SSH_CONFIG(5)

🦋 NAME 🦋

ok so this is ssh_config — the OpenSSH client configuration file, bestie. That's SSH_CONFIG(5), straight out of the File Formats Manual. yes, we're doing a whole manpage. grab a snack. 🍓

---

🌸 DESCRIPTION 🌸

so ssh(1) is out here collecting its configuration data from a bunch of places, and the order matters so much. it goes like this:

1. command-line options
2. user's configuration file (~/.ssh/config)
3. system-wide configuration file (/etc/ssh/ssh_config)

and unless the docs say otherwise, for each parameter, the first obtained value wins. periodt. 💅

the config files contain sections separated by Host specifications, and that section only applies for hosts that match one of the patterns in the specification. the matched host name is usually the one you gave on the command line (peep the CanonicalizeHostname option for the exceptions, iykyk).

since the first obtained value for each parameter is the one that gets used, the more host-specific declarations should live near the beginning of the file, and the general defaults go at the end. like, general stuff last. no cap. ✨

the file is made of keyword-argument pairs, one per line. lines starting with # and empty lines are read as comments. arguments may optionally be wrapped in double quotes (") so you can represent arguments that contain spaces. configuration options may be separated by whitespace, or by optional whitespace and exactly one = — and that second format is genuinely useful to avoid needing to quote whitespace when you're specifying config options with the -o option of ssh, scp, and sftp. 🫶

the possible keywords and their meanings are as follows (keywords are case-insensitive, arguments are case-sensitive — keep that energy in mind):

👑 Host

restricts the following declarations (all the way up to the next Host or Match keyword) so they only apply to those hosts that match one of the patterns given after the keyword. if you give more than one pattern, separate them with whitespace. a single * as a pattern gives you global defaults for all hosts. the host is usually the hostname argument you gave on the command line (see the CanonicalizeHostname keyword for exceptions).

a pattern entry may be negated by prefixing it with an exclamation mark (!). if a negated entry is matched, then the Host entry is ignored, no matter whether any other patterns on the line match. so negated matches are lowkey perfect for providing exceptions to wildcard matches. 💀

see “PATTERNS” for way more on patterns.

🔮 Match

restricts the following declarations (up to the next Host or Match keyword) so they're used only when the conditions after the Match keyword are satisfied. match conditions get specified using one or more criteria, or the single token all which always matches. the available criteria keywords are: canonical, final, exec, localnetwork, host, originalhost, tagged, command, user, localuser, and version. the all criteria must appear alone or immediately after canonical or final. other criteria may be combined arbitrarily. every criteria except all, canonical, and final requires an argument. criteria may be negated by prepending an exclamation mark (!). 🎀

ok now the criteria one by one:

• the canonical keyword matches only when the configuration file is being re-parsed after hostname canonicalization (see the CanonicalizeHostname option). this one's handy for specifying conditions that work with canonical host names only.
• the final keyword requests that the configuration be re-parsed (regardless of whether CanonicalizeHostname is enabled), and matches only during this final pass. if CanonicalizeHostname is enabled, then canonical and final both match during the same pass.
• the exec keyword executes the specified command under the user's shell. if the command returns a zero exit status then the condition is considered true. commands containing whitespace characters must be quoted. arguments to exec accept the tokens described in the “TOKENS” section.
• the localnetwork keyword matches the addresses of active local network interfaces against the supplied list of networks in CIDR format. this can be convenient for varying the effective configuration on devices that roam between networks. but note that network address is not a trustworthy criteria in many situations (e.g. when the network is automatically configured using DHCP), so be careful if you're using it to control security-sensitive configuration. 🙈

the other keywords' criteria must be single entries or comma-separated lists, and may use the wildcard and negation operators described in the “PATTERNS” section.

the criteria for the host keyword are matched against the target hostname, after any substitution by the Hostname or CanonicalizeHostname options. the originalhost keyword matches against the hostname as it was specified on the command line. 🦋

the tagged keyword matches a tag name specified by a prior Tag directive, or on the ssh(1) command line using the -P flag. the command keyword matches the remote command that has been requested, or the subsystem name that's being invoked (e.g. "sftp" for an SFTP session). the empty string will match the case where a command or tag has not been specified, i.e. 'Match tag ""'. the version keyword matches against the version string of ssh(1), for example "OpenSSH_10.0". ⭐

the user keyword matches against the target username on the remote host. the localuser keyword matches against the name of the local user running ssh(1) (this keyword may be useful in system-wide ssh_config files).

and finally, the sessiontype keyword matches the requested session type, which may be one of: shell for interactive sessions, exec for command execution sessions, subsystem for subsystem invocations such as sftp(1), or none for transport-only sessions, such as when ssh(1) is started with the -N flag. slay. 💅

🎀 AddKeysToAgent

specifies whether keys should be automatically added to a running ssh-agent(1). here's the tea:

• yes → if a key is loaded from a file, the key and its passphrase get added to the agent with the default lifetime, as if by ssh-add(1).
• ask → ssh(1) will require confirmation using the SSH_ASKPASS program before adding a key (see ssh-add(1) for details).
• confirm → every use of the key must be confirmed, as if the -c option was specified to ssh-add(1).
• no → no keys are added to the agent.

alternately, this option may be specified as a time interval using the format described in the “TIME FORMATS” section of sshd_config(5) to specify the key's lifetime in ssh-agent(1), after which it will automatically be removed. so the argument must be no (the default), yes, confirm (optionally followed by a time interval), ask, or a time interval. ✨

🌈 AddressFamily

specifies which address family to use when connecting. valid arguments are any (the default), inet (use IPv4 only), or inet6 (use IPv6 only).

🧚 BatchMode

if set to yes, user interaction like password prompts and host key confirmation requests gets disabled. this option is useful in scripts and other batch jobs where no user is around to interact with ssh(1). the argument must be yes or no (the default).

💜 BindAddress

use the specified address on the local machine as the source address of the connection. only useful on systems with more than one address.

🫶 BindInterface

use the address of the specified interface on the local machine as the source address of the connection.

🌸 CanonicalDomains

when CanonicalizeHostname is enabled, this option specifies the list of domain suffixes in which to search for the specified destination host.

🦋 CanonicalizeFallbackLocal

specifies whether to fail with an error when hostname canonicalization fails. the default, yes, will attempt to look up the unqualified hostname using the system resolver's search rules. a value of no will cause ssh(1) to fail instantly if CanonicalizeHostname is enabled and the target hostname cannot be found in any of the domains specified by CanonicalDomains. 😭

👑 CanonicalizeHostname

controls whether explicit hostname canonicalization is performed. the default, no, doesn't do any name rewriting at all and lets the system resolver handle all hostname lookups. if set to yes then, for connections that do not use a ProxyCommand or ProxyJump, ssh(1) will attempt to canonicalize the hostname specified on the command line using the CanonicalDomains suffixes and CanonicalizePermittedCNAMEs rules. if CanonicalizeHostname is set to always, then canonicalization gets applied to proxied connections too. 🔥

if this option is enabled, then the configuration files are processed again using the new target name to pick up any new configuration in matching Host and Match stanzas. a value of none disables the use of a ProxyJump host.

💅 CanonicalizeMaxDots

specifies the maximum number of dot characters in a hostname before canonicalization is disabled. the default, 1, allows a single dot (i.e. hostname.subdomain).

