ssh

jxwpx

SYNOPSIS
     ssh [-l login_name] hostname | user@hostname [command]
     ssh [-afgknqstvxACNTX1246] [-b bind_address] [-c cipher_spec]
     [-e escape_char] [-i identity_file] [-l login_name] [-m mac_spec]
     [-o option] [-p port] [-F configfile] [-L port:host:hostport] [-R
     port:host:hostport] [-D port] hostname | user@hostname [command]
DESCRIPTION
     ssh (SSH client) is a program for logging into a remote machine and for
     executing commands on a remote machine.  It is intended to replace rlogin
     and rsh, and provide secure encrypted communications between two
     untrusted hosts over an insecure network.  X11 connections and arbitrary
     TCP/IP ports can also be forwarded over the secure channel.
     ssh connects and logs into the specified hostname. The user must prove
     his/her identity to the remote machine using one of several methods
     depending on the protocol version used:
   SSH protocol version 1
     First, if the machine the user logs in from is listed in /etc/hosts.equiv
     or /etc/ssh/shosts.equiv on the remote machine, and the user names are
     the same on both sides, the user is immediately permitted to log in.
     Second, if .rhosts or .shosts exists in the user’s home directory on the
     remote machine and contains a line containing the name of the client
     machine and the name of the user on that machine, the user is permitted
     to log in. This form of authentication alone is normally not allowed by
     the server because it is not secure.
     The second authentication method is the rhosts or hosts.equiv method com-
     bined with RSA-based host authentication.  It means that if the login
     would be permitted by $HOME/.rhosts, $HOME/.shosts, /etc/hosts.equiv, or
     /etc/ssh/shosts.equiv, and if additionally the server can verify the
     client’s host key (see /etc/ssh/ssh_known_hosts and
     $HOME/.ssh/known_hosts in the FILES section), only then login is permit-
     ted.  This authentication method closes security holes due to IP spoof-
     ing, DNS spoofing and routing spoofing.  [Note to the administrator:
     /etc/hosts.equiv, $HOME/.rhosts, and the rlogin/rsh protocol in general,
     are inherently insecure and should be disabled if security is desired.]
     As a third authentication method, ssh supports RSA based authentication.
     The scheme is based on public-key cryptography: there are cryptosystems
     where encryption and decryption are done using separate keys, and it is
     not possible to derive the decryption key from the encryption key. RSA
     is one such system.  The idea is that each user creates a public/private
     key pair for authentication purposes.  The server knows the public key,
     and only the user knows the private key.  The file
     $HOME/.ssh/authorized_keys lists the public keys that are permitted for
     logging in.  When the user logs in, the ssh program tells the server
     which key pair it would like to use for authentication.  The server
     checks if this key is permitted, and if so, sends the user (actually the
     ssh program running on behalf of the user) a challenge, a random number,
     encrypted by the user’s public key.  The challenge can only be decrypted
     using the proper private key.  The user’s client then decrypts the chal-
     lenge using the private key, proving that he/she knows the private key
     but without disclosing it to the server.
     ssh implements the RSA authentication protocol automatically.  The user
     creates his/her RSA key pair by running ssh-keygen(1).  This stores the
     private key in $HOME/.ssh/identity and the public key in
     $HOME/.ssh/identity.pub in the user’s home directory.  The user should
     then copy the identity.pub to $HOME/.ssh/authorized_keys in his/her home
     directory on the remote machine (the authorized_keys file corresponds to
     the conventional $HOME/.rhosts file, and has one key per line, though the
     lines can be very long).  After this, the user can log in without giving
     the password.  RSA authentication is much more secure than rhosts authen-
     tication.
     The most convenient way to use RSA authentication may be with an authen-
     tication agent.  See ssh-agent(1) for more information.
     If other authentication methods fail, ssh prompts the user for a pass-
     word.  The password is sent to the remote host for checking; however,
     since all communications are encrypted, the password cannot be seen by
     someone listening on the network.
   SSH protocol version 2
     When a user connects using protocol version 2 similar authentication
     methods are available.  Using the default values for
     PreferredAuthentications, the client will try to authenticate first using
     the hostbased method; if this method fails public key authentication is
     attempted, and finally if this method fails keyboard-interactive and
     password authentication are tried.
     The public key method is similar to RSA authentication described in the
     previous section and allows the RSA or DSA algorithm to be used: The
     client uses his private key, $HOME/.ssh/id_dsa or $HOME/.ssh/id_rsa, to
     sign the session identifier and sends the result to the server.  The
     server checks whether the matching public key is listed in
     $HOME/.ssh/authorized_keys and grants access if both the key is found and
     the signature is correct.  The session identifier is derived from a
     shared Diffie-Hellman value and is only known to the client and the
     server.
     If public key authentication fails or is not available a password can be
     sent encrypted to the remote host for proving the user’s identity.
