|
2016-10-24
§
|
| 15:04 |
<bblack> |
enabling/running puppet on caches for 8x varnish ports changes - T107749 |
[production] |
| 14:57 |
<paravoid> |
restarting ferm on es2015 |
[production] |
| 14:54 |
<bblack> |
starting ferm server on eeden, radon |
[production] |
| 14:41 |
<gehel@puppetmaster1001> |
conftool action : set/pooled=yes; selector: dc=eqiad,cluster=maps,service=kartotherian,name=maps1002.eqiad.wmnet |
[production] |
| 14:38 |
<dereckson@mira> |
Synchronized wmf-config/CommonSettings.php: Toggle wgDefaultUserOptions['watchdefault'] on for cs.wikipedia, off elsewhere (T148328, 2/2) (duration: 00m 50s) |
[production] |
| 14:36 |
<dereckson@mira> |
Synchronized wmf-config/InitialiseSettings.php: Toggle wgDefaultUserOptions['watchdefault'] on for cs.wikipedia, off elsewhere (T148328, 1/2) (duration: 00m 54s) |
[production] |
| 14:36 |
<bblack> |
disabling puppet on all caches ahead of port# work, to test - T107749 / https://gerrit.wikimedia.org/r/#/c/317405 |
[production] |
| 14:29 |
<yurik> |
re-deployed current kartotherian to all servers (maps1002 & maps-test* were stale) |
[production] |
| 14:11 |
<marostegui> |
Deploy schema change s5 dewiki.revision - only codfw T148967 |
[production] |
| 14:03 |
<l10nupdate@mira> |
ResourceLoader cache refresh completed at Mon Oct 24 14:03:07 UTC 2016 (duration 6m 17s) |
[production] |
| 13:56 |
<dereckson@mira> |
scap sync-l10n completed (1.28.0-wmf.22) (duration: 10m 46s) |
[production] |
| 13:42 |
<bblack> |
restarting all varnish frontends (serially per-cluster with proper depooling, etc) |
[production] |
| 13:20 |
<elukey> |
reimaging mc120[89] and mc1030 |
[production] |
| 13:18 |
<Dereckson> |
Started manually l10nupdate, as it didn't run for 6 days, and more especially to fix T148921 user-facing issue. |
[production] |
| 13:13 |
<dereckson@mira> |
Synchronized wmf-config/throttle.php: Edit-a-thon BDA (Poitiers) throttle rule (T148852) (duration: 01m 13s) |
[production] |
| 10:47 |
<elukey> |
reimaged mc102[56], currently doing mc1027 |
[production] |
| 10:21 |
<_joe_> |
rebooting kubernetes1002 |
[production] |
| 09:20 |
<mobrovac> |
change-prop deploying c7feda2 |
[production] |
| 09:09 |
<mobrovac> |
restbase deploy end of f9017ad |
[production] |
| 08:55 |
<akosiaris> |
rebooting cobalt (gerrit) for kernel upgrades |
[production] |
| 08:53 |
<elukey> |
reimaging mc1024 |
[production] |
| 08:46 |
<mobrovac> |
restbase deploy start of f9017ad |
[production] |
| 08:38 |
<gehel> |
continue rolling restart of elasticsearch eqiad cluster |
[production] |
| 08:38 |
<hashar> |
Restarting gallium (Jenkins/Zuul) for kernel upgrades |
[production] |
| 08:36 |
<akosiaris> |
rebooting labnodepool1001 for kernel upgrades |
[production] |
| 08:36 |
<akosiaris> |
rebooting scandium for kernel upgrades |
[production] |
| 08:33 |
<hashar> |
rebooting contint1001 |
[production] |
| 08:20 |
<elukey> |
reimaging mc1023.eqiad.wmnet |
[production] |
| 07:46 |
<elukey> |
reimaging mc1022.eqiad.wmnet (T137345) |
[production] |
| 07:09 |
<marosteg1i> |
Deploying alter table s1.enwiki on codfw - T147166 |
[production] |
|
2016-10-21
§
|
| 23:45 |
<mutante> |
depooling maps1002 (by running "depool" on the server itself) |
[production] |
| 23:35 |
<yurik> |
maps1002.eqiad is running older/incorrect/misbehaving software for some reason, restart didn't help. Need to depool |
[production] |
| 22:17 |
<mutante> |
cp4006,cp4014 gzipped some logs in home for disk space |
[production] |
| 22:08 |
<mutante> |
cp4006, cp4014 were running out of disk, apt-get clean |
[production] |
| 21:40 |
<mutante> |
phab2001 that IP was also on iridium/phab1001, it should not be hardcoded in puppet, causing issues in T143363 |
[production] |
| 21:37 |
<mutante> |
phab2001 - ip addr del 10.64.32.186/21 dev eth0 |
[production] |
| 21:06 |
<bblack> |
restarting varnish backends (depooled, etc) for eqiad cache_upload: cp1049, cp1072, cp1074 |
[production] |
| 19:50 |
<cmjohnson1> |
dataset1001 array 1 swap failed disk slot 4 |
[production] |
| 19:40 |
<cmjohnson1> |
labvirt1005 swapping disk 0 |
[production] |
| 19:40 |
<gehel> |
routing traffic for cache-maps in codfw -> eqiad |
[production] |
| 19:29 |
<gehel> |
running puppet on eqiad cache nodes to activate maps traffic redirection |
[production] |
| 19:06 |
<gehel> |
shutting down cassandra on maps2004, seems to have lost data |
[production] |
| 18:22 |
<ejegg> |
updated SmashPig from d1ca0632d00dfb608f70ca4b70251a5ba49f4411 to e28b2cd9f0c1429acdd2a08c68f95884dbffb594 |
[production] |
| 16:45 |
<ejegg> |
updated fundraising tools from 09ae6e24d8ca8350dc099d63a6ca0d9ec9fdef2b to f83e39291adc55677fc4b49307dc4807eba18019 |
[production] |