⭐ CanonicalizePermittedCNAMEs

specifies rules to determine whether CNAMEs should be followed when canonicalizing hostnames. the rules consist of one or more arguments of source_domain_list:target_domain_list, where source_domain_list is a pattern-list of domains that may follow CNAMEs in canonicalization, and target_domain_list is a pattern-list of domains that they may resolve to.

for example, "*.a.example.com:*.b.example.com,*.c.example.com" will allow hostnames matching "*.a.example.com" to be canonicalized to names in the "*.b.example.com" or "*.c.example.com" domains.

a single argument of "none" causes no CNAMEs to be considered for canonicalization. this is the default behaviour. 🎀

CASignatureAlgorithms 🖊️

okay so this one is all about which algorithms are allowed for signing of certificates by certificate authorities (CAs). bestie, the default is handled system-wide by crypto-policies(7) — and all the info about defaults, how to modify the defaults, and how to customize existing policies with sub-policies lives in update-crypto-policies(8).

Here's the vibe for the list:

• If your list starts with a + character, the algorithms you named get appended to the default set instead of replacing it ✨
• If it starts with a - character, the ones you named (wildcards included, iykyk) get removed from the default set instead of replacing it

And like, ssh(1) will simply not accept host certificates signed using algorithms other than the ones you specified. no cap, it just won't. 💅

---

CertificateFile 📜

Specifies a file from which the user's certificate is read. Important bit: a corresponding private key must be provided separately in order to actually use this certificate, and you can feed it in either from an IdentityFile directive or the -i flag to ssh(1), via ssh-agent(1), or via a PKCS11Provider or SecurityKeyProvider.

Arguments to CertificateFile may use the tilde syntax to refer to a user's home directory, the tokens described in the “TOKENS” section, and environment variables as described in the “ENVIRONMENT VARIABLES” section. 🏠

And yes, you can absolutely have multiple certificate files specified in configuration files — they'll be tried in sequence. Multiple CertificateFile directives just add to the list of certificates used for authentication. stacking them like 🎀

---

ChannelTimeout ⏳

Specifies whether and how quickly ssh(1) should close inactive channels. The timeouts are written as one or more “type=interval” pairs separated by whitespace, where the “type” has to be the special keyword global or a channel type name from the list below — and it can optionally contain wildcard characters.

The timeout value “interval” is in seconds, or you can use any of the units documented in the “TIME FORMATS” section. Example time: session=5m would cause interactive sessions to terminate after five minutes of inactivity. Specifying a zero value disables the inactivity timeout. 🕛

The special timeout global applies to all active channels, taken together. Traffic on any active channel will reset the timeout, but when the timeout expires then all open channels get closed. Heads up: this global timeout is not matched by wildcards and must be specified explicitly.

The available channel type names include:

• agent-connection — open connections to ssh-agent(1).
• direct-tcpip, direct-streamlocal@openssh.com — open TCP or Unix socket (respectively) connections that have been established from an ssh(1) local forwarding, i.e. LocalForward or DynamicForward.
• forwarded-tcpip, forwarded-streamlocal@openssh.com — open TCP or Unix socket (respectively) connections that have been established to an sshd(8) listening on behalf of an ssh(1) remote forwarding, i.e. RemoteForward.
• session — the interactive main session, including shell session, command execution, scp(1), sftp(1), etc.
• tun-connection — open TunnelForward connections.
• x11-connection — open X11 forwarding sessions.

⚠️ Real talk: in all the above cases, terminating an inactive session does not guarantee to remove all resources associated with the session — e.g. shell processes or X11 clients relating to the session may continue to execute. 😭

Moreover, terminating an inactive channel or session does not necessarily close the SSH connection, nor does it prevent a client from requesting another channel of the same type. In particular, expiring an inactive forwarding session does not prevent another identical forwarding from being subsequently created. it's giving "the door closes but the party continues" 🫠

The default is not to expire channels of any type for inactivity.

---

CheckHostIP 🔍

If set to yes, ssh(1) will additionally check the host IP address in the known_hosts file. This is how it can detect if a host key changed due to DNS spoofing — and it'll add addresses of destination hosts to ~/.ssh/known_hosts in the process, regardless of the setting of StrictHostKeyChecking. If the option is set to no (the default), the check will not be executed. simple as that. 🦋

---

Ciphers 🔐

The default is handled system-wide by crypto-policies(7). Info about defaults, how to modify the defaults and how to customize existing policies with sub-policies are present in manual page update-crypto-policies(8).

Specifies the ciphers allowed and their order of preference. Multiple ciphers must be comma-separated. Now the prefix lore:

• + → the specified ciphers get appended to the built-in openssh default set instead of replacing them
• - → the specified ciphers (including wildcards) get removed from the built-in openssh default set instead of replacing them
• ^ → the specified ciphers get placed at the head of the built-in openssh default set 👑

The supported ciphers are:

• 3des-cbc
• aes128-cbc
• aes192-cbc
• aes256-cbc
• aes128-ctr
• aes192-ctr
• aes256-ctr
• aes128-gcm@openssh.com
• aes256-gcm@openssh.com
• chacha20-poly1305@openssh.com

You can also grab the list of available ciphers using "ssh -Q cipher". so handy fr fr 🌸

---

ClearAllForwardings 🧹

Specifies that all local, remote, and dynamic port forwardings specified in the configuration files or on the command line be cleared. This one is primarily useful when used from the ssh(1) command line to clear port forwardings set in configuration files, and it's automatically set by scp(1) and sftp(1). The argument must be yes or no (the default). decluttering era ✨

---

Compression 🗜️

Specifies whether to use compression. The argument must be yes or no (the default). that's it, that's the whole option 💜

---

ConnectionAttempts 🔁

Specifies the number of tries (one per second) to make before exiting. The argument must be an integer. This may be useful in scripts if the connection sometimes fails. The default is 1. persistence, but make it polite 🫶

---

ConnectTimeout ⏱️

Specifies the timeout (in seconds) used when connecting to the SSH server, instead of using the default system TCP timeout. This timeout is applied both to establishing the connection and to performing the initial SSH protocol handshake and key exchange. so it's guarding both the knock and the secret handshake 🤝

---

ControlMaster 🎛️

Enables the sharing of multiple sessions over a single network connection — the multiplexing icon 👑. When set to yes, ssh(1) will listen for connections on a control socket specified using the ControlPath argument. Additional sessions can connect to this socket using the same ControlPath with ControlMaster set to no (the default). These sessions will try to reuse the master instance's network connection rather than initiating new ones, but will fall back to connecting normally if the control socket does not exist, or is not listening.

Setting this to ask will cause ssh(1) to listen for control connections, but require confirmation using ssh-askpass(1). If the ControlPath cannot be opened, ssh(1) will continue without connecting to a master instance.

X11 and ssh-agent(1) forwarding is supported over these multiplexed connections, however the display and agent forwarded will be the one belonging to the master connection — i.e. it is not possible to forward multiple displays or agents. one master, one vibe 😤

Two additional options allow for opportunistic multiplexing: try to use a master connection but fall back to creating a new one if one does not already exist:

• auto
• autoask — the latter requires confirmation like the ask option

---

ControlPath 🛤️

Specify the path to the control socket used for connection sharing as described in the ControlMaster section above, or the string none to disable connection sharing. Arguments to ControlPath may use the tilde syntax to refer to a user's home directory, the tokens described in the “TOKENS” section, and environment variables as described in the “ENVIRONMENT VARIABLES” section.

