|
2017-06-21
§
|
| 10:50 |
<akosiaris> |
rebooting all ganeti200X nodes |
[production] |
| 10:47 |
<akosiaris> |
shutdown all VMs on the ganeti01.svc.codfw.wmnet cluster |
[production] |
| 10:43 |
<elukey> |
reboot analytics1001 (Hadoop master) for kernel update |
[production] |
| 10:35 |
<akosiaris> |
rebooting the entire codfw ganeti cluster for kernel upgrades. Silenced hosts in icinga already. T167643 |
[production] |
| 10:30 |
<moritzm> |
rebooting bast4001 for kernel update |
[production] |
| 10:21 |
<ema> |
reboot lvs[1001-1003] (eqiad primaries) for kernel update |
[production] |
| 10:17 |
<elukey> |
running a script in tmux on rdb[12]003 called "check" to dump periodically LLEN enwiki:jobqueue:enqueue:l-unclaimed and stopped the one on rdb2004 |
[production] |
| 10:07 |
<ema> |
reboot lvs[1004-1006] (eqiad secondaries) for kernel update |
[production] |
| 10:01 |
<elukey> |
reboot analytics1002 (Hadoop master standby) for kernel update |
[production] |
| 10:01 |
<moritzm> |
rebooting auth* servers for kernel update |
[production] |
| 09:48 |
<ema> |
reboot lvs[1010-1012] for kernel update |
[production] |
| 09:48 |
<elukey> |
reboot aqs1005 for kernel update |
[production] |
| 09:10 |
<elukey> |
reboot kafka2001 for kernel update (eventbus codfw) |
[production] |
| 09:06 |
<moritzm> |
rebooting restbase1017 for kernel update |
[production] |
| 08:52 |
<oblivian@puppetmaster1001> |
conftool action : set/pooled=inactive; selector: name=restbase2001.codfw.wmnet,dc=codfw,service=restbase |
[production] |
| 08:49 |
<_joe_> |
correction: restarting pybal |
[production] |
| 08:49 |
<_joe_> |
restarting etcd on lvs2003/2006, connection lost to etcd |
[production] |
| 08:34 |
<elukey> |
reboot kafka1012 for kernel upgrades |
[production] |
| 08:34 |
<marostegui> |
Deploy alter table db1070 s5 - T166207 |
[production] |
| 08:33 |
<marostegui@tin> |
Synchronized wmf-config/db-eqiad.php: Depool db1070 - T166207 (duration: 00m 44s) |
[production] |
| 08:27 |
<marostegui@tin> |
Synchronized wmf-config/db-eqiad.php: Repool db1082 - T166207 (duration: 00m 45s) |
[production] |
| 08:26 |
<godog> |
reimage ms-be1014 / 1015 with jessie |
[production] |
| 07:37 |
<marostegui> |
Stop and reset slave s5 on dbstore2001 - T168354 |
[production] |
| 06:23 |
<mutante> |
planet2001 wget missing unpuppetized logo file from https://en.planet.wikimedia.org/images/planet-wm2.png - should fix puppet run |
[production] |
| 06:19 |
<marostegui> |
Stop replication and puppet on db2066 for maintenance - T168354 |
[production] |
| 06:18 |
<marostegui@tin> |
Synchronized wmf-config/db-codfw.php: Depool db2066 - T168354 (duration: 00m 43s) |
[production] |
| 06:08 |
<elukey> |
reboot thorium for kernel upgrades (outage to all the analytics websites) |
[production] |
| 06:05 |
<marostegui> |
Deploy alter table s5 - db1082 - T166207 |
[production] |
| 06:04 |
<marostegui@tin> |
Synchronized wmf-config/db-eqiad.php: Depool db1082 - T166207 (duration: 00m 44s) |
[production] |
| 06:04 |
<marostegui> |
Deploy alter table s5 - dbstore1002 - T166207 |
[production] |
| 05:59 |
<elukey> |
reboot stat100[2,3,4] for kernel upgrades |
[production] |
| 05:57 |
<marostegui@tin> |
Synchronized wmf-config/db-eqiad.php: Repool db1087 - T166207 (duration: 00m 44s) |
[production] |
| 05:54 |
<marostegui> |
Deploy alter table s5 - labsdb1011 - T166207 |
[production] |
| 05:50 |
<marostegui@tin> |
Synchronized wmf-config/db-eqiad.php: Repool db1021 - T166205 (duration: 01m 00s) |
[production] |
| 05:41 |
<marostegui> |
Start relearn BBU cycle on db1016 - T166344 |
[production] |
| 03:13 |
<mutante> |
planet - copying HTML files from docroot from planet1001 to planet2001 - (don't serve Debian default page) |
[production] |
| 03:03 |
<mutante> |
planet1001 - remove/purge all php5* packages |
[production] |
| 02:57 |
<l10nupdate@tin> |
ResourceLoader cache refresh completed at Wed Jun 21 02:57:19 UTC 2017 (duration 6m 41s) |
[production] |
| 02:50 |
<l10nupdate@tin> |
scap sync-l10n completed (1.30.0-wmf.6) (duration: 06m 06s) |
[production] |
| 02:26 |
<l10nupdate@tin> |
scap sync-l10n completed (1.30.0-wmf.5) (duration: 06m 52s) |
[production] |
| 01:45 |
<mutante> |
planet1001 - remove php5 package |
[production] |
| 00:34 |
<mutante> |
planet2001 - revoke old puppet cert, salt-key, re-add new cert/key after reinstall |
[production] |
| 00:24 |
<mutante> |
planet2001 - scheduled downtime, reinstall with stretch |
[production] |
| 00:06 |
<mutante> |
tin (deployment): manually remove l10nupdate cron, let puppet re-create it after gerrit:350749. stops l10nupdate cron from running on weekends. naos didn't need an action. (T164035). |
[production] |