|
2018-01-18
ยง
|
| 17:39 |
<moritzm> |
rebooting sodium (and temporarily disable icinga-wm due to some expected spam due to clients failing to run apt-get update) |
[production] |
| 17:33 |
<jynus> |
starting compare.py on s3 codfw (it triggered db2036 crash before) |
[production] |
| 17:31 |
<ema> |
cache_upload: repool cp3039 (varnish 5) |
[production] |
| 17:26 |
<ema> |
cache_upload: upgrade cp3039 to varnish 5 |
[production] |
| 17:02 |
<ema> |
cache_upload: repool cp3036 (varnish 5) |
[production] |
| 16:55 |
<ema> |
cache_upload: upgrade cp3036 to varnish 5 |
[production] |
| 15:54 |
<ema> |
cache_upload: repool cp3048 (varnish 5) |
[production] |
| 15:49 |
<ema> |
cache_upload: upgrade cp3048 to varnish 5 |
[production] |
| 15:40 |
<moritzm> |
rebooting labsdb1004 for kernel security update |
[production] |
| 15:40 |
<ema> |
cache_upload: repool cp3047 (varnish 5) |
[production] |
| 15:34 |
<ema> |
cache_upload: upgrade cp3047 to varnish 5 |
[production] |
| 15:33 |
<moritzm> |
reboot labsdb1006 (OSM slave) for kernel security update |
[production] |
| 15:20 |
<marostegui@tin> |
Synchronized wmf-config/db-eqiad.php: Repool db1066 - T162807 (duration: 01m 12s) |
[production] |
| 15:15 |
<mforns@tin> |
Finished deploy [analytics/refinery@78f98d9]: deploying refinery to add ISO codes to pageviews by country (duration: 04m 12s) |
[production] |
| 15:11 |
<mforns@tin> |
Started deploy [analytics/refinery@78f98d9]: deploying refinery to add ISO codes to pageviews by country |
[production] |
| 15:01 |
<marostegui> |
Stop replication in sync db1089 and db1066 - T162807 |
[production] |
| 15:00 |
<marostegui@tin> |
Synchronized wmf-config/db-eqiad.php: Depool db1066 - T162807 (duration: 01m 11s) |
[production] |
| 14:58 |
<moritzm> |
installing bind security updates (we only use the client-side tools) |
[production] |
| 14:58 |
<volans> |
reprepro includedeb jessie-wikimedia python-requests-mock_1.3.0-3_all.deb |
[production] |
| 14:45 |
<ema> |
cache_upload: repool cp3038 (varnish 5) |
[production] |
| 14:44 |
<herron> |
disabling puppet agents during deploy of 404587, 404689 |
[production] |
| 14:39 |
<ema> |
cache_upload: upgrade cp3038 to varnish 5 |
[production] |
| 14:39 |
<godog> |
restart hhvm on mw1233 |
[production] |
| 14:31 |
<_joe_> |
restarting hhvm on a few API appservers |
[production] |
| 14:30 |
<marostegui@tin> |
Synchronized wmf-config/db-eqiad.php: Repool db1087 - T174569 (duration: 01m 12s) |
[production] |
| 14:28 |
<ema> |
cache_upload: repool cp3035 (varnish 5) |
[production] |
| 14:25 |
<marostegui@tin> |
Synchronized wmf-config/db-codfw.php: Promote db2043 to s3 master after db2036 crash (duration: 01m 12s) |
[production] |
| 14:25 |
<godog> |
restart hhvm on mw1227 |
[production] |
| 14:23 |
<ema> |
cache_upload: upgrade cp3035 to varnish 5 |
[production] |
| 14:19 |
<jynus> |
starting mysql on db2043 |
[production] |
| 14:17 |
<jynus> |
stopping mysql on db2043 |
[production] |
| 14:10 |
<zeljkof> |
EU SWAT finished |
[production] |
| 14:10 |
<ema> |
cache_upload: repool cp3037 (varnish 5) |
[production] |
| 14:09 |
<zfilipin@tin> |
Synchronized wmf-config/InitialiseSettings.php: SWAT: [[gerrit:404911|Change autoconfirmed settings and Enable flood group at zhwikibooks (T185182)]] (duration: 01m 13s) |
[production] |
| 13:54 |
<ema> |
cache_upload: upgrade cp3037 to varnish 5 |
[production] |
| 13:49 |
<moritzm> |
upgrade mw* servers in eqiad running 3.18.5+dfsg-1+wmf3 (recent installations) to 3.18.5+dfsg-1+wmf4 |
[production] |
| 13:19 |
<jynus> |
changing topology of codfw s3 databases |
[production] |
| 13:05 |
<akosiaris> |
reboot poolcounter2001 for PCID/INVPCID CPU feature enabling |
[production] |
| 13:03 |
<akosiaris> |
reboot webperf1001 for PCID, INVPCID feature enabling (INVPCID not supported on current hardware, but still enabling it cluster wide) |
[production] |
| 12:57 |
<akosiaris> |
enable puppet across the fleet after nitrogen (puppetdb) reboot |
[production] |
| 12:56 |
<akosiaris> |
reboot nitrogen for PCID, INVPCID feature enabling (INVPCID not supported on current hardware, but still enabling it cluster wide) |
[production] |
| 12:52 |
<jgleeson> |
turned on donations queue consumer process-control job (actual time of change 17/01/18 ~16:20) |
[production] |
| 12:45 |
<akosiaris> |
reboot seaborgium for PCID, INVPCID feature enabling (INVPCID not supported on current hardware, but still enabling it cluster wide) |
[production] |
| 12:43 |
<elukey> |
bohrium rebooted for kernel upgrades |
[production] |
| 12:43 |
<akosiaris> |
disable puppet across the fleet for nitrogen (puppetdb) reboot |
[production] |
| 12:40 |
<elukey> |
set piwik in readonly mode and stopped mysql on bohrium (prep step for reboot) |
[production] |
| 12:36 |
<akosiaris> |
reboot chlorine.eqiad.wmnet etcd1003.eqiad.wmnet etcd1005.eqiad.wmnet fermium.wikimedia.org install1002.wikimedia.org krypton.eqiad.wmnet kubestagetcd1003.eqiad.wmnet logstash1009.eqiad.wmnet mwdebug1001.eqiad.wmnet sca1004.eqiad.wmnet for PCID, INVPCID feature enabling (INVPCID not supported on current hardware, but still enabling it cluster wide) |
[production] |
| 11:34 |
<akosiaris> |
reboot logstash1008 etcd1002 kubestagetcd1002.eqiad.wmnet for PCID, INVPCID feature enabling (INVPCID not supported on current hardware, but still enabling it cluster wide) |
[production] |
| 11:12 |
<ema> |
cp3046: restart varnish-be due to mbox lag |
[production] |
| 11:06 |
<volans> |
disabled puppet on tegmen to test impact on puppetdb - T170740 |
[production] |