It is recommended that any ControlPath used for opportunistic connection sharing include at least %h, %p, and %r (or alternatively %C) and be placed in a directory that is not writable by other users. This ensures that shared connections are uniquely identified. like a name tag for your socket 🏷️

---

ControlPersist 🌙

When used in conjunction with ControlMaster, specifies that the master connection should remain open in the background (waiting for future client connections) after the initial client connection has been closed. The breakdown:

• no (the default) → the master connection will not be placed into the background, and will close as soon as the initial client connection is closed
• yes or 0 → the master connection will remain in the background indefinitely (until killed or closed via a mechanism such as the "ssh -O exit")
• a time in seconds, or a time in any of the formats documented in sshd_config(5) → the backgrounded master connection will automatically terminate after it has remained idle (with no client connections) for the specified time

the master connection said "i'll wait" 💅

---

DynamicForward 🌈

Specifies that a TCP port on the local machine be forwarded over the secure channel, and the application protocol is then used to determine where to connect to from the remote machine.

Okay so the argument has to be [bind_address:]port, no cap. IPv6 addresses can be specified by enclosing addresses in square brackets. By default, the local port is bound in accordance with the GatewayPorts setting. However, an explicit bind_address may be used to bind the connection to a specific address. The bind_address of localhost means the listening port gets bound for local use only, while an empty address or * means the port should be available from all interfaces.

Currently the SOCKS4 and SOCKS5 protocols are supported, and ssh(1) will act as a SOCKS server. Multiple forwardings may be specified, and additional forwardings can be given on the command line. Only the superuser can forward privileged ports. 💅

✨ EnableEscapeCommandline ✨

Enables the command line option in the EscapeChar menu for interactive sessions (default ~C). By default, the command line is disabled. That’s the whole vibe.

🎀 EnableSSHKeysign 🎀

Setting this option to yes in the global client configuration file /etc/ssh/ssh_config enables the use of the helper program ssh-keysign(8) during HostbasedAuthentication. The argument must be yes or no (the default). This option should be placed in the non-hostspecific section. See ssh-keysign(8) for more information. ssh-keysign should be installed explicitly.

🦋 EscapeChar 🦋

Sets the escape character (default: ~). The escape character can also be set on the command line. The argument should be a single character, ^ followed by a letter, or none to disable the escape character entirely (making the connection transparent for binary data).

🌸 ExitOnForwardFailure 🌸

Specifies whether ssh(1) should terminate the connection if it cannot set up all requested dynamic, tunnel, local, and remote port forwardings (e.g. if either end is unable to bind and listen on a specified port). Note that ExitOnForwardFailure does not apply to connections made over port forwardings and will not, for example, cause ssh(1) to exit if TCP connections to the ultimate forwarding destination fail. The argument must be yes or no (the default).

💅 FingerprintHash 💅

Specifies the hash algorithm used when displaying key fingerprints. Valid options are: md5 and sha256 (the default).

🫶 ForkAfterAuthentication 🫶

Requests ssh to go to background just before command execution. This is useful if ssh is going to ask for passwords or passphrases, but the user wants it in the background. This implies the StdinNull configuration option being set to “yes”. The recommended way to start X11 programs at a remote site is with something like ssh -f host xterm, which is the same as ssh host xterm if the ForkAfterAuthentication configuration option is set to “yes”.

If the ExitOnForwardFailure configuration option is set to “yes”, then a client started with the ForkAfterAuthentication configuration option being set to “yes” will wait for all remote port forwards to be successfully established before placing itself in the background. The argument to this keyword must be yes (same as the -f option) or no (the default).

😭 ForwardAgent 😭

Specifies whether the connection to the authentication agent (if any) will be forwarded to the remote machine. The argument may be yes, no (the default), an explicit path to an agent socket or the name of an environment variable (beginning with $) in which to find the path.

⚠️ Agent forwarding should be enabled with caution. Users with the ability to bypass file permissions on the remote host (for the agent's Unix-domain socket) can access the local agent through the forwarded connection. An attacker cannot obtain key material from the agent, however they can perform operations on the keys that enable them to authenticate using the identities loaded into the agent.

🔥 ForwardX11 🔥

Specifies whether X11 connections will be automatically redirected over the secure channel and DISPLAY set. The argument must be yes or no (the default).

⚠️ X11 forwarding should be enabled with caution. Users with the ability to bypass file permissions on the remote host (for the user's X11 authorization database) can access the local X11 display through the forwarded connection. An attacker may then be able to perform activities such as keystroke monitoring if the ForwardX11Trusted option is also enabled.

👑 ForwardX11Timeout 👑

Specify a timeout for untrusted X11 forwarding using the format described in the “TIME FORMATS” section of sshd_config(5). X11 connections received by ssh(1) after this time will be refused. Setting ForwardX11Timeout to zero will disable the timeout and permit X11 forwarding for the life of the connection. The default is to disable untrusted X11 forwarding after twenty minutes has elapsed.

🌈 ForwardX11Trusted 🌈

If this option is set to yes, remote X11 clients will have full access to the original X11 display.

If this option is set to no (the default), remote X11 clients will be considered untrusted and prevented from stealing or tampering with data belonging to trusted X11 clients. Furthermore, the xauth(1) token used for the session will be set to expire after 20 minutes. Remote clients will be refused access after this time.

See the X11 SECURITY extension specification for full details on the restrictions imposed on untrusted clients.

⭐ GatewayPorts ⭐

Specifies whether remote hosts are allowed to connect to local forwarded ports. By default, ssh(1) binds local port forwardings to the loopback address. This prevents other remote hosts from connecting to forwarded ports. GatewayPorts can be used to specify that ssh should bind local port forwardings to the wildcard address, thus allowing remote hosts to connect to forwarded ports. The argument must be yes or no (the default).

🍓 GlobalKnownHostsFile 🍓

Specifies one or more files to use for the global host key database, separated by whitespace. The default is /etc/ssh/ssh_known_hosts, /etc/ssh/ssh_known_hosts2.

🧚 GSSAPIAuthentication 🧚

Specifies whether user authentication based on GSSAPI is allowed. The default is no.

💜 GSSAPIClientIdentity 💜

If set, specifies the GSSAPI client identity that ssh should use when connecting to the server. The default is unset, which means that the default identity will be used.

🎀 GSSAPIDelegateCredentials 🎀

Forward (delegate) credentials to the server. The default is no.

🦋 GSSAPIKeyExchange 🦋

Specifies whether key exchange based on GSSAPI may be used. When using GSSAPI key exchange the server need not have a host key. The default is “no”.

🌸 GSSAPIRenewalForcesRekey 🌸

If set to “yes” then renewal of the client's GSSAPI credentials will force the rekeying of the ssh connection. With a compatible server, this will delegate the renewed credentials to a session on the server.

Checks are made to ensure that credentials are only propagated when the new credentials match the old ones on the originating client and where the receiving server still has the old set in its cache.

The default is “no”.

For this to work GSSAPIKeyExchange needs to be enabled in the server and also used by the client.

💅 GSSAPIServerIdentity 💅

If set, specifies the GSSAPI server identity that ssh should expect when connecting to the server. The default is unset, which means that the expected GSSAPI server identity will be determined from the target hostname.

🫶 GSSAPITrustDns 🫶

Set to “yes” to indicate that the DNS is trusted to securely canonicalize the name of the host being connected to. If “no”, the hostname entered on the command line will be passed untouched to the GSSAPI library. The default is “no”.

😭 GSSAPIKexAlgorithms 😭

The default is handled system-wide by crypto-policies(7). Information about defaults, how to modify the defaults and how to customize existing policies with sub-policies are present in manual page update-crypto-policies(8).

The list of key exchange algorithms that are offered for GSSAPI key exchange. Possible values are:

• gss-gex-sha1-
• gss-group1-sha1-
• gss-group14-sha1-
• gss-group14-sha256-
• gss-group16-sha512-
• gss-nistp256-sha256-
• gss-curve25519-sha256-

This option only applies to connections using GSSAPI.

