rsh connects to the specified hostname and executes the specified command. rsh copies its standard input to the remote command, the standard output of the remote command to its standard output, and the standard error of the remote command to its standard error. Interrupt, quit, and terminate signals are propagated to the remote command. rsh normally terminates when the remote command does.

If you omit command, instead of executing a single command, rsh logs you in on the remote host using rlogin(1).

rsh will not return the exit status code of command.

Shell metacharacters which are not quoted are interpreted on the local machine, while quoted metacharacters are interpreted on the remote machine. See EXAMPLES.

If there is no locale setting in the initialization file of the login shell (.cshrc, . . .) for a particular user, rsh always executes the command in the "C" locale instead of using the default locale of the remote machine.

The following options are supported:

-l username

Uses username as the remote username instead of your local username. In the absence of this option, the remote username is the same as your local username.

-n

Redirects the input of rsh to /dev/null. You sometimes need this option to avoid unfortunate interactions between rsh and the shell which invokes it. For example, if you are running rsh and invoke a rsh in the background without redirecting its input away from the terminal, it will block even if no reads are posted by the remote command. The -n option will prevent this.

The type of remote shell (sh, rsh, or other) is determined by the user's entry in the file /etc/passwd on the remote system.

The following operand is supported:

command

The command to be executed on the specified hostname.

See largefile(5) for the description of the behavior of rsh and remsh when encountering files greater than or equal to 2 Gbyte ( 2³¹ bytes).

The rsh and remsh commands are IPv6-enabled. See ip6(7P).

Hostnames are given in the hosts database, which may be contained in the /etc/hosts file, the Internet domain name database, or both. Each host has one official name (the first name in the database entry) and optionally one or more nicknames. Official hostnames or nicknames may be given as hostname.

If the name of the file from which rsh is executed is anything other than rsh, rsh takes this name as its hostname argument. This allows you to create a symbolic link to rsh in the name of a host which, when executed, will invoke a remote shell on that host. By creating a directory and populating it with symbolic links in the names of commonly used hosts, then including the directory in your shell's search path, you can run rsh by typing hostname to your shell.

If rsh is invoked with the basename remsh, rsh will check for the existence of the file /usr/bin/remsh. If this file exists, rsh will behave as if remsh is an alias for rsh. If /usr/bin/remsh does not exist, rsh will behave as if remsh is a host name.

Each remote machine may have a file named /etc/hosts.equiv containing a list of trusted hostnames with which it shares usernames. Users with the same username on both the local and remote machine may run rsh from the machines listed in the remote machine's /etc/hosts file. Individual users may set up a similar private equivalence list with the file .rhosts in their home directories. Each line in this file contains two names: a hostname and a username separated by a space. The entry permits the user named username who is logged into hostname to use rsh to access the remote machine as the remote user. If the name of the local host is not found in the /etc/hosts.equiv file on the remote machine, and the local username and hostname are not found in the remote user's .rhosts file, then the access is denied. The hostnames listed in the /etc/hosts.equiv and .rhosts files must be the official hostnames listed in the hosts database; nicknames may not be used in either of these files.

rsh will not prompt for a password if access is denied on the remote machine unless the command argument is omitted.

The following exit values are returned:

0

Successful completion.

1

An error occurred.

/etc/hosts

Internet host table

/etc/hosts.equiv

trusted remote hosts and users

/etc/passwd

system password file

See attributes(5) for descriptions of the following attributes:

on(1), rlogin(1), telnet(1), vi(1), in.named(1M), hosts(4), hosts.equiv(4), ipnodes(4), attributes(5), largefile(5), ip6(7P)

When a system is listed in hosts.equiv, its security must be as good as local security. One insecure system listed in hosts.equiv can compromise the security of the entire system.

You cannot run an interactive command (such as vi(1)). Use rlogin if you wish to do this.

Stop signals stop the local rsh process only. This is arguably wrong, but currently hard to fix for reasons too complicated to explain here.

The current local environment is not passed to the remote shell.

Sometimes the -n option is needed for reasons that are less than obvious. For example, the command:

example% rsh somehost dd if=/dev/nrmt0 bs=20b | tar xvpBf -

will put your shell into a strange state. Evidently, what happens is that the tar terminates before the rsh. The rsh then tries to write into the ``broken pipe'' and, instead of terminating neatly, proceeds to compete with your shell for its standard input. Invoking rsh with the -n option avoids such incidents.

This bug occurs only when rsh is at the beginning of a pipeline and is not reading standard input. Do not use the -n if rsh actually needs to read standard input. For example,

example% tar cf - . | rsh sundial dd of=/dev/rmt0 obs=20b

does not produce the bug. If you were to use the -n in a case like this, rsh would incorrectly read from /dev/null instead of from the pipe.