     Additionally, ssh supports hostbased or challenge response authentica-
     tion.
     Protocol 2 provides additional mechanisms for confidentiality (the traf-
     fic is encrypted using 3DES, Blowfish, CAST128 or Arcfour) and integrity
     (hmac-md5, hmac-sha1).  Note that protocol 1 lacks a strong mechanism for
     ensuring the integrity of the connection.
   Login session and remote execution
     When the user’s identity has been accepted by the server, the server
     either executes the given command, or logs into the machine and gives the
     user a normal shell on the remote machine. All communication with the
     remote command or shell will be automatically encrypted.
     If a pseudo-terminal has been allocated (normal login session), the user
     may use the escape characters noted below.
     If no pseudo tty has been allocated, the session is transparent and can
     be used to reliably transfer binary data.  On most systems, setting the
     escape character to “none” will also make the session transparent even if
     a tty is used.
     The session terminates when the command or shell on the remote machine
     exits and all X11 and TCP/IP connections have been closed. The exit sta-
     tus of the remote program is returned as the exit status of ssh.
   Escape Characters
     When a pseudo terminal has been requested, ssh supports a number of func-
     tions through the use of an escape character.
     A single tilde character can be sent as ~~ or by following the tilde by a
     character other than those described below.  The escape character must
     always follow a newline to be interpreted as special.  The escape charac-
     ter can be changed in configuration files using the EscapeChar configura-
     tion directive or on the command line by the -e option.
     The supported escapes (assuming the default ‘~’) are:
     ~.      Disconnect
     ~^Z     Background ssh
     ~#      List forwarded connections
     ~&      Background ssh at logout when waiting for forwarded connection /
         X11 sessions to terminate
     ~?      Display a list of escape characters
     ~C      Open command line (only useful for adding port forwardings using
         the -L and -R options)
     ~R      Request rekeying of the connection (only useful for SSH protocol
         version 2 and if the peer supports it)
   X11 and TCP forwarding
     If the ForwardX11 variable is set to “yes” (or, see the description of
     the -X and -x options described later) and the user is using X11 (the
     DISPLAY environment variable is set), the connection to the X11 display
     is automatically forwarded to the remote side in such a way that any X11
     programs started from the shell (or command) will go through the
     encrypted channel, and the connection to the real X server will be made
     from the local machine.  The user should not manually set DISPLAY. For-
     warding of X11 connections can be configured on the command line or in
     configuration files.
     The DISPLAY value set by ssh will point to the server machine, but with a
     display number greater than zero.  This is normal, and happens because
     ssh creates a “proxy” X server on the server machine for forwarding the
     connections over the encrypted channel.
     ssh will also automatically set up Xauthority data on the server machine.
     For this purpose, it will generate a random authorization cookie, store
     it in Xauthority on the server, and verify that any forwarded connections
     carry this cookie and replace it by the real cookie when the connection
     is opened. The real authentication cookie is never sent to the server
     machine (and no cookies are sent in the plain).
     If the ForwardAgent variable is set to “yes” (or, see the description of
     the -A and -a options described later) and the user is using an authenti-
     cation agent, the connection to the agent is automatically forwarded to
     the remote side.
     Forwarding of arbitrary TCP/IP connections over the secure channel can be
     specified either on the command line or in a configuration file.  One
     possible application of TCP/IP forwarding is a secure connection to an
     electronic purse; another is going through firewalls.
   Server authentication
     ssh automatically maintains and checks a database containing identifica-
     tions for all hosts it has ever been used with.  Host keys are stored in
     $HOME/.ssh/known_hosts in the user’s home directory.  Additionally, the
     file /etc/ssh/ssh_known_hosts is automatically checked for known hosts.
     Any new hosts are automatically added to the user’s file. If a host’s
     identification ever changes, ssh warns about this and disables password
     authentication to prevent a trojan horse from getting the user’s pass-
     word.  Another purpose of this mechanism is to prevent man-in-the-middle
     attacks which could otherwise be used to circumvent the encryption.  The
     StrictHostKeyChecking option can be used to prevent logins to machines
     whose host key is not known or has changed.
     The options are as follows:
     -a      Disables forwarding of the authentication agent connection.
     -A      Enables forwarding of the authentication agent connection. This
         can also be specified on a per-host basis in a configuration
         file.
         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.
     -b bind_address
         Specify the interface to transmit from on machines with multiple
         interfaces or aliased addresses.
     -c blowfish|3des|des
         Selects the cipher to use for encrypting the session.  3des is
         used by default.  It is believed to be secure.  3des (triple-des)
         is an encrypt-decrypt-encrypt triple with three different keys.
         blowfish is a fast block cipher, it appears very secure and is
         much faster than 3des.  des is only supported in the ssh client
         for interoperability with legacy protocol 1 implementations that
         do not support the 3des cipher.  Its use is strongly discouraged
         due to cryptographic weaknesses.
