|
2021-09-09
ยง
|
| 13:11 |
<mutante> |
planet1002 - re-enabling disabled puppet |
[production] |
| 13:06 |
<jmm@cumin2002> |
END (PASS) - Cookbook sre.idm.logout (exit_code=0) Logging Muehlenhoff out of all services on: 2 hosts |
[production] |
| 13:06 |
<jmm@cumin2002> |
START - Cookbook sre.idm.logout Logging Muehlenhoff out of all services on: 2 hosts |
[production] |
| 13:05 |
<jmm@cumin2002> |
END (PASS) - Cookbook sre.idm.logout (exit_code=0) Logging Muehlenhoff out of all services on: 2 hosts |
[production] |
| 13:05 |
<jmm@cumin2002> |
START - Cookbook sre.idm.logout Logging Muehlenhoff out of all services on: 2 hosts |
[production] |
| 13:03 |
<jmm@cumin2002> |
END (PASS) - Cookbook sre.idm.logout (exit_code=0) Logging Muehlenhoff out of all services on: 2 hosts |
[production] |
| 13:03 |
<jmm@cumin2002> |
START - Cookbook sre.idm.logout Logging Muehlenhoff out of all services on: 2 hosts |
[production] |
| 13:00 |
<mwdebug-deploy@deploy1002> |
helmfile [codfw] Ran 'sync' command on namespace 'mwdebug' for release 'pinkunicorn' . |
[production] |
| 12:56 |
<mwdebug-deploy@deploy1002> |
helmfile [eqiad] Ran 'sync' command on namespace 'mwdebug' for release 'pinkunicorn' . |
[production] |
| 10:49 |
<hnowlan@puppetmaster1001> |
conftool action : set/pooled=no; selector: name=maps1007.eqiad.wmnet |
[production] |
| 10:48 |
<hnowlan@cumin1001> |
END (PASS) - Cookbook sre.hosts.downtime (exit_code=0) for 5:00:00 on maps1007.eqiad.wmnet with reason: Resyncing from master |
[production] |
| 10:48 |
<hnowlan@cumin1001> |
START - Cookbook sre.hosts.downtime for 5:00:00 on maps1007.eqiad.wmnet with reason: Resyncing from master |
[production] |
| 10:48 |
<hnowlan@puppetmaster1001> |
conftool action : set/pooled=yes; selector: name=maps1007.eqiad.wmnet |
[production] |
| 10:48 |
<hnowlan@puppetmaster1001> |
conftool action : set/pooled=yes; selector: name=maps1006.eqiad.wmnet |
[production] |
| 10:47 |
<topranks> |
Removing peering to old IPs of AS139931 (BSCCL) at Equinix Singapore (cr3-eqsin). |
[production] |
| 10:45 |
<topranks> |
Removing peering to AS24218 at Equinix Singapore (cr3-eqsin) - network no longer uses this ASN. |
[production] |
| 10:22 |
<volans> |
upgrading spicerack on cumin1001 |
[production] |
| 10:20 |
<volans@cumin2002> |
END (FAIL) - Cookbook sre.hosts.decommission (exit_code=1) for hosts mc1027.eqiad.wmnet |
[production] |
| 10:10 |
<volans@cumin2002> |
START - Cookbook sre.hosts.decommission for hosts mc1027.eqiad.wmnet |
[production] |
| 09:56 |
<jmm@cumin2002> |
START - Cookbook sre.hosts.reboot-single for host mx2002.wikimedia.org |
[production] |
| 09:47 |
<volans@cumin2002> |
END (ERROR) - Cookbook sre.hosts.decommission (exit_code=97) for hosts mc1027.eqiad.wmnet |
[production] |
| 09:46 |
<volans@cumin2002> |
START - Cookbook sre.hosts.decommission for hosts mc1027.eqiad.wmnet |
[production] |
| 09:37 |
<godog> |
swift eqiad add ms-be10[64-67] with initial weight - T290546 |
[production] |
| 09:19 |
<filippo@puppetmaster1001> |
conftool action : set/pooled=false; selector: dnsdisc=swift-ro,name=eqiad |
[production] |
| 09:19 |
<filippo@puppetmaster1001> |
conftool action : set/pooled=false; selector: dnsdisc=swift,name=eqiad |
[production] |
| 09:15 |
<volans> |
rebooting sretest1001 to test ipmi reboot via spicerack |
[production] |
| 09:15 |
<volans@cumin2002> |
END (PASS) - Cookbook sre.hosts.downtime (exit_code=0) for 0:20:00 on sretest1001.eqiad.wmnet with reason: testing reboot via ipmi |