The list of supported key exchange algorithms may also be obtained using ssh -Q GSSAPIKexAlgorithms.

🔥 HashKnownHosts 🔥

Indicates that ssh(1) should hash host names and addresses when they are added to ~/.ssh/known_hosts. These hashed names may be used normally by ssh(1) and sshd(8), but they do not visually reveal identifying information if the file's contents are disclosed. The default is no. Note that existing names and addresses in known hosts files will not be converted automatically, but may be manually hashed using ssh-keygen(1).

👑 HostbasedAcceptedAlgorithms 👑

The default is handled system-wide by crypto-policies(7). Information about defaults, how to modify the defaults and how to customize existing policies with sub-policies are present in manual page update-crypto-policies(8).

Specifies the signature algorithms that will be used for hostbased authentication, as a comma-separated list of patterns. And okay the prefix thing is genuinely iconic, so listen up:

• + at the start — the algorithms you list get appended to the built-in openssh default set instead of replacing it 💫
• - at the start — the algorithms you list (wildcards included, bestie) get removed from the built-in openssh default set instead of replacing it
• ^ at the start — the algorithms you list get placed at the head of the built-in openssh default set, main character position 👑

The -Q option of ssh(1) may be used to list supported signature algorithms. This was formerly named HostbasedKeyTypes.

---

🫶 HostbasedAuthentication

Specifies whether to try rhosts based authentication with public key authentication. The argument must be yes or no (the default).

👑 HostKeyAlgorithms

Specifies the host key signature algorithms that the client wants to use, in order of preference. The prefix rules are the same cute little family as above:

• + at the start — your list gets appended to the default set instead of replacing it
• - at the start — your algorithms (including wildcards) get removed from the default set instead of replacing it
• ^ at the start — your algorithms get placed at the head of the default set

The default for this option is literally this whole list, no cap:

ssh-ed25519-cert-v01@openssh.com,
ecdsa-sha2-nistp256-cert-v01@openssh.com,
ecdsa-sha2-nistp384-cert-v01@openssh.com,
ecdsa-sha2-nistp521-cert-v01@openssh.com,
sk-ssh-ed25519-cert-v01@openssh.com,
sk-ecdsa-sha2-nistp256-cert-v01@openssh.com,
rsa-sha2-512-cert-v01@openssh.com,
rsa-sha2-256-cert-v01@openssh.com,
ssh-ed25519,
ecdsa-sha2-nistp256,ecdsa-sha2-nistp384,ecdsa-sha2-nistp521,
sk-ecdsa-sha2-nistp256@openssh.com,
sk-ssh-ed25519@openssh.com,
rsa-sha2-512,rsa-sha2-256

If hostkeys are known for the destination host then this default is modified to prefer their algorithms. Cute of it, honestly.

The list of available signature algorithms may also be obtained using "ssh -Q HostKeyAlgorithms".

Bestie, heads up: the proposed HostKeyAlgorithms during KEX are limited to the set of algorithms that is defined in PubkeyAcceptedAlgorithms and therefore they are indirectly affected by system-wide crypto_policies(7). And crypto_policies(7) can not handle the list of host key algorithms directly as doing so would break the order given by the known_hosts file. So, like, don't try it 💀

🎀 HostKeyAlias

Specifies an alias that should be used instead of the real host name when looking up or saving the host key in the host key database files and when validating host certificates. This option is useful for tunneling SSH connections or for multiple servers running on a single host. Basically a little nickname for your host 🏷️

🦋 Hostname

Specifies the real host name to log into. This can be used to specify nicknames or abbreviations for hosts. Arguments to Hostname accept the tokens described in the “TOKENS” section. Numeric IP addresses are also permitted (both on the command line and in Hostname specifications). The default is the name given on the command line.

💅 IdentitiesOnly

Specifies that ssh(1) should only use the configured authentication identity and certificate files (either the default files, or those explicitly configured in the ssh_config files or passed on the ssh(1) command-line), even if ssh-agent(1) or a PKCS11Provider or SecurityKeyProvider offers more identities. The argument to this keyword must be yes or no (the default). This option is intended for situations where ssh-agent offers many different identities — like, when your agent is being a little extra and you need it to chill 🧚

🌸 IdentityAgent

Specifies the Unix-domain socket used to communicate with the authentication agent.

This option overrides the SSH_AUTH_SOCK environment variable and can be used to select a specific agent. Setting the socket name to none disables the use of an authentication agent. If the string "SSH_AUTH_SOCK" is specified, the location of the socket will be read from the SSH_AUTH_SOCK environment variable. Otherwise if the specified value begins with a ‘$’ character, then it will be treated as an environment variable containing the location of the socket.

Arguments to IdentityAgent may use the tilde syntax to refer to a user's home directory, the tokens described in the “TOKENS” section and environment variables as described in the “ENVIRONMENT VARIABLES” section.

💖 IdentityFile

Specifies a file from which the user's ECDSA, authenticator-hosted ECDSA, Ed25519, authenticator-hosted Ed25519 or RSA authentication identity is read. You can also specify a public key file to use the corresponding private key that is loaded in ssh-agent(1) when the private key file is not present locally. The default is ~/.ssh/id_rsa, ~/.ssh/id_ecdsa, ~/.ssh/id_ecdsa_sk, ~/.ssh/id_ed25519 and ~/.ssh/id_ed25519_sk. Additionally, any identities represented by the authentication agent will be used for authentication unless IdentitiesOnly is set. If no certificates have been explicitly specified by CertificateFile, ssh(1) will try to load certificate information from the filename obtained by appending -cert.pub to the path of a specified IdentityFile.

Arguments to IdentityFile may use the tilde syntax to refer to a user's home directory or the tokens described in the “TOKENS” section. Alternately an argument of none may be used to indicate no identity files should be loaded.

It is possible to have multiple identity files specified in configuration files; all these identities will be tried in sequence. Multiple IdentityFile directives will add to the list of identities tried (this behaviour differs from that of other configuration directives). Yeah, she's special, she plays by her own rules 💁‍♀️

IdentityFile may be used in conjunction with IdentitiesOnly to select which identities in an agent are offered during authentication. IdentityFile may also be used in conjunction with CertificateFile in order to provide any certificate also needed for authentication with the identity.

The authentication identity can be also specified in a form of PKCS#11 URI starting with a string pkcs11:. There is supported a subset of the PKCS#11 URI as defined in RFC 7512 (implemented path arguments id, manufacturer, object, token and query arguments module-path and pin-value). The URI can not be in quotes.

🌈 IgnoreUnknown

Specifies a pattern-list of unknown options to be ignored if they are encountered in configuration parsing. This may be used to suppress errors if ssh_config contains options that are unrecognised by ssh(1). It is recommended that IgnoreUnknown be listed early in the configuration file as it will not be applied to unknown options that appear before it. So put it first, bestie, or it literally cannot help you 🙈

✨ Include

Include the specified configuration file(s). Multiple pathnames may be specified and each pathname may contain glob(7) wildcards, tokens as described in the “TOKENS” section, environment variables as described in the “ENVIRONMENT VARIABLES” section and, for user configurations, shell-like ‘~’ references to user home directories. Wildcards will be expanded and processed in lexical order. Files without absolute paths are assumed to be in ~/.ssh if included in a user configuration file or /etc/ssh if included from the system configuration file. Include directive may appear inside a Match or Host block to perform conditional inclusion. Nesting! We love a nesting moment 🪆

🍓 IPQoS

Specifies the Differentiated Services Field Codepoint (DSCP) value for connections. Accepted values are af11, af12, af13, af21, af22, af23, af31, af32, af33, af41, af42, af43, cs0, cs1, cs2, cs3, cs4, cs5, cs6, cs7, ef, le, a numeric value, or none to use the operating system default. This option may take one or two arguments, separated by whitespace. If one argument is specified, it is used as the packet class unconditionally. If two values are specified, the first is automatically selected for interactive sessions and the second for non-interactive sessions. The default is ef (Expedited Forwarding) for interactive sessions and none (the operating system default) for non-interactive sessions.