     -c cipher_spec
         Additionally, for protocol version 2 a comma-separated list of
         ciphers can be specified in order of preference.  See Ciphers for
         more information.
     -e ch|^ch|none
         Sets the escape character for sessions with a pty (default: ‘~’).
         The escape character is only recognized at the beginning of a
         line.  The escape character followed by a dot (‘.’) closes the
         connection, followed by control-Z suspends the connection, and
         followed by itself sends the escape character once.  Setting the
         character to “none” disables any escapes and makes the session
         fully transparent.
     -f     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 -n.  The recommended way to start X11 programs at a
         remote site is with something like ssh -f host xterm.
     -g      Allows remote hosts to connect to local forwarded ports.
     -i identity_file
         Selects a file from which the identity (private key) for RSA or
         DSA authentication is read.  The default is $HOME/.ssh/identity
         for protocol version 1, and $HOME/.ssh/id_rsa and
         $HOME/.ssh/id_dsa for protocol version 2.  Identity files may
         also be specified on a per-host basis in the configuration file.
         It is possible to have multiple -i options (and multiple identi-
         ties specified in configuration files).
     -I smartcard_device
         Specifies which smartcard device to use. The argument is the
         device ssh should use to communicate with a smartcard used for
         storing the user’s private RSA key.
     -k      Disables forwarding of Kerberos tickets and AFS tokens.  This may
         also be specified on a per-host basis in the configuration file.
     -l login_name
         Specifies the user to log in as on the remote machine.  This also
         may be specified on a per-host basis in the configuration file.
     -m mac_spec
         Additionally, for protocol version 2 a comma-separated list of
         MAC (message authentication code) algorithms can be specified in
         order of preference.  See the MACs keyword for more information.
     -n      Redirects stdin from /dev/null (actually, prevents reading from
         stdin).  This must be used when ssh is run in the background.  A
         common trick is to use this to run X11 programs on a remote
         machine.  For example, ssh -n shadows.cs.hut.fi emacs & will
         start an emacs on shadows.cs.hut.fi, and the X11 connection will
         be automatically forwarded over an encrypted channel.  The ssh
         program will be put in the background.  (This does not work if
         ssh needs to ask for a password or passphrase; see also the -f
         option.)
     -N      Do not execute a remote command.  This is useful for just for-
         warding ports (protocol version 2 only).
     -o option
         Can be used to give options in the format used in the configura-
         tion file. This is useful for specifying options for which there
         is no separate command-line flag.
     -p port
         Port to connect to on the remote host.  This can be specified on
         a per-host basis in the configuration file.
     -q      Quiet mode.  Causes all warning and diagnostic messages to be
         suppressed.
     -s      May be used to request invocation of a subsystem on the remote
         system. Subsystems are a feature of the SSH2 protocol which
         facilitate the use of SSH as a secure transport for other appli-
         cations (eg. sftp). The subsystem is specified as the remote com-
         mand.
     -t      Force pseudo-tty allocation.  This can be used to execute arbi-
         trary screen-based programs on a remote machine, which can be
         very useful, e.g., when implementing menu services.  Multiple -t
         options force tty allocation, even if ssh has no local tty.
     -T      Disable pseudo-tty allocation.
     -v      Verbose mode.  Causes ssh to print debugging messages about its
         progress.  This is helpful in debugging connection, authentica-
         tion, and configuration problems.  Multiple -v options increases
         the verbosity.  Maximum is 3.
     -x      Disables X11 forwarding.
     -X      Enables X11 forwarding.  This can also be specified on a per-host
         basis in a configuration file.
         X11 forwarding should be enabled with caution.  Users with the
         ability to bypass file permissions on the remote host (for the
         user’s X 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.
     -C      Requests compression of all data (including stdin, stdout,
         stderr, and data for forwarded X11 and TCP/IP connections).  The
         compression algorithm is the same used by gzip(1), and the
         “level” can be controlled by the CompressionLevel option for pro-
         tocol version 1.  Compression is desirable on modem lines and
         other slow connections, but will only slow down things on fast
         networks.  The default value can be set on a host-by-host basis
         in the configuration files; see the Compression option.
     -F configfile
         Specifies an alternative per-user configuration file.  If a con-
         figuration file is given on the command line, the system-wide
         configuration file (/etc/ssh/ssh_config) will be ignored.  The
         default for the per-user configuration file is $HOME/.ssh/config.
