|
2016-07-05
§
|
| 09:59 |
<elukey@palladium> |
conftool action : set/weight=30; selector: mw1272.eqiad.wmnet |
[production] |
| 09:31 |
<_joe_> |
shutting down mw1009-16 for decommissioning |
[production] |
| 09:06 |
<_joe_> |
decommissioning mw1009-16 |
[production] |
| 08:38 |
<elukey@palladium> |
conftool action : set/pooled=yes; selector: mw1275.eqiad.wmnet |
[production] |
| 08:36 |
<elukey@palladium> |
conftool action : set/pooled=yes; selector: mw1274.eqiad.wmnet |
[production] |
| 08:32 |
<gehel> |
deleting enwikisource_titlesuggest on elasticsearch codfw (index creation issue during cluster restart) |
[production] |
| 08:31 |
<elukey@palladium> |
conftool action : set/pooled=yes; selector: mw1273.eqiad.wmnet |
[production] |
| 08:23 |
<elukey@palladium> |
conftool action : set/pooled=yes; selector: mw1272.eqiad.wmnet |
[production] |
| 08:21 |
<elukey> |
adding and pooling new appservers - mw127[2345].eqiad |
[production] |
| 08:07 |
<godog> |
swift codfw-prod: ms-be202[567] weight 1500 |
[production] |
| 07:55 |
<jynus> |
dropping etherpad_restore2 database from m1 T138516 |
[production] |
| 07:40 |
<akosiaris> |
T138516 forcing a puppet run on cache::misc hosts after merging https://gerrit.wikimedia.org/r/297352 |
[production] |
| 07:29 |
<akosiaris> |
T138516 stop the secondary etherpad instance on etherpad1001. etherpad-restore.wikimedia.org has served its purpose, killing it |
[production] |
| 02:44 |
<l10nupdate@tin> |
ResourceLoader cache refresh completed at Tue Jul 5 02:44:09 UTC 2016 (duration 6m 12s) |
[production] |
| 02:37 |
<mwdeploy@tin> |
scap sync-l10n completed (1.28.0-wmf.8) (duration: 17m 13s) |
[production] |
|
2016-07-04
§
|
| 20:28 |
<jynus> |
removing /tmp/joal/sstables on all analytics10* hosts |
[production] |
| 20:22 |
<jynus> |
deleted 21GB worth of temporary files from analytics1050 |
[production] |
| 19:58 |
<aaron@tin> |
Synchronized wmf-config/filebackend-production.php: Increase redis lockmanager timeout to 2 (duration: 00m 31s) |
[production] |
| 19:57 |
<legoktm@tin> |
Synchronized php-1.28.0-wmf.8/extensions/MassMessage/: MassMessage is no longer accepting lists in the MassMessageList content model - T139303 (duration: 00m 39s) |
[production] |
| 17:37 |
<jynus> |
testing slave_parallel_threads=5 on db1073 |
[production] |
| 14:27 |
<moritzm> |
rebooting lithium for kernel update |
[production] |
| 14:22 |
<moritzm> |
installing tomcat7/ libservlet3.0-java security update on the kafka brokers |
[production] |
| 14:06 |
<_joe_> |
shutting down mw1001-1008 for decommissioning |
[production] |
| 14:03 |
<gehel> |
rolling restart of elasticsearch codfw/eqiad for kernel upgrade (T138811) |
[production] |
| 13:47 |
<_joe_> |
stopping jobrunner on mw1011-16 as well, befor decommissioning |
[production] |
| 13:46 |
<moritzm> |
depooling mw1153-mw1160 (trusty image scalers), replaced by mw1291-mw1298 (jessie image scalers) |
[production] |
| 13:44 |
<godog> |
ack all mr1-codfw related alerts in librenms |
[production] |
| 13:43 |
<akosiaris> |
restart smokeping on netmon1001, temporarily disabled msw1-codfw |
[production] |
| 13:38 |
<gehel> |
resuming writes on Cirrus / elasticsearch, this did not speedup cluster recovery |
[production] |
| 13:18 |
<godog> |
bounce redis on rcs1001 |
[production] |
| 13:16 |
<gehel> |
restarting elastic1021 for kernel upgrade (T138811) |
[production] |
| 13:07 |
<elukey> |
Bootstrapping again Cassandra on aqs100[456] (rack awareness + 2.2.6 - testing environment) |
[production] |
| 13:02 |
<gehel> |
pausing writes on Cirrus / elasticsearch for faster cluster restart |
[production] |
| 12:43 |
<hashar> |
Nodepool back up with 10 instances (instead of 20) to accomodate for labs capacity T139285 |
[production] |
| 12:39 |
<godog> |
nodetool-b stop -- COMPACTION on restbase1014 |
[production] |
| 12:29 |
<moritzm> |
rolling reboot of rcs* cluster for kernel security update |
[production] |
| 12:10 |
<moritzm> |
rolling reboot of ocg* cluster for kernel security update |
[production] |
| 11:40 |
<jynus@tin> |
Synchronized wmf-config/db-eqiad.php: Failover db1053 to db1072 (duration: 00m 40s) |
[production] |
| 10:56 |
<moritzm> |
rolling reboot of swift frontends in eqiad for kernel security update |
[production] |
| 10:30 |
<yuvipanda> |
stop nodepool on labnodepool1001 and disable puppet to keep it down, to allow stabilizing labs first |
[production] |
| 10:28 |
<yuvipanda> |
restart rabbitmq-server on labcontrol1001 |
[production] |
| 10:14 |
<moritzm> |
installing chromium security update on osmium |
[production] |
| 10:07 |
<moritzm> |
installing xerces-c security updates on Ubuntu systems (jessie already fixed) |
[production] |
| 10:01 |
<_joe_> |
stopping jobchron and jobrunner on mw1001-10 before decommission |
[production] |
| 09:50 |
<godog> |
reimage ms-be300[234] with jessie |
[production] |
| 09:44 |
<hashar> |
Labs infra cant delete instances anymore (impacts CI as well) T139285 |
[production] |
| 09:41 |
<moritzm> |
installing p7zip security updates |
[production] |
| 09:38 |
<hashar> |
CI is out of Nodepool instances, the pool has drained because instances can no more be deleted over the OpenStack API |
[production] |
| 09:25 |
<elukey> |
Added new jobrunners in service - mw130[256].eqiad.wmnet (https://etherpad.wikimedia.org/p/jessie-install) |
[production] |
| 08:16 |
<moritzm> |
rolling reboot of swift backends in eqiad for kernel security update |
[production] |