⭐ KbdInteractiveAuthentication

Specifies whether to use keyboard-interactive authentication. The argument to this keyword must be yes (the default) or no. ChallengeResponseAuthentication is a deprecated alias for this — deprecated, so, like, move on 💀

🔥 KbdInteractiveDevices

Specifies the list of methods to use in keyboard-interactive authentication. Multiple method names must be comma-separated. The default is to use the server specified list. The methods available vary depending on what the server supports. For an OpenSSH server, it may be zero or more of: bsdauth and pam.

🧚 KexAlgorithms

The default is handled system-wide by crypto-policies(7). Information about defaults, how to modify the defaults and how to customize existing policies with sub-policies are present in manual page update-crypto-policies(8).

Specifies the permitted KEX (Key Exchange) algorithms that will be used and their preference order. The selected algorithm will be the first algorithm in this list that the server also supports. Multiple algorithms must be comma-separated.

• If your list starts with a +, those algorithms get appended to the built-in openssh default set instead of replacing it ✨
• Start with - and the algorithms you name (wildcards included, bestie 🌈) get removed from the built-in openssh default set instead of replacing it
• Start with ^ and your algorithms get placed at the head of the built-in openssh default set, front of the line, main character energy 👑

You can also just ask for the list of supported key exchange algorithms with ssh -Q kex — iykyk 💅

---

🌸 KnownHostsCommand

Tells ssh where to run a command that gives it a list of host keys — in addition to the ones already listed in UserKnownHostsFile and GlobalKnownHostsFile. This command gets executed after the files have been read 📖

It can write host key lines to standard output in the exact same format as the usual files (that format lives in the “VERIFYING HOST KEYS” section in ssh(1)). Arguments to KnownHostsCommand accept the tokens described in the “TOKENS” section 🎀

Heads up, because this one's a little extra: the command may be invoked multiple times per connection — once when preparing the preference list of host key algorithms to use, again to obtain the host key for the requested host name and, if CheckHostIP is enabled, one more time to obtain the host key matching the server's address 👀

And if the command exits abnormally or returns a non-zero exit status, the connection is terminated. No vibes, no negotiation, just gone 💀

🦋 LocalCommand

Specifies a command to execute on the local machine after successfully connecting to the server. The command string extends to the end of the line, and it's executed with the user's shell. Arguments to LocalCommand accept the tokens described in the “TOKENS” section ✨

The command runs synchronously and does not have access to the session of the ssh(1) that spawned it. It should not be used for interactive commands — so like, don't try to put your whole interactive terminal app in here, it's not that girl 🙈

Also this directive is ignored unless PermitLocalCommand has been enabled. Gatekept by design 🔒

🎀 LocalForward

Specifies that a TCP port or Unix-domain socket on the local machine be forwarded over the secure channel to the specified host and port (or Unix-domain socket) from the remote machine 💫

• For a TCP port, the first argument must be [bind_address:]port or a Unix domain socket path
• The second argument is the destination and may be host:hostport or a Unix domain socket path if the remote host supports it

IPv6 addresses can be specified by enclosing addresses in square brackets 🧚

If either argument contains a / in it, that argument will be interpreted as a Unix-domain socket (on the corresponding host) rather than a TCP port. Slash = socket, that's the rule 🔥

Multiple forwardings may be specified, and additional forwardings can be given on the command line. Only the superuser can forward privileged ports 👑

By default, the local port is bound in accordance with the GatewayPorts setting. However, an explicit bind_address may be used to bind the connection to a specific address:

• A bind_address of localhost means the listening port is bound for local use only
• An empty address or * means the port should be available from all interfaces

Unix domain socket paths may use the tokens described in the “TOKENS” section and environment variables as described in the “ENVIRONMENT VARIABLES” section 🌈

⭐ LogLevel

Gives the verbosity level used when logging messages from ssh(1). The possible values are: QUIET, FATAL, ERROR, INFO, VERBOSE, DEBUG, DEBUG1, DEBUG2, and DEBUG3. The default is INFO 📋

DEBUG and DEBUG1 are equivalent. DEBUG2 and DEBUG3 each specify higher levels of verbose output — yes it escalates, yes it gets loud 🔥

💜 LogVerbose

Lets you specify one or more overrides to LogLevel. An override consists of one or more pattern lists that match the source file, function and line number to force detailed logging for 🕵️‍♀️

For example, an override pattern of:

kex.c:*:1000,*:kex_exchange_identification():*,packet.c:*

would enable detailed logging for line 1000 of kex.c, everything in the kex_exchange_identification() function, and all code in the packet.c file. Very "I know exactly which line is misbehaving" of you 💅

This option is intended for debugging and no overrides are enabled by default. Ngl that's probably for the best 😌

🍓 MACs

The default is handled system-wide by crypto-policies(7). Information about defaults, how to modify the defaults and how to customize existing policies with sub-policies are present in manual page update-crypto-policies(8) 📚

Specifies the MAC (message authentication code) algorithms in order of preference. The MAC algorithm is used for data integrity protection 🛡️ Multiple algorithms must be comma-separated.

• If the specified list begins with a + character, then the specified algorithms will be appended to the built-in openssh default set instead of replacing them
• If the specified list begins with a - character, then the specified algorithms (including wildcards) will be removed from the built-in openssh default set instead of replacing them
• If the specified list begins with a ^ character, then the specified algorithms will be placed at the head of the built-in openssh default set

The algorithms that contain -etm calculate the MAC after encryption (encrypt-then-mac). These are considered safer and their use recommended ✨ Encrypt-then-mac is the moment, periodt 💖

The list of available MAC algorithms may also be obtained using ssh -Q mac 🎀

🌷 NoHostAuthenticationForLocalhost

Disables host authentication for localhost (loopback addresses). The argument to this keyword must be yes or no (the default) 💫

🧚 NumberOfPasswordPrompts

Specifies the number of password prompts before giving up. The argument to this keyword must be an integer. The default is 3 🔢 Three tries and then we're done, bestie 😭

🦋 ObscureKeystrokeTiming

Specifies whether ssh(1) should try to obscure inter-keystroke timings from passive observers of network traffic 👀

If enabled, then for interactive sessions, ssh(1) will send keystrokes at fixed intervals of a few tens of milliseconds and will send fake keystroke packets for some time after typing ceases. Literally the "no, I'm always typing" of network protocols 💅

The argument to this keyword must be yes, no or an interval specifier of the form interval:milliseconds (e.g. interval:80 for 80 milliseconds). The default is to obscure keystrokes using a 20ms packet interval.

Note that smaller intervals will result in higher fake keystroke packet rates 🔥 More noise, more packets, that's the tradeoff.

