From gmoellen@cv3.cv.nrao.edu Wed Nov 14 17:53:31 2001
Date: Wed, 14 Nov 2001 17:02:52 -0700 (MST)
From: George Moellenbrock <gmoellen@cv3.cv.nrao.edu>
To: aips2-naug@zia.aoc.NRAO.EDU
Subject: filling data at the AOC


Hi everyone,

Today, "new-aips++-user" Claire and I became acutely aware of some current
problems filling vla data via remote tape drives, problems which I hadn't
fully appreciated until now.  Wes and I have partially remedied the
situation, to a point where it should be possible to easily fill (via
remote tapes) your VLA data at the AOC.

Here's the situation, as best I can discern:

There have been three independent problems, depending upon what machines
you might attempt a remote vlafiller to/from.  Briefly, these are (1) lack
of rsh access to some of the new linux boxes, (2) Solaris version
incompatibilities between the remote tape drive machine and the aips++
build machine, and (3) solaris standard library version problems. You will
have vlafiller problems if the machine on which the remove tape drive
resides suffers any of these.  Together, these problems have made it
rather difficult to fill VLA data into aips++ at the AOC recently.

If the machine you are sitting at suffers from either #2 or #3, you will
likely have many other problems running aips++.

Now, Wes has discussed with James and George (Martin) the solutions to
problems #1 and #2, which will hopefully be implemented soon.  The only
public machine currently affected by #2 is cochiti (Claire's unfortunate
choice this morning).  Wes has investigated #3, and he has repaired the
current stable, weekly, and daily versions of vlafiller.  By next Monday,
these fixes will have propogated to all of aips++ (not just vlafiller, to
the extent that other tools are affected) in all versions.

So, the current remote vlafiller status is:

USE OF ANY PUBLIC *SOLARIS* MACHINE (EXCEPT COCHITI) SHOULD WORK TO
REMOTELY FILL YOUR VLA DATA.

Thanks Claire, for your patience, and thanks Wes, for your help.

Cheers,
George