Hello all,
wikidata has made it on the toolserver today. The database is called
"wikidatawiki_p" (don't ask me…) and its is home on s3. Because I guessed that
many joins to it will happen in future, I created also copies on all clusters
with the exception of s6 and s7 (will happen later); so you can join it in the
same way as commons.
Sincerely,
DaB.
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885
Hello all,
this is only for the tomcat-users, so most of you can ignore this eMail.
To solve TS-1550 [1] I decided to remove the postgresql-jar from tomcat's lib-
directory. If you need postgresql in your war, please include it there,
because I will remove the postgresql-jar at
Friday, 2. November 20 UTC.
I will also restart the tomcat-program afterwards.
Sincerely,
DaB.
[1] https://jira.toolserver.org/browse/TS-1550
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885
Hello all,
I thought it is time for another status-update, because some things happened
in the last days.
==s1-dump==
WMF and WMDE finally found an agreement that makes it possible for the
toolserver to get dumps again. As a first step the wmf-techs created a
s1(enwp)-dump for us. I transfer this dump at the moment to our cluster.
Somewhere on the way between Virginia and Amsterdam is a rate-limiting we have
to find, but otherwise the hole thing looks quite good.
When the dump is there, I will import it on thyme (during the import sql-s1-rr
will point to rosemary), than switch rosemary read-only, dump the user-
databases, import the user-databases to thyme and let point sql-s1-user to
thyme. After that Nosy will dump commons from rosemary to re-create sql-s4-
user on z-dat-s4-a. After that I will import the wmf-dump also on rosemary and
we should have 2 correct s1-hosts again.
Details when something will happen will be in a future eMail.
==TS-future==
It was not possible for me the convince the Vorstand of WMDE to create a
budget-position for the toolserver for the year 2013. To solve this I have
sent a request for the general member meeting few minutes ago. You can find the
request at [1] in german language. To summarize the account:
-Non-ending of the toolserver.
-Buying of 5 additional database-servers for 2012 at the beginning of 2013.
-Further extension in 2013 and behind.
-WMDE will pay for the hosting if WMF decide it will no longer do that.
-WMDE does all it can to convince WMF to provide db-replication and dumps.
-WMDE will continue to pay a sysop for the TS.
The general meeting can (and I guess will) change the request before it vote
about it, so we have to see what we get.
==University==
This week my university started again and so I have less time for the
toolserver than in the last months. So please accept if some thing will take a
little longer. I also can not be in IRC as often as I would like. So if you
find a problem or need something: Fill in jira-request and try to reach Nosy.
Sincerely,
DaB.
[1] de.wikipedia.org/wiki/Benutzer:DaB./MV-TS-Antrag
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885
Hello all,
yesterday I removed ([1],[2]) the last reminders of /sge62 (our old SGE-
installation we left in June) from our systems.
Until now 3 users had problems with missing SGE-components ([3]). If you have
problems too, please take a look at your PATH and if it looks ok, please add a
comment to [3] so we can find the problem together.
Sincerely,
DaB.
P.S: I removed only /sge62 — /sge (our current SGE-installation) is good and
fine.
[1] https://fisheye.toolserver.org/changelog/Puppet?cs=47
[2] https://fisheye.toolserver.org/changelog/Puppet?cs=48
[3] https://jira.toolserver.org/browse/TS-1544
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885
Hello all,
like announced in July [1] the default arch of SGE will switch soon from
solaris to *. Originally that should happen at 1. October, but I forgot
the re-announce it and I'm sure some of you forgot about it too. So the switch
is hereby announced for
9. October, 20:00 UTC.
Jobs that are executed after this timestamp and has no arch-option will run on
any host (linux or solaris) instead of a solaris-host.
There are 4 ways for you to prepare (sorted):
-Make sure that you program runs on linux AND solaris, add "arch=*".
-Make sure that you program runs only on linux, add "arch=lx".
-Make sure that you program runs only on solaris, add "arch=sol".
-Do nothing, pray and see things break.
(Somehow I have the feeling most of you will choose 4…, please make me wrong).
Sincerely,
DaB.
[1] http://lists.wikimedia.org/pipermail/toolserver-announce/2012-
July/000506.html
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885
Hello all,
doing last check-ups before the switch of SGE's arch tonight I discovered that
the linux-hosts had a problem with the mail-system. I fixed the problem, but
now all mails that were stored will be delivered to you – so if you receive
many emails about SGE-jobs (and other things) that are a few days old, that's
nothing to worry.
Sincerely,
DaB.
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885
Hello,
I will add more disk space to rosemary / sql-s1-user on
12th October 19:00-22:00 UTC
The user databases for s1 will not be accessible for about two hours.
sql-s1-rr (without user tables) will stay available during the maintenance.
Cheers
nosy
Hello,
I will add more disk space to adenia / sql-user on
10th October 19:00-21:00 UTC
The user databases will not be accessible for about one hour.
Cheers
nosy
Hello all,
like announced we have set up a temporary workaround for the user-store until
hemlock is fixed. It should work now, the only exception is willow. I can't
mount or umount the user-store on willow and the directory looks strange. So I
hereby announce a reboot of willow for tomorrow,
Friday, 16:00 UTC.
The downtime should be 10 minutes. You can follow the process at [1].
[1] https://jira.toolserver.org/browse/MNT-1272
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885
Hello all,
hemlock had another disc-problem and needed a emergency-reboot. It is back in
action now (more or less) but it lost one of its partitions. This partition
carries the user-store. A check-up (fsck) is running at the moment and it will
take a while to complete (I guess the hole european night). Most things should
work, but nothing that needs /mnt/user-store.
I will send an update when the user-store is back.
Sincerely,
DaB.
--
Userpage: [[:w:de:User:DaB.]] — PGP: 2B255885