     -L port:host:hostport
         Specifies that the given port on the local (client) host is to be
         forwarded to the given host and port on the remote side.  This
         works by allocating a socket to listen to port on the local side,
         and whenever a connection is made to this port, the connection is
         forwarded over the secure channel, and a connection is made to
         host port hostport from the remote machine.  Port forwardings can
         also be specified in the configuration file.  Only root can for-
         ward privileged ports.  IPv6 addresses can be specified with an
         alternative syntax: port/host/hostport
     -R port:host:hostport
         Specifies that the given port on the remote (server) host is to
         be forwarded to the given host and port on the local side. This
         works by allocating a socket to listen to port on the remote
         side, and whenever a connection is made to this port, the connec-
         tion is forwarded over the secure channel, and a connection is
         made to host port hostport from the local machine. Port forward-
         ings can also be specified in the configuration file.  Privileged
         ports can be forwarded only when logging in as root on the remote
         machine.  IPv6 addresses can be specified with an alternative
         syntax: port/host/hostport
     -D port
         Specifies a local “dynamic” application-level port forwarding.
         This works by allocating a socket to listen to port on the local
         side, and whenever a connection is made to this port, the connec-
         tion is forwarded over the secure channel, and the application
         protocol is then used to determine where to connect to from the
         remote machine.  Currently the SOCKS4 protocol is supported, and
         ssh will act as a SOCKS4 server.  Only root can forward privi-
         leged ports.  Dynamic port forwardings can also be specified in
         the configuration file.
     -1      Forces ssh to try protocol version 1 only.
     -2      Forces ssh to try protocol version 2 only.
     -4      Forces ssh to use IPv4 addresses only.
     -6      Forces ssh to use IPv6 addresses only.
CONFIGURATION FILES
     ssh may additionally obtain configuration data from a per-user configura-
     tion file and a system-wide configuration file.  The file format and con-
     figuration options are described in ssh_config(5).
ENVIRONMENT
     ssh will normally set the following environment variables:
     DISPLAY
         The DISPLAY variable indicates the location of the X11 server.
         It is automatically set by ssh to point to a value of the form
         “hostname:n” where hostname indicates the host where the shell
         runs, and n is an integer >= 1.  ssh uses this special value to
         forward X11 connections over the secure channel.  The user should
         normally not set DISPLAY explicitly, as that will render the X11
         connection insecure (and will require the user to manually copy
         any required authorization cookies).
     HOME    Set to the path of the user’s home directory.
     LOGNAME
         Synonym for USER; set for compatibility with systems that use
         this variable.
     MAIL    Set to the path of the user’s mailbox.
     PATH    Set to the default PATH, as specified when compiling ssh.
     SSH_ASKPASS
         If ssh needs a passphrase, it will read the passphrase from the
         current terminal if it was run from a terminal.  If ssh does not
         have a terminal associated with it but DISPLAY and SSH_ASKPASS
         are set, it will execute the program specified by SSH_ASKPASS and
         open an X11 window to read the passphrase. This is particularly
         useful when calling ssh from a .Xsession or related script.
         (Note that on some machines it may be necessary to redirect the
         input from /dev/null to make this work.)
     SSH_AUTH_SOCK
         Identifies the path of a unix-domain socket used to communicate
         with the agent.
     SSH_CONNECTION
         Identifies the client and server ends of the connection.  The
         variable contains four space-separated values: client ip-address,
         client port number, server ip-address and server port number.
     SSH_ORIGINAL_COMMAND
         The variable contains the original command line if a forced com-
         mand is executed.  It can be used to extract the original argu-
         ments.
     SSH_TTY
         This is set to the name of the tty (path to the device) associ-
         ated with the current shell or command.  If the current session
         has no tty, this variable is not set.
     TZ      The timezone variable is set to indicate the present timezone if
         it was set when the daemon was started (i.e., the daemon passes
         the value on to new connections).
     USER    Set to the name of the user logging in.
     Additionally, ssh reads $HOME/.ssh/environment, and adds lines of the
     format “VARNAME=value” to the environment if the file exists and if users
     are allowed to change their environment.  See the PermitUserEnvironment
     option in sshd_config(5).
FILES
     $HOME/.ssh/known_hosts
         Records host keys for all hosts the user has logged into that are
         not in /etc/ssh/ssh_known_hosts.  See sshd(8).
     $HOME/.ssh/identity, $HOME/.ssh/id_dsa, $HOME/.ssh/id_rsa
         Contains the authentication identity of the user.  They are for
         protocol 1 RSA, protocol 2 DSA, and protocol 2 RSA, respectively.
         These files contain sensitive data and should be readable by the
         user but not accessible by others (read/write/execute).  Note
         that ssh ignores a private key file if it is accessible by oth-
         ers.  It is possible to specify a passphrase when generating the
         key; the passphrase will be used to encrypt the sensitive part of
         this file using 3DES.
     $HOME/.ssh/identity.pub, $HOME/.ssh/id_dsa.pub, $HOME/.ssh/id_rsa.pub
<span style

本文来自ChinaUnix博客，如果查看原文请点：http://blog.chinaunix.net/u/5403/showart_17550.html