From: Jeff Squyres; Thursday, May 13, 2004 6:10 AM
> 
> However, two problems still remain -- but I wanted to get these speed
> fixes in so that NCSA GUI people could work much faster.  Here's
what's
> still broken:
> 
> 1. There's something wrong when LAM's RPMs get installed in the server
> non-core section.  I don't know yet whether this is a problem with the
LAM
> RPMs, switcher, or RPM doing some funky ordering that I didn't
anticipate
> (I'm leaning towards the last one -- stupid RPM!).   I'll keep poking
> around here.
> 
> 2. There's still some oddness in DM/update-rpms interaction if you try
to
> run install_cluster again such that DM reports a failure from
update-rpms
> in the case of "everything has already been installed -- nothing to
do."
> I think this is another case of "Mat expected one thing; Dave expected
> another."  The solution is not quite trivial, however; Dave and I will
> work this out and get it solved soon.

I just fixed the update-rpms end of this, i.e., update-rpms now returns
the correct exit code for this case.

BTW, you want watch what update-rpms does, including the error code
being returned, by setting DEBUG_UPDATE_RPMS to some value before
launching install_cluster, e.g.,

  $ export DEBUG_UPDATE_RPMS=y
  $ ./install_cluster device

This will write out the actual command line, what's being returned, and
the exit code to /dev/tty.

-- 
David N. Lombard
 
My comments represent my opinions, not those of Intel Corporation


-------------------------------------------------------
This SF.Net email is sponsored by: SourceForge.net Broadband
Sign-up now for SourceForge Broadband and get the fastest
6.0/768 connection for only $19.95/mo for the first 3 months!
http://ads.osdn.com/?ad_id%62&alloc_ida84&op=click
_______________________________________________
Oscar-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/oscar-devel

Reply via email to