This is a minor but long-standing problem on the courier build (under my
version of freebsd, at least) that I am just now remembering to mention:

In courier/configure, the following test appears ...

  if ulimit -v 2>/dev/null
  then
          ULIMIT='ulimit -v'
  else
          ULIMIT='ulimit -d'
  fi

However, this doesn't test the actual usage of the "ulimit" command
within the courier system.  In all places where "ulimit" appears with
the "-v" or "-d" option, it always takes a subsequent argument.  For
example:

  ulimit -v $IMAP_ULIMITD
  ulimit -d "$ULIMIT"
  ... etc. ...

However, under some values of CONFIG_SHELL (mine ends up being
/bin/bash, and my bash version is 2.05b), the following fails as
follows:

  % ulimit -v 500000
  bash: ulimit: virtual memory: cannot modify limit: Invalid argument

... even though the following works fine:

  % ulimit -v
  589824

... as does this:

  % ulimit -d
  589824
  % ulimit -d 500000
  % ulimit -d
  500000

Therefore, if the above test indicates that "ulimit -v" is correct for a
given system, this doesn't say anything about whether or not
"ulimit -v ARGUMENT" will work on that same system.

Because of this, I would like to propose that the above test in
courier/configure be modified as follows (or at least similarly):

  if $CONFIG_SHELL -c "ulimit -v 0"
  then
          ULIMIT='ulimit -v'
  else
          ULIMIT='ulimit -d'
  fi

The "$CONFIG_SHELL -c" thing is needed because we don't want this test
to change the ulimit of the shell that's running courier/configure.

Thanks.


-- 
 Lloyd Zusman
 [EMAIL PROTECTED]



-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
courier-users mailing list
[EMAIL PROTECTED]
Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users

Reply via email to