Hello all,
there was a mail-sending-problem[1] on the linux-boxes. The problem is fixed
now and the mails are on their ways. Just to let you know if you wonder where
all these old mails are coming from (I doubt the issue was serious, otherwise
someone had notice me before about the missing mails).
Sincerely,
DaB.
[1] https://jira.toolserver.org/browse/MNT-1256
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885
Hello all,
yesterday Merlissimo and I successfully tested the installation of the SGE-
version for the toolserver. The last step is now to install the new version on
the live-system. For that, the SGE-service needs to stop completely on the
cluster, the old version has to be removed and the new one has to be
installed. We plan to to this on
Thursday 5. July between 17:30 and 22:30 UTC.
During this time no SGE will work. There will be no restarting (and no
migration) of stopped things after the update.
After the update is done, we will start to use the 2 Linux-boxes for tools too
(I will send details than).
Sincerely,
DaB.
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885
Hello all,
it is finally done, the linux-login-servers (yarrow and nightshade) are ready
to use :-). You should be able to login from the outside to these boxes and to
hop from any other toolserver-host too.
I installed most of the packages at [1]; if any package is missing, please use
JIRA now to request it. We are not able at the moment to package software for
which there is no debian-package, because we miss a build-host (I willl speak
with the WMDE about that).
We run Debian stable on the linux-boxes, and both servers (like willow too)
are managed by puppet to keep them in sync in manner of configuration and
software. I installed some aliases (mostly gsometing->something) to make to
switch from Solaris (back) to linux easier for you – if I miss a important
one, please talk to me in IRC or open a jira-ticket; for less important ones
you can define them yourself in .bashrc or .profile in your home.
While you can (at least at the moment) use the cron-system on linux, you
should not – use the submit-hosts and SGE! Speaking of SGE: To let your sge-
jobs run on linux only, you have to add "-l arch=lx" as parameter. By adding
"-l arch='*' " you let SGE decide the best server of both operating systems
for your job. This will always result in a minimum waiting time and is the
recommended usage if possible. Solaris will be the default for SGE until 30.
September 2012, after that the default will be switched to any available os (I
will send a reminder or 2 somewhen before).
If there are any problems, please open a JIRA-report and the roots will try to
resolve them. What will not happen: Installing of packages from testing or
unstable, or installing of non-packaged-software (see above).
The next step in my plan, is to convert our webservers to Debian (and using
apache instead of ZWS). The step will start on Monday or Tuesday with the
removal of wolfsbane from the web-cluster (so ortelius will be the only
webserver for some time). I doubt that this step will take as long as the
previous (because most work in puppet is already done), but you can never
know.
I would hereby like to thank Merlissimo; without him the conversion would have
been much harder or impossible. Another thank to Nosy who freed me from
Solaris-work.
Sincerely,
DaB.
[1] https://wiki.toolserver.org/view/User:Dab/Debian-Packages
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885