[production] |
| 09:15 |
<volans@cumin2002> |
START - Cookbook sre.hosts.downtime for 0:20:00 on sretest1001.eqiad.wmnet with reason: testing reboot via ipmi |
[production] |
| 09:13 |
<btullis@cumin1001> |
END (PASS) - Cookbook sre.aqs.roll-restart (exit_code=0) for AQS aqs cluster: Roll restart of all AQS's nodejs daemons. - btullis@cumin1001 |
[production] |
| 09:09 |
<btullis@cumin1001> |
START - Cookbook sre.aqs.roll-restart for AQS aqs cluster: Roll restart of all AQS's nodejs daemons. - btullis@cumin1001 |
[production] |
| 08:59 |
<godog> |
move swift traffic fully to codfw to rebalance eqiad - T287539 |
[production] |
| 08:59 |
<filippo@puppetmaster1001> |
conftool action : set/pooled=true; selector: dnsdisc=swift,name=codfw |
[production] |
| 08:58 |
<filippo@puppetmaster1001> |
conftool action : set/pooled=true; selector: dnsdisc=swift-ro,name=codfw |
[production] |
| 08:56 |
<volans> |
upgrading spicerack on cumin2002 to test the new release |
[production] |
| 08:50 |
<volans> |
uploaded spicerack_0.0.59 to apt.wikimedia.org buster-wikimedia,bullseye-wikimedia |
[production] |
| 08:23 |
<jelto> |
run ansible change 719041 on gitlab1001 |
[production] |
| 08:13 |
<jelto> |
run ansible change 719041 on gitlab2001 |
[production] |
| 07:07 |
<dzahn@cumin1001> |
END (PASS) - Cookbook sre.ganeti.makevm (exit_code=0) for new host durum1002.eqiad.wmnet |
[production] |
| 06:47 |
<dzahn@cumin1001> |
START - Cookbook sre.ganeti.makevm for new host durum1002.eqiad.wmnet |
[production] |
| 04:37 |
<ryankemper> |
[WDQS] Dispatched e-mail to the banned user agent (dailymotion) |
[production] |
| 03:57 |
<ryankemper> |
[WDQS] Dispatched e-mail to WDQS public mailing list informing them the outage is over; all that's left is the e-mail to the banned UA |
[production] |
| 03:47 |
<ryankemper> |
[WDQS] Restarting `wdqs-blazegraph` on `wdqs[2001-2008].codfw.wmnet`; if banning the dailymotion UA was sufficient then servers should come back up healthy and not drop back into deadlock |
[production] |
| 03:43 |
<ryankemper> |
[WDQS] Running puppet agent on `wdqs[2001-2008].codfw.wmnet` to roll out https://gerrit.wikimedia.org/r/719753 |
[production] |
| 03:29 |
<ryankemper> |
[WDQS] There's no clear indication of them being a culprit, but by far the most common user agent is a dailymotion VideocatalogTopic UA (see https://logstash.wikimedia.org/goto/51f238e9010d0220e5d33c6c210be93e) |
[production] |
| 03:12 |
<bstorm> |
attempting to start replication on clouddb1017 s1 T290630 |
[production] |
| 03:11 |
<bstorm> |
stopping and restarting mariadb on clouddb1017 s1 |
[production] |
| 03:04 |
<ryankemper> |
[WDQS] Dispatched email to Wikidata public mailing list about reduced service availability |
[production] |
| 02:36 |
<ryankemper> |
[WDQS] https://grafana.wikimedia.org/d/000000489/wikidata-query-service?viewPanel=7&orgId=1&from=1631152574841&to=1631154942992 shows the availability pattern, anywhere we see missing data (null) represents time that blazegraph was locked up and therefore unable to report metrics |
[production] |
| 02:34 |
<ryankemper> |
[WDQS] For context I glanced at `ryankemper@cumin1001:~$ sudo -E cumin 'P{wdqs2*}' 'sudo systemctl status wdqs-blazegraph'` before doing the aforementioned restarts and they'd all last restarted between 25-28 minutes ago |
[production] |
| 02:33 |
<ryankemper> |
[WDQS] Restarting `wdqs-blazegraph` across all of `wdqs2*` |
[production] |