|
2021-06-08
§
|
| 05:46 |
<marostegui@cumin1001> |
END (PASS) - Cookbook sre.hosts.downtime (exit_code=0) for 2:00:00 on db2123.codfw.wmnet with reason: REIMAGE |
[production] |
| 05:44 |
<marostegui@cumin1001> |
START - Cookbook sre.hosts.downtime for 2:00:00 on db2123.codfw.wmnet with reason: REIMAGE |
[production] |
| 05:17 |
<marostegui@cumin1001> |
END (PASS) - Cookbook sre.hosts.downtime (exit_code=0) for 2:00:00 on db2123.codfw.wmnet with reason: REIMAGE |
[production] |
| 05:15 |
<marostegui@cumin1001> |
START - Cookbook sre.hosts.downtime for 2:00:00 on db2123.codfw.wmnet with reason: REIMAGE |
[production] |
| 04:54 |
<marostegui> |
Repool clouddb1019:3314 |
[production] |
| 04:07 |
<ryankemper@cumin1001> |
END (PASS) - Cookbook sre.wdqs.data-transfer (exit_code=0) |
[production] |
| 02:38 |
<ryankemper@cumin1001> |
START - Cookbook sre.wdqs.data-transfer |
[production] |
| 02:38 |
<ryankemper> |
T284445 `sudo -i cookbook sre.wdqs.data-transfer --source wdqs1011.eqiad.wmnet --dest wdqs1012.eqiad.wmnet --reason "repairing overinflated blazegraph journal" --blazegraph_instance blazegraph` on `ryankemper@cumin1001` tmux session `wdqs` |
[production] |
| 02:37 |
<ryankemper> |
T284445 after manually stopping blazegraph/wdqs-updater, `sudo rm -fv /srv/wdqs/wikidata.jnl` on `wdqs1012` (clearing old overinflated journal file away before xferring new one) |
[production] |
| 02:34 |
<ryankemper> |
[WDQS] `ryankemper@wdqs1005:~$ sudo depool` (catching up on ~7h of lag) |
[production] |
|
2021-06-07
§
|
| 21:26 |
<otto@cumin1001> |
END (PASS) - Cookbook sre.kafka.roll-restart-brokers (exit_code=0) |
[production] |
| 21:12 |
<sbassett> |
Deployed security patch for T284364 |
[production] |
| 19:30 |
<ryankemper> |
T284479 [Cirrussearch] We'll keep monitoring. For now this incident is resolved. Glancing at our current volume relative to what we'd expect, the numbers we see match what we'd expect. If we're accidentally banning any innocent requests they must be an incredibly small percentage of the total otherwise we'd see significantly lower volume than expected |
[production] |
| 19:25 |
<ryankemper> |
T284479 [Cirrussearch] Seeing the expected drop in `entity_full_text` requests here: https://grafana-rw.wikimedia.org/d/000000455/elasticsearch-percentiles?viewPanel=47&orgId=1&from=now-12h&to=now As a result we're no longer rejecting any requests |
[production] |
| 19:21 |
<ryankemper> |
T284479 [Cirrussearch] We're working on rolling out https://gerrit.wikimedia.org/r/698607, which will ban search API requests that match the Google App Engine IP range `2600:1900::0/28` AND whose user agent includes `HeadlessChrome` |
[production] |
| 19:19 |
<cdanis> |
T284479 ✔️ cdanis@cumin1001.eqiad.wmnet ~ 🕞🍵 sudo cumin -b16 'A:cp-text' "run-puppet-agent" |
[production] |
| 19:07 |
<andrew@deploy1002> |
Finished deploy [horizon/deploy@6199b67]: disable shelve/unshelve T284462 (duration: 04m 53s) |
[production] |
| 19:02 |
<andrew@deploy1002> |
Started deploy [horizon/deploy@6199b67]: disable shelve/unshelve T284462 |
[production] |
| 19:01 |
<andrew@deploy1002> |
Finished deploy [horizon/deploy@6199b67]: disable shelve/unshelve (duration: 02m 01s) |
[production] |
| 18:59 |
<andrew@deploy1002> |
Started deploy [horizon/deploy@6199b67]: disable shelve/unshelve |
[production] |
| 18:57 |
<herron> |
prometheus3001: moved /srv back to vda1 filesystem T243057 |
[production] |
| 18:25 |
<urbanecm> |
[urbanecm@mwmaint1002 /srv/mediawiki/php-1.37.0-wmf.7]$ mwscript extensions/GrowthExperiments/maintenance/initWikiConfig.php --wiki=skwiki --phab=T284149 |
[production] |
| 18:24 |
<urbanecm@deploy1002> |
Synchronized php-1.37.0-wmf.7/extensions/GrowthExperiments/includes/WelcomeSurvey.php: 368b5d9: 0e79aee: WelcomeSurvey backports (T284127, T284257; 2/2) (duration: 00m 57s) |
[production] |
| 18:22 |
<urbanecm@deploy1002> |
Synchronized php-1.37.0-wmf.7/extensions/GrowthExperiments/extension.json: 368b5d9: 0e79aee: WelcomeSurvey backports (T284127, T284257; 1/2) (duration: 00m 56s) |