🌈 PasswordAuthentication

Specifies whether to use password authentication. The argument to this keyword must be yes (the default) or no 🔑

🎀 PermitLocalCommand

Allow local command execution via the LocalCommand option or using the !command escape sequence in ssh(1). The argument must be yes or no (the default) ✨

👑 PermitRemoteOpen

Specifies the destinations to which remote TCP port forwarding is permitted when RemoteForward is used as a SOCKS proxy. The forwarding specification must be one of the following forms:

PermitRemoteOpen host:port
PermitRemoteOpen IPv4_addr:port
PermitRemoteOpen [IPv6_addr]:port

Multiple forwards may be specified by separating them with whitespace 📝

• An argument of any can be used to remove all restrictions and permit any forwarding requests
• An argument of none can be used to prohibit all forwarding requests
• The wildcard * can be used for host or port to allow all hosts or ports respectively

Otherwise, no pattern matching or address lookups are performed on supplied names — it takes them at face value, no cap 🧐

💜 PKCS11Provider

Specifies which PKCS#11 provider to use or none to indicate that no provider should be used (the default). The argument to this keyword is a path to the PKCS#11 shared library ssh(1) should use to communicate with a PKCS#11 token providing keys for user authentication 🔐

⭐ Port

Specifies the port number to connect on the remote host. The default is 22 🚪 The classic. Iconic. Never not relevant.

🌸 PreferredAuthentications

Specifies the order in which the client should try authentication methods. This allows a client to prefer one method (e.g. keyboard-interactive) over another method (e.g. password) 🥇

The default is:

gssapi-with-mic,hostbased,publickey,
keyboard-interactive,password

🦋 ProxyCommand

Specifies the command to use to connect to the server. The command string extends to the end of the line, and is executed using the user's shell exec directive to avoid a lingering shell process 🐚

Arguments to ProxyCommand accept the tokens described in the “TOKENS” section. The command can be basically anything, and should read from its standard input and write to its standard output. It should eventually connect an sshd(8) server running on some machine, or execute sshd -i somewhere 📡

Host key management will be done using the Hostname of the host being connected (defaulting to the name typed by the user). Setting the command to none disables this option entirely 🚫

Note that CheckHostIP is not available for connects with a proxy command. Just so you know 😌

This directive is useful in conjunction with nc(1) and its proxy support. For example, the following directive would connect via an HTTP proxy at 192.0.2.0:

ProxyCommand /usr/bin/nc -X connect -x 192.0.2.0:8080 %h %p

🎀 ProxyJump

Specifies one or more jump proxies as either [user@]host[:port] or an ssh URI 🦘

Multiple proxies may be separated by comma characters and will be visited sequentially. Setting this option will cause ssh(1) to connect to the target host by first making an ssh(1) connection to the specified ProxyJump host and then establishing a TCP forwarding to the ultimate target from there — hop by hop, look at her go ✨

Setting the host to none disables this option entirely 💫

Also lowkey important: this option is gonna fight with the ProxyCommand option — whichever one shows up first wins, and the later instance of the other one just… doesn't take effect. Periodt. 💅

And also (sorry, two notes in one, I know 😭): the config for your destination host — whether you gave it on the command line or in the config file — is not generally applied to the jump hosts. So if you need specific config for the jump hosts, put it in ~/.ssh/config. That's the move. 🫶

• ProxyUseFdpass — says that ProxyCommand will hand a connected file descriptor back to ssh(1) instead of sticking around to execute and pass data. The default is no. ✨
• PubkeyAcceptedAlgorithms — okay so first thing's first, the default here is handled system-wide by crypto-policies(7). All the info about what the defaults are, how to change them, and how to customize existing policies with sub-policies lives in update-crypto-policies(8). Go read that if you're nosy like me 🧚

Now the actual deal: this specifies the signature algorithms used for public key authentication, as a comma-separated list of patterns. And the leading characters are a whole thing:

• If your list starts with a '+' character, the algorithms after it get appended to the built-in openssh default instead of replacing it.
• If it starts with a '-' character, the algorithms you specify (wildcards included!) get removed from the built-in openssh default set instead of replacing them.
• If it starts with a '^' character, your specified algorithms get placed at the head of the built-in openssh default set.

Wanna see the list of available signature algorithms? Just run "ssh -Q PubkeyAcceptedAlgorithms". Slay. 🔥

Heads up bestie: this option affects HostKeyAlgorithms too. Iykyk.

• PubkeyAuthentication — specifies whether to try public key authentication. The argument has to be yes (the default), no, unbound or host-bound. The last two enable public key authentication while respectively disabling or enabling the OpenSSH host-bound authentication protocol extension that restricted ssh-agent(1) forwarding needs. 👑
• RefuseConnection — literally lets a connection be refused by the configuration file itself. Iconic behavior honestly. If you specify this option, then ssh(1) terminates immediately before even attempting to connect to the remote host, displays an error message containing the argument to this keyword, and returns a non-zero exit status. So you can use this to leave reminders or warnings for the user via ssh_config. It's giving passive-aggressive sticky note 💀
• RekeyLimit — specifies the max amount of data that may be transmitted or received before the session key is renegotiated, optionally followed by a max amount of time that may pass before the session key gets renegotiated. First argument is in bytes and can have a suffix of 'K', 'M', or 'G' for Kilobytes, Megabytes, or Gigabytes respectively. Default is between '1G' and '4G', depending on the cipher. The optional second value is in seconds and can use any of the units documented in the TIME FORMATS section of sshd_config(5). Default value for RekeyLimit is default none, which means rekeying happens after the cipher's default amount of data has been sent or received, and no time based rekeying is done. 🙈
• RemoteCommand — specifies a command to execute on the remote machine after successfully connecting to the server. The command string extends to the end of the line, and it's executed with the user's shell. Arguments to RemoteCommand accept the tokens described in the “TOKENS” section. 🌸
• RemoteForward — specifies that a TCP port or Unix-domain socket on the remote machine gets forwarded over the secure channel. The remote port may either be forwarded to a specified host and port or Unix-domain socket from the local machine, or it can act as a SOCKS 4/5 proxy so a remote client can connect to arbitrary destinations from the local machine. First argument is the listening specification and may be [bind_address:]port or, if the remote host supports it, a Unix domain socket path. If you're forwarding to a specific destination then the second argument must be host:hostport or a Unix domain socket path — otherwise, if no destination argument is given, the remote forwarding gets established as a SOCKS proxy. When acting as a SOCKS proxy, the destination of the connection can be restricted by PermitRemoteOpen. 🦋

IPv6 addresses can be specified by enclosing addresses in square brackets.

If either argument contains a / in it, that argument is interpreted as a Unix-domain socket (on the corresponding host) rather than a TCP port.

Multiple forwardings may be specified, and additional forwardings can be given on the command line. Privileged ports can be forwarded only when logging in as root on the remote machine. Unix domain socket paths may use the tokens described in the “TOKENS” section and environment variables as described in the “ENVIRONMENT VARIABLES” section.

If the port argument is 0, the listen port gets dynamically allocated on the server and reported to the client at run time. ✨

If bind_address isn't specified, the default is to only bind to loopback addresses. If bind_address is '*' or an empty string, then the forwarding is requested to listen on all interfaces. Specifying a remote bind_address will only succeed if the server's GatewayPorts option is enabled (see sshd_config(5)).

• RequestTTY — specifies whether to request a pseudo-tty for the session. The argument may be one of:
  • no — never request a TTY
  • yes — always request a TTY when standard input is a TTY
  • force — always request a TTY
  • auto — request a TTY when opening a login session

This option mirrors the -t and -T flags for ssh(1). 🎀

• RequiredRSASize — specifies the minimum RSA key size (in bits) that ssh(1) will accept. User authentication keys smaller than this limit get ignored. Servers that present host keys smaller than this limit will cause the connection to be terminated. The default is 1024 bits. And ngl, note that this limit may only be raised from the default — you can't lower it. 🔥
• RevokedHostKeys — specifies revoked host public keys. Keys listed in this file get refused for host authentication. Big warning bestie: if this file doesn't exist or isn't readable, then host authentication will be refused for all hosts. Keys may be specified as a text file listing one public key per line, or as an OpenSSH Key Revocation List (KRL) as generated by ssh-keygen(1). For more info on KRLs, see the KEY REVOCATION LISTS section in ssh-keygen(1). Arguments to RevokedHostKeys may use the tilde syntax to refer to a user's home directory, the tokens described in the “TOKENS” section, and environment variables as described in the “ENVIRONMENT VARIABLES” section. 💀
• SecurityKeyProvider — specifies a path to a library that'll be used when loading any FIDO authenticator-hosted keys, overriding the default of using the built-in USB HID support. ✨

If the specified value begins with a '$' character, then it gets treated as an environment variable containing the path to the library.

