|
2009-03-20
ยง
|
| 23:54 |
<Tim> |
depooled srv125 from ES, has been down for 12 days. cluster12 is now down to 1 server |
[production] |
| 23:54 |
<tstarling> |
synchronized php-1.5/db.php |
[production] |
| 23:52 |
<Tim> |
install ganglia-metrics on db24 |
[production] |
| 23:13 |
<Tim> |
set up logrotate on locke, using the same script that we use for MW debug logs |
[production] |
| 22:56 |
<domas> |
kickbanned security threat from #wikimedia-tech, was trying to install keylogger and steal our passwords |
[production] |
| 22:54 |
<Tim> |
removed old squid log stream going to iris. Set up a log stream going from all the squids to locke. |
[production] |
| 22:37 |
<Tim> |
depooled adler, it's down |
[production] |
| 22:36 |
<tstarling> |
synchronized php-1.5/db.php 'depooled adler' |
[production] |
| 21:01 |
<mark> |
removed vlan 5 on csw5-pmtpa that was accidently created/left behind by Tim |
[production] |
| 20:47 |
<domas> |
installed snaprotate on db26, enabled snapshots with 2x8h schedule, updated [[Database snapshots]] |
[production] |
| 20:44 |
<midom> |
synchronized php-1.5/includes/api/ApiQueryRevisions.php 'livemerging r48642' |
[production] |
| 20:43 |
<midom> |
synchronized php-1.5/includes/filerepo/ArchivedFile.php 'livemerging r48644' |
[production] |
| 20:27 |
<domas> |
someone who wrote ArchivedFile::load, needs some pain and torture applied (query doesn't use index... ;-) |
[production] |
| 20:25 |
<midom> |
synchronized php-1.5/db.php 'reduced load for db26 from 200 to 100, as it has reduced amount of RAM and increased amount of other work' |
[production] |
| 20:09 |
<domas> |
added db26 to s1 pool |
[production] |
| 20:08 |
<midom> |
synchronized php-1.5/db.php |
[production] |
| 18:47 |
<mark> |
JeLuF> |log \\o/ |
[production] |
| 18:46 |
<domas> |
pooled db22 back in, as it caught up on replication! |
[production] |
| 18:46 |
<mark> |
raised karma of /h/w/b/reset-mysql-slave |
[production] |
| 18:46 |
<midom> |
synchronized php-1.5/db.php |
[production] |
| 18:45 |
<domas> |
ran /h/w/b/reset-mysql/slave on db26, after copy from db22! |
[production] |
| 18:45 |
<domas> |
wrote /h/w/b/reset-mysql-slave to reset mysql slaves! |
[production] |
| 18:32 |
<jeluf> |
synchronized php-1.5/InitialiseSettings.php '15880 Pseudo-Namespace on Korean Wikipedia' |
[production] |
| 18:25 |
<jeluf> |
synchronized php-1.5/InitialiseSettings.php '17699 Create Appendix namespace on Spanish Wiktionary' |
[production] |
| 18:19 |
<jeluf> |
synchronized php-1.5/InitialiseSettings.php '17987 Set $wgBlockAllowsUTEdit = true for zh.wikipedia' |
[production] |
| 18:16 |
<jeluf> |
synchronized php-1.5/InitialiseSettings.php '16446 tlwikibooks namespaces to be searched by default' |
[production] |
| 18:14 |
<jeluf> |
synchronized php-1.5/InitialiseSettings.php '16446 Add Pagluluto: namespace to tlwikibooks' |
[production] |
| 18:12 |
<domas> |
db26 bootstrap problem was inconsistent ibdata specification, probably first copy would've been enough :) |
[production] |
| 17:55 |
<jeluf> |
synchronized php-1.5/InitialiseSettings.php '17818 Create "Wikijunior" namespace on Polish Wikibooks (fix)' |
[production] |
| 17:53 |
<jeluf> |
synchronized php-1.5/InitialiseSettings.php '17818 Create "Wikijunior" namespace on Polish Wikibooks' |
[production] |
| 17:41 |
<domas> |
I'm out of luck, copy from db22 to db26 has failed again |
[production] |
| 17:32 |
<JeLuF> |
changed sync-common-file to automatically log messages to [[Server admin log]] |
[production] |
| 17:27 |
<jeluf> |
synchronized php-1.5/InitialiseSettings.php "17739 nowiktionary logo" |
[production] |
| 17:18 |
<JeLuF> |
changed slwikibooks, slwikisource logos to $stdlogo |
[production] |
| 17:17 |
<domas> |
tried to poke Werdna about AbuseFilter regexp warnings, but he didn't listen (check remotelogtail @ db20 ) |
[production] |
| 17:16 |
<domas> |
restarted srv170 - was throwing occasional segfaults |
[production] |
| 17:15 |
<domas> |
'RESET SLAVE" on db22, to clean the relay log congestion |
[production] |
| 17:14 |
<JeLuF> |
changed mgwiki logo to $stdlogo |
[production] |
| 17:13 |
<domas> |
attempted to reboot adler using sysrq, failed at it. adler needs datacenter service |
[production] |
| 17:12 |
<domas> |
did set up Jens access rights on wikitech, because nobody did that before |
[production] |
| 17:11 |
<JeLuF> |
sahwiki and stqwiki logos changed to $stdlogo |
[production] |
| 17:11 |
<domas> |
tested block compression on 'pagelinks' table on lomaria's innodb/1.0.3 |
[production] |
| 17:11 |
<domas> |
tested key packing on 'pagelinks' table on lomaria's innodb/1.0.3 |
[production] |
| 17:10 |
<domas> |
erased test mysql 5.1 builds on db26 |
[production] |
| 17:10 |
<domas> |
restarted data copy from db22 to db26 |
[production] |
| 17:10 |
<domas> |
forgot to stop mysqld on db26 :) |
[production] |
| 17:09 |
<domas> |
started copying data from db22 to db26 |
[production] |
| 17:09 |
<domas> |
took out db22 for copy to db26 |
[production] |
| 16:53 |
<JeLuF> |
locked rn.wiktionary |
[production] |
| 12:48 |
<domas> |
truncated log/remote on db20, had 6G of adler kernel noise, firewalled out adler syslog stream |
[production] |