[production] |
| 18:20 |
<urbanecm@deploy1002> |
Synchronized php-1.37.0-wmf.7/extensions/GrowthExperiments/maintenance/initWikiConfig.php: 7089728: b2482fb: initWikiConfig GE backports (T284072) (duration: 00m 58s) |
[production] |
| 18:16 |
<urbanecm@deploy1002> |
Synchronized wmf-config/InitialiseSettings.php: 15e09109b7c45de967a496a0eb58ad267dbc5079: skwiki: Make Growth features available in dark mode (T284149; 3/3) (duration: 00m 56s) |
[production] |
| 18:14 |
<urbanecm@deploy1002> |
Synchronized dblists/growthexperiments.dblist: 15e09109b7c45de967a496a0eb58ad267dbc5079: skwiki: Make Growth features available in dark mode (T284149; 2/3) (duration: 00m 56s) |
[production] |
| 18:14 |
<otto@cumin1001> |
START - Cookbook sre.kafka.roll-restart-brokers |
[production] |
| 18:14 |
<ottomata> |
rolling restart of kafka jumbo brokers - T283067 |
[production] |
| 18:13 |
<urbanecm@deploy1002> |
Synchronized wmf-config/config/skwiki.yaml: 15e09109b7c45de967a496a0eb58ad267dbc5079: skwiki: Make Growth features available in dark mode (T284149; 1/3) (duration: 00m 59s) |
[production] |
| 18:12 |
<otto@cumin1001> |
END (PASS) - Cookbook sre.kafka.roll-restart-mirror-maker (exit_code=0) |
[production] |
| 18:04 |
<urbanecm> |
[urbanecm@mwmaint1002 ~]$ mwscript extensions/WikimediaMaintenance/createExtensionTables.php --wiki=skwiki growthexperiments # T284149 |
[production] |
| 18:04 |
<urbanecm@deploy1002> |
Synchronized wmf-config/InitialiseSettings.php: 5de2f8b27b016a2cd8f424d8e40318edde5e5704: Set WelcomeSurveyEnableWithHomepage (T281896, T284257) (duration: 00m 59s) |
[production] |
| 17:53 |
<otto@cumin1001> |
START - Cookbook sre.kafka.roll-restart-mirror-maker |
[production] |
| 17:53 |
<ottomata> |
rolling restart of kafka jumbo mirror makers - T283067 |
[production] |
| 17:17 |
<ryankemper> |
[Cirrussearch] We're seeing ~10% of current requests being rejected by poolcounter, due to ~2x expected `eqiad.full_text` query volume and ~30x expected `eqiad.entity_full_text` query volume |
[production] |
| 16:56 |
<ryankemper> |
[WDQS] `ryankemper@wdqs1005:~$ sudo systemctl restart wdqs-blazegraph` (blazegraph locked up) |
[production] |
| 16:51 |
<razzi> |
run homer '*.eqiad.wmnet' diff |
[production] |
| 16:49 |
<ottomata> |
restarting mysqld analytics-meta replica on db1108 to apply config change - T272973 |
[production] |
| 16:31 |
<ebernhardson@deploy1002> |
Finished deploy [wikimedia/discovery/analytics@19313f7]: Bump glent jar to 0.2.6 (duration: 04m 29s) |
[production] |
| 16:27 |
<ebernhardson@deploy1002> |
Started deploy [wikimedia/discovery/analytics@19313f7]: Bump glent jar to 0.2.6 |
[production] |
| 16:09 |
<ebernhardson@deploy1002> |
Finished deploy [wikimedia/discovery/analytics@f236b95]: Bump glent jar to 0.2.6 (duration: 00m 35s) |
[production] |
| 16:09 |
<ebernhardson@deploy1002> |
Started deploy [wikimedia/discovery/analytics@f236b95]: Bump glent jar to 0.2.6 |
[production] |
| 14:57 |
<moritzm> |
installing remaining lz4 security updates on buster |
[production] |
| 14:35 |
<moritzm> |
installing isc-dhcp security updates |
[production] |
| 14:17 |
<marostegui@cumin1001> |
dbctl commit (dc=all): 'Repool db1113 (s5,s6) after upgrade', diff saved to https://phabricator.wikimedia.org/P16315 and previous config saved to /var/cache/conftool/dbconfig/20210607-141722-marostegui.json |
[production] |
| 14:13 |
<marostegui@cumin1001> |
dbctl commit (dc=all): 'Depool db1113 (s5,s6) for upgrade', diff saved to https://phabricator.wikimedia.org/P16314 and previous config saved to /var/cache/conftool/dbconfig/20210607-141307-marostegui.json |
[production] |
| 13:35 |
<volans@deploy1002> |
Finished deploy [netbox/deploy@c70df91]: Force deploy of gerrit/672831 to netbox-next (3) (duration: 00m 52s) |
[production] |
| 13:34 |
<volans@deploy1002> |
Started deploy [netbox/deploy@c70df91]: Force deploy of gerrit/672831 to netbox-next (3) |
[production] |
| 13:34 |
<moritzm> |
installing libxml2 security updates on stretch |
[production] |