• SendEnv — specifies what variables from the local environ(7) should be sent to the server. The server has to support it too, and the server must be configured to accept these environment variables. Note that the TERM environment variable is always sent whenever a pseudo-terminal is requested, because the protocol requires it. Refer to AcceptEnv in sshd_config(5) for how to configure the server. Variables are specified by name, which may contain wildcard characters. Multiple environment variables may be separated by whitespace or spread across multiple SendEnv directives. 🫶

See “PATTERNS” for more info on patterns.

Also you can clear previously set SendEnv variable names by prefixing patterns with -. The default is not to send any environment variables. 🙈

• ServerAliveCountMax — sets the number of server alive messages (see below 👇) that may be sent without ssh(1) receiving any messages back from the server. If this threshold is reached while server alive messages are being sent, ssh disconnects from the server, terminating the session. And it's important to note that server alive messages are very different from TCPKeepAlive (below). The server alive messages are sent through the encrypted channel and therefore will not be spoofable. The TCP keepalive option enabled by TCPKeepAlive is spoofable. The server alive mechanism is valuable when the client or server depend on knowing when a connection has become unresponsive. 💅

The default value is 3. So if, for example, ServerAliveInterval (see below) is set to 15 and ServerAliveCountMax is left at the default, then if the server becomes unresponsive, ssh will disconnect after approximately 45 seconds. Math ate. 🔥

• ServerAliveInterval — sets a timeout interval in seconds after which, if no data has been received from the server, ssh(1) will send a message through the encrypted channel to request a response from the server. The default is 0, which means these messages will not be sent to the server. ⭐
• SessionType — may be used to either request invocation of a subsystem on the remote system, or to prevent the execution of a remote command at all. That second one is useful for just forwarding ports, no cap. The argument to this keyword must be:
  • none — same as the -N option
  • subsystem — same as the -s option
  • default — shell or command execution
• SetEnv — like, just hand the server one or more environment variables and their contents directly, in the form “NAME=VALUE”. It works similarly to SendEnv, except that (with the exception of the TERM variable) the server has to be prepared to accept the environment variable, okay? 💅

The “VALUE” can use the tokens from the “TOKENS” section and environment variables as described in the “ENVIRONMENT VARIABLES” section. Use the good stuff.

• StdinNull — redirects stdin from /dev/null (which, real talk, prevents reading from stdin at all). You must use this — or the equivalent -n option — when ssh is running in the background. The argument has to be yes (same as the -n option) or no (the default). no cap 🫶
• StreamLocalBindMask — sets the octal file creation mode mask (umask) used when creating a Unix-domain socket file for local or remote port forwarding. Heads up: this option is only used for port forwarding to a Unix-domain socket file.

The default value is 0177, which makes a Unix-domain socket file that's readable and writable only by the owner. But bestie, note that not all operating systems honor the file mode on Unix-domain socket files 😭 so don't come for me if it doesn't behave.

• StreamLocalBindUnlink — controls whether to remove an existing Unix-domain socket file for local or remote port forwarding before creating a new one. If the socket file is already sitting there and StreamLocalBindUnlink isn't enabled, ssh literally cannot forward the port to that Unix-domain socket file. Again: only used for port forwarding to a Unix-domain socket file.

The argument must be yes or no (the default).

• StrictHostKeyChecking — the big one, the main character of host keys 🔥

If this flag is yes, ssh(1) will never automatically add host keys to the ~/.ssh/known_hosts file, and it flat-out refuses to connect to hosts whose host key has changed. Maximum protection against man-in-the-middle (MITM) attacks — iconic — though it can be super annoying when the /etc/ssh/ssh_known_hosts file is poorly maintained, or when you're constantly connecting to new hosts. Basically this option forces the user to manually add every single new host 🙈

If it's set to accept-new, then ssh will automatically add new host keys to the user's known_hosts file, but will not permit connections to hosts with changed host keys. If it's set to no or off, ssh will automatically add new host keys to the user known hosts files and allow connections to hosts with changed hostkeys to proceed, subject to some restrictions. And if it's set to ask (the default ✨), new host keys get added to the user known host files only after the user has confirmed that's what they actually want to do, and ssh will refuse to connect to hosts whose host key has changed. Either way, the host keys of known hosts get verified automatically in all cases. periodt.

• SyslogFacility — gives the facility code used when logging messages from ssh(1). The possible values are: DAEMON, USER, AUTH, LOCAL0, LOCAL1, LOCAL2, LOCAL3, LOCAL4, LOCAL5, LOCAL6, LOCAL7. The default is USER. pick your vibe 💜
• TCPKeepAlive — specifies whether the system should send TCP keepalive messages to the other side. If they're sent, death of the connection or a crash of one of the machines gets properly noticed 👀 However, that means connections will die if the route is down temporarily, and some people find that so annoying.

The default is yes (send TCP keepalive messages), and the client will notice if the network goes down or the remote host dies. This is important in scripts, and honestly many users want it too.

To disable TCP keepalive messages, set the value to no. Also peep ServerAliveInterval for protocol-level keepalives 🦋

• Tag — specify a configuration tag name that can later be used by a Match directive to select a block of configuration. Tag it and bag it 🎀
• Tunnel — requests tun(4) device forwarding between the client and the server. The argument must be yes, point-to-point (layer 3), ethernet (layer 2), or no (the default). Specifying yes requests the default tunnel mode, which is point-to-point. tunneling era 💫
• TunnelDevice — specifies the tun(4) devices to open on the client (local_tun) and the server (remote_tun).

The argument must be local_tun[:remote_tun]. The devices may be specified by numerical ID or the keyword any, which uses the next available tunnel device. If remote_tun is not specified, it defaults to any. The default is any:any. slay 🌈

• UpdateHostKeys — specifies whether ssh(1) should accept notifications of additional hostkeys from the server, sent after authentication has completed, and add them to UserKnownHostsFile. The argument must be yes, no or ask. This option lets you learn alternate hostkeys for a server and supports graceful key rotation by letting a server send replacement public keys before the old ones are removed. forward-thinking queen 👑

Additional hostkeys are only accepted if the key used to authenticate the host was already trusted or explicitly accepted by the user, the host was authenticated via UserKnownHostsFile (i.e. not GlobalKnownHostsFile) and the host was authenticated using a plain key and not a certificate. Lots of conditions, ikyk.

UpdateHostKeys is enabled by default if the user has not overridden the default UserKnownHostsFile setting and has not enabled VerifyHostKeyDNS — otherwise UpdateHostKeys gets set to no.

If UpdateHostKeys is set to ask, then the user is asked to confirm the modifications to the known_hosts file. Warning bestie: confirmation is currently incompatible with ControlPersist, and will be disabled if it is enabled 🫠

Presently, only sshd(8) from OpenSSH 6.8 and greater support the "hostkeys@openssh.com" protocol extension used to inform the client of all the server's hostkeys.

