|
2015-09-17
§
|
| 10:00 |
<dcausse> |
elastic in eqiad plugin updates: unfreezing indices |
[production] |
| 09:48 |
<dcausse> |
elastic in eqiad plugin updates: no more groovy in warmers, waiting for few more shards to move in elastic1001 and will unfreeze indices to test warmers |
[production] |
| 09:39 |
<dcausse> |
elastic in eqiad plugin updates: deleting warmers manually for old unused indices (eswikisource_content_1415240352, ruwiki_content_1415302164, thwiki_content_1415318677). We will have to remove these indices. |
[production] |
| 09:39 |
<paravoid> |
repooling ulsfo US-West traffic back to ulsfo for the first time since May :) |
[production] |
| 09:01 |
<dcausse> |
elastic in eqiad plugin updates: updating warmers on all wikis |
[production] |
| 08:58 |
<paravoid> |
penalizing ulsfo-eqiad direct MPLS links to higher OSPF weights |
[production] |
| 08:57 |
<paravoid> |
adjusting OSPF weights to be latency-based across the US network |
[production] |
| 08:53 |
<_joe_> |
removed iptables rules for dropping traffic to helium on mw1017 |
[production] |
| 08:52 |
<dcausse> |
elastic in eqiad plugin updates: index warmer queries are outdated with inline groovy script, updating warmers on warwiki first to test |
[production] |
| 08:05 |
<paravoid> |
eqiad-codfw -> eqiad-eqord-codfw migration |
[production] |
| 07:49 |
<moritzm> |
repooled mw1180-mw1188 (T104968) |
[production] |
| 07:42 |
<dcausse> |
elastic in eqiad plugin updates: restarting elastic1001 |
[production] |
| 07:42 |
<moritzm> |
depooled mw1180-mw1188 (T104968) |
[production] |
| 07:36 |
<moritzm> |
repooled mw1170-mw1179 (T104968) |
[production] |
| 07:36 |
<dcausse> |
elastic in eqiad plugin updates: freezing indices |
[production] |
| 07:27 |
<moritzm> |
depooled mw1170-mw1179 (T104968) |
[production] |
| 07:14 |
<_joe_> |
uploading new HHVM package |
[production] |
| 07:07 |
<moritzm> |
repooled mw1161-1168 (T104968) |
[production] |
| 06:57 |
<moritzm> |
depooled mw1161-1168 (T104968) |
[production] |
| 06:45 |
<moritzm> |
repooled mw1209-mw1220 with ferm enabled |
[production] |
| 06:33 |
<moritzm> |
depooling mw1209-mw1220 (in two steps) |
[production] |
| 05:47 |
<l10nupdate@tin> |
ResourceLoader cache refresh completed at Thu Sep 17 05:47:47 UTC 2015 (duration 47m 46s) |
[production] |
| 03:06 |
<l10nupdate@tin> |
LocalisationUpdate completed (1.26wmf23) at 2015-09-17 03:06:33+00:00 |
[production] |
| 03:03 |
<l10nupdate@tin> |
Synchronized php-1.26wmf23/cache/l10n: l10nupdate for 1.26wmf23 (duration: 06m 30s) |
[production] |
| 02:45 |
<l10nupdate@tin> |
LocalisationUpdate completed (1.26wmf22) at 2015-09-17 02:45:48+00:00 |
[production] |
| 02:39 |
<l10nupdate@tin> |
Synchronized php-1.26wmf22/cache/l10n: l10nupdate for 1.26wmf22 (duration: 11m 11s) |
[production] |
|
2015-09-16
§
|
| 23:27 |
<bblack> |
updating eqiad switch configs for lvs1007-1012 vlan/trunk settings |
[production] |
| 23:19 |
<krenair@tin> |
Synchronized php-1.26wmf23/extensions/MobileFrontend/resources/mobile.overlays/Overlay.less: https://gerrit.wikimedia.org/r/#/c/238865/ (duration: 00m 11s) |
[production] |
| 23:13 |
<gwicke> |
started `nodetool rebuild -- eqiad` on restbase-test200{1,2 |
[production] |
| 22:08 |
<urandom> |
disabling puppet in RESTBase eqiad staging cluster to test new code and config |
[production] |
| 22:08 |
<ottomata> |
powercycling analytcis1029, it is down? |
[production] |
| 20:47 |
<cscott> |
updated OCG to version 4032a596ce6eb442b02cc6ee9b79263b1eb23275 |
[production] |
| 19:38 |
<ori> |
Deployed statsv 0bfd9f06f / change I050a12d3b |
[production] |
| 18:47 |
<twentyafterfour@tin> |
rebuilt wikiversions.cdb and synchronized wikiversions files: group1 wikis to 1.26wmf23 |
[production] |
| 18:38 |
<twentyafterfour@tin> |
Synchronized php-1.26wmf23: syncing wmf23 ahead of deployment to group1 (duration: 01m 35s) |
[production] |
| 17:34 |
<paravoid> |
asw-d-eqiad: toggling RE mastership again |
[production] |
| 17:26 |
<godog> |
stop puppet on restbase* to apply https://gerrit.wikimedia.org/r/#/c/238738/ / merge / reenable puppet |
[production] |
| 16:54 |
<_joe_> |
turned on the hhvm tmh, stopping the zend ones for testing |
[production] |
| 16:44 |
<oblivian@tin> |
Synchronized wmf-config/CommonSettings.php: use ffmpeg whereever possible (duration: 00m 12s) |
[production] |
| 16:16 |
<bblack> |
upgrading pybal on lvs400[12] |
[production] |
| 16:12 |
<bblack> |
upgrading pybal on lvs400[34], lvs300[34] |
[production] |
| 16:08 |
<bblack> |
upgrading pybal on lvs200[123] |
[production] |
| 16:05 |
<bblack> |
upgrading pybal on lvs200[456] |
[production] |
| 15:44 |
<_joe_> |
uploading pybal 1.10 to reprepro, installing to the test cluster |
[production] |
| 15:24 |
<moritzm> |
uploaded debdeploy 0.0.6 to apt.wikimedia.org |
[production] |
| 15:10 |
<hashar> |
Started using Nodepool spawned instances. Moved integration-jjb-config-diff Jenkins job to Nodepool with https://gerrit.wikimedia.org/r/#/c/238752/ . See also: https://phabricator.wikimedia.org/T112750 |
[production] |
| 15:05 |
<demon@tin> |
Synchronized wmf-config/CommonSettings.php: Add m.wikidata.org to wgCrossSiteAJAXdomains (duration: 00m 12s) |
[production] |
| 14:51 |
<_joe_> |
experimenting on testwiki for poolcounter failure scenarios |
[production] |
| 14:45 |
<moritzm> |
enabled ferm on mw1010 (jobrunner) in eqiad |
[production] |
| 14:27 |
<paravoid> |
asw-d-eqiad: toggling RE mastership |
[production] |