine. Windows XP comes equipped with such a firewall; you should use it. Software firewall solutions are available for earlier versions of Windows.
There is another type of firewall called an egress firewall, or a firewall that filters outbound connections from your machine to the internet. This is generally software which tries to control what programs on your machine access the internet. This is intended to block malicious software from doing too much damage after it has already infected your computer. However, cleverly written malware can fool an egress firewall like this with fairly simple and straightforward deceptions. The only real medicine against malware is therefore to prevent it from infecting your computer in the first place.