• User — specifies the user to log in as. This is useful when a different user name is used on different machines, and it saves you the trouble of having to remember to give the user name on the command line. Arguments to User may use the tokens described in the “TOKENS” section (with the exception of %r and %C) and environment variables as described in the “ENVIRONMENT VARIABLES” section. no more forgetting, we love to see it 💖
• UserKnownHostsFile — specifies one or more files to use for the user host key database, separated by whitespace. Each filename may use tilde notation to refer to the user's home directory, the tokens described in the “TOKENS” section, and environment variables as described in the “ENVIRONMENT VARIABLES” section. A value of none causes ssh(1) to ignore any user-specific known hosts files. The default is ~/.ssh/known_hosts, ~/.ssh/known_hosts2.
• VerifyHostKeyDNS — specifies whether to verify the remote key using DNS and SSHFP resource records. If it's set to yes, the client will implicitly trust keys that match a secure fingerprint from DNS. Insecure fingerprints get handled as if this option was set to ask. If it's set to ask, info on the fingerprint match gets displayed, but the user still needs to confirm new host keys according to the StrictHostKeyChecking option. The default is no.

See also “VERIFYING HOST KEYS” in ssh(1) 📖

• VersionAddendum — optionally specifies additional text to append to the SSH protocol banner sent by the client upon connection. The default is none. little signature on your banner, cute ✨
• VisualHostKey — if this flag is set to yes, an ASCII art representation of the remote host key fingerprint is printed in addition to the fingerprint string at login and for unknown host keys. If it's set to no (the default), no fingerprint strings are printed at login and only the fingerprint string will be printed for unknown host keys. ascii art moment 🎨
• WarnWeakCrypto — controls whether the user gets warned when the cryptographic algorithms negotiated for the connection are weak or otherwise recommended against. Warnings may be disabled by turning off a specific warning, or by disabling all warnings. Warnings about connections that don't use a post-quantum key exchange may be disabled using the no-pq-kex flag. no will disable all warnings. The default, equivalent to yes, is to enable all warnings. stay safe out there bestie 💀
• XAuthLocation — specifies the full pathname of the xauth(1) program. The default is /usr/bin/xauth.

---

🦋 PATTERNS 🦋

okay so a pattern consists of zero or more non-whitespace characters, ‘*’ (a wildcard that matches zero or more characters), or ‘?’ (a wildcard that matches exactly one character). Simple as.

For example, to specify a set of declarations for any host in the .co.uk set of domains, the following pattern could be used:

Host *.co.uk

The following pattern would match any host in the 192.168.0.[0-9] network range:

Host 192.168.0.?

A pattern-list is a comma-separated list of patterns. Patterns within pattern-lists may be negated by preceding them with an exclamation mark (‘!’). For example, to allow a key to be used from anywhere within an organization except from the "dialup" pool, the following entry (in authorized_keys) could be used:

from="!*.dialup.example.com,*.example.com"

⚠️ Note that a negated match will never produce a positive result by itself. fr fr. For example, attempting to match "host3" against the following pattern-list will fail:

from="!host1,!host2"

The solution here is to include a term that will yield a positive match, like a wildcard:

from="!host1,!host2,*"

see? always give it something to say yes to 💅

🎀 TOKENS 🎀

okay bestie so some of these keywords take arguments that aren't just plain text — they're tokens, and they get expanded at runtime ✨ like little magic spells that fill themselves in when ssh actually does the thing.

Here's the whole token squad, no cap:

• %% — a literal %. yes, it's the escape hatch for when you actually want a percent sign 💅
• %C — a hash of %l%h%p%r%j
• %d — the local user's home directory
• %f — the fingerprint of the server's host key
• %H — the known_hosts hostname or address that is being searched for
• %h — the remote hostname
• %I — a string describing the reason a KnownHostsCommand is being run. it's one of: ADDRESS when looking up a host by address (only when CheckHostIP is enabled), HOSTNAME when searching by hostname, or ORDER when preparing the host key algorithm preference list to use for the destination host 🧚
• %i — the local user ID
• %j — the contents of the ProxyJump option, or the empty string if that option is unset
• %K — the base64 encoded host key
• %k — the host key alias if one was specified, otherwise the original remote hostname given on the command line
• %L — the local hostname
• %l — the local hostname, including the domain name
• %n — the original remote hostname, as given on the command line
• %p — the remote port
• %r — the remote username
• %T — the local tun(4) or tap(4) network interface assigned if tunnel forwarding was requested, or "NONE" otherwise
• %t — the type of the server host key, e.g. ssh-ed25519
• %u — the local username

---

and now the important part: who's allowed to use what 👑

CertificateFile, ControlPath, IdentityAgent, IdentityFile, Include, KnownHostsCommand, LocalForward, Match exec, RemoteCommand, RemoteForward, RevokedHostKeys, UserKnownHostsFile and VersionAddendum all accept the tokens %%, %C, %d, %h, %i, %j, %k, %L, %l, %n, %p, %r, and %u.

KnownHostsCommand gets extra privileges 💖 — it additionally accepts the tokens %f, %H, %I, %K and %t.

Hostname is on a strict little diet: it accepts the tokens %% and %h.

LocalCommand gets all tokens. every single one. main character energy ✨

ProxyCommand and ProxyJump accept the tokens %%, %h, %n, %p, and %r.

⚠️ read this one, seriously — some of these directives build commands that get executed via the shell. and because ssh(1) performs no filtering or escaping of characters that have special meaning in shell commands (think: quotes), it is the user's responsibility to make sure the arguments passed to ssh(1) don't contain such characters, and that tokens are appropriately quoted when they're used 😭 like ssh is not your bodyguard here, bestie, you gotta do it yourself.

🌸 ENVIRONMENT VARIABLES 🌸

some keywords' arguments can also be expanded at runtime from environment variables on the client, and you do that by wrapping them in ${}. so for example ${HOME}/.ssh would refer to the user's .ssh directory 🏠✨

🚨 and if a specified environment variable doesn't exist, then an error will be returned and the setting for that keyword will be ignored. not "fallback to something sensible" — just... ignored. ok? ok 💀

the keywords that support environment variables are: CertificateFile, ControlPath, IdentityAgent, IdentityFile, Include, KnownHostsCommand, and UserKnownHostsFile.

and LocalForward and RemoteForward support environment variables only for Unix domain socket paths. that's it, that's the rule 🫶

🦋 FILES 🦋

~/.ssh/config

this is the per-user configuration file 💖 the format of this file is described way up above (hi, that was the whole rest of this page). this file is used by the SSH client.

and listen — because of the potential for abuse, this file has to have strict permissions: read/write for the user, and not writable by others. if you've been sloppy with your chmod, this is your sign to go fix it 🙈

/etc/ssh/ssh_config

the systemwide configuration file 👑 this one provides defaults for those values that aren't specified in the user's configuration file, and it also covers those users who don't have a configuration file at all.

this file must be world-readable. yes, the opposite vibe from the user one, don't ask, it's just how it is 🌈

⭐ SEE ALSO ⭐

go read these next, i promise they're worth it:

• ssh(1)
• crypto-policies(7)
• update-crypto-policies(8)

👑 AUTHORS 👑

OpenSSH is a derivative of the original and free ssh 1.2.12 release by Tatu Ylonen. Aaron Campbell, Bob Beck, Markus Friedl, Niels Provos, Theo de Raadt and Dug Song removed many bugs, re-added newer features and created OpenSSH. Markus Friedl contributed the support for SSH protocol versions 1.5 and 2.0. absolute legends, every one of them 🫶✨

---

and that's the whole page, bestie!! you made it to the end, i'm so proud 😭💖 this one shipped from Debian, dated October 4, 2025, and it's the SSH_CONFIG(5) manual page — go forth and configure responsibly 🎀🦋💅