Administration Guide

Controlling the rah Command

You can use the following environment variables to control the rah command.

Table 152.

Name Meaning Default
$RAHBUFDIR
Note: Available on UNIX-based platforms only.
directory for buffer /tmp/$USER
$RAHBUFNAME
Note: Available on UNIX-based platforms only.
filename for buffer rahout
$RAHOSTFILE (on UNIX-based platforms); RAHOSTFILE (on Windows NT) file containing list of hosts db2nodes.cfg
$RAHOSTLIST (on UNIX-based platforms); RAHOSTLIST (on Windows NT) list of hosts as a string extracted from $RAHOSTFILE
$RAHCHECKBUF
Note: Available on UNIX-based platforms only.
if set to "no", bypass checks not set
$RAHSLEEPTIME (on UNIX-based platforms); RAHSLEEPTIME (on Windows NT) time in seconds this script will wait for initial output from commands run in parallel 86400 seconds for db2_kill, 200 seconds for all other
$RAHWAITTIME (on UNIX-based platforms); RAHWAITTIME (on Windows NT) on Windows NT, interval in seconds between successive checks that remote jobs are still running.
On UNIX-based platforms, interval in seconds between successive checks that remote jobs are still running and rah: waiting for <pid> ... messages.
On all platforms, specify any positive integer. Prefix value with a leading zero to suppress messages, for example, export RAHWAITTIME=045.
It is not necessary to specify a low value as rah does not rely on these checks to detect job completion.
45 seconds
$RAHENV
Note: Available on UNIX-based platforms only.
specifies filename to be executed if $RAHDOTFILES=E or K or PE or B $ENV
$RAHUSER (on UNIX-based platforms); RAHUSER (on Windows NT) on UNIX-based platforms, user ID under which the remote command is to be run.
On Windows NT, the logon account associated with the DB2 Remote Command Service
$USER

Note: On UNIX-based platforms, the value of $RAHENV where rah is run is used, not the value (if any) set by the remote shell.

$RAHDOTFILES on UNIX-Based Platforms

Note: The information in this section applies to UNIX-based platforms only.
Following are the .files that are run if no prefix sequence is specified:

P: .profile
E: File named in $RAHENV (probably .kshrc)
K: Same as E
PE: .profile followed by file named in $RAHENV (probably .kshrc)
B: Same as PE
N: None (or Neither)

Note:

If your login shell is not a Korn shell, any dot files you specify to be executed will be executed in a Korn shell process, and so must conform to Korn shell syntax. So, for example, if your login shell is a C shell, to have your .cshrc environment set up for commands executed by rah, you should either create a Korn shell INSTHOME/.profile equivalent to your .cshrc and specify in your INSTHOME/.cshrc:

  setenv RAHDOTFILES P

or you should create a Korn shell INSTHOME/.kshrc equivalent to your .cshrc and specify in your INSTHOME/.cshrc:

  setenv RAHDOTFILES E
  setenv RAHENV INSTHOME/.kshrc

Also, it is essential that your .cshrc does not write to stdout if there is no tty (as when invoked by rsh). You can ensure this by enclosing any lines which write to stdout by, for example,

  if { tty -s } then echo "executed .cshrc";
  endif

Setting the Default Environment Profile on Windows NT

Note: The information in this section applies to Windows NT only.
To set the default environment profile for the rah command, use a file called db2rah.env, which should be created in the instance directory. The file should have the following format:

   ; This is a comment line
   DB2INSTANCE=instancename
   DB2DBDFT=database
   ; End of file

You can specify all the environment variables that you need to initialize the environment for rah.

[ Top of Page | Previous Page | Next Page | Table of Contents | Index ]

[ DB2 List of Books | Search the DB2 Books ]