Jump to content

Incidents/2022-12-12 wdqs codfw brief outage

From Wikitech

document status: draft

Summary

Incident metadata (see Incident Scorecard)
Incident ID 2022-12-12 wdqs codfw brief outage Start 2022-12-12 20:14:17
Task End 2022-12-12 20:30:00
People paged 3 Responder count 1
Coordinators 1 Affected metrics/SLOs WDQS Uptime SLO
Impact For about 15 minutes, wdqs queries routed to codfw failed


A large influx in requests led to excessive thread pool usage from codfw blazegraph backends, with concomitant increases in CPU load and throttling filter state size. This triggered a known bug in Blazegraph where its improper thread management leads to deadlock.

The system self healed, perhaps due to application-level throttling state throttling the offender.

Most or all CODFW-routed wdqs requests during the incident window failed.

Timeline

Beginning of incident

2022-12-12 20:13:18 <+jinxer-wm> (ProbeDown) firing: Service wdqs-ssl:443 has failed probes (http_wdqs-ssl_ip4) #page - https://wikitech.wikimedia.org/wiki/Runbook#wdqs-ssl:443 - https://grafana.wikimedia.org/d/O0nHhdhnz/network-probes-overview?var-job=probes/service&var-module=All - https://alerts.wikimedia.org/?q=alertname%3DProbeDown

2022-12-12 20:13:18 <+jinxer-wm> (ProbeDown) firing: Service wdqs-ssl:443 has failed probes (http_wdqs-ssl_ip4) - https://wikitech.wikimedia.org/wiki/Runbook#wdqs-ssl:443 - https://grafana.wikimedia.org/d/O0nHhdhnz/network-probes-overview?var-job=probes/service&var-module=All - https://alerts.wikimedia.org/?q=alertname%3DProbeDown

2022-12-12 20:14:17 <+icinga-wm> PROBLEM - PyBal backends health check on lvs2009 is CRITICAL: PYBAL CRITICAL - CRITICAL - wdqs-heavy-queries_8888: Servers wdqs2003.codfw.wmnet, wdqs2001.codfw.wmnet, wdqs2004.codfw.wmnet, wdqs2002.codfw.wmnet are marked down but pooled: wdqs-ssl_443: Servers wdqs2003.codfw.wmnet, wdqs2001.codfw.wmnet, wdqs2004.codfw.wmnet, wdqs2002.codfw.wmnet are marked down but pooled: wdqs_80: Servers wdqs2003.codfw.wmnet, wdqs2001.codfw.wmnet, wdqs20

2022-12-12 20:14:17 <+icinga-wm> .wmnet, wdqs2002.codfw.wmnet are marked down but pooled https://wikitech.wikimedia.org/wiki/PyBal

2022-12-12 20:14:17 <+icinga-wm> PROBLEM - WDQS SPARQL on wdqs2001 is CRITICAL: CRITICAL - Socket timeout after 10 seconds https://wikitech.wikimedia.org/wiki/Wikidata_query_service/Runbook

2022-12-12 20:14:17 <+icinga-wm> PROBLEM - WDQS SPARQL on wdqs2003 is CRITICAL: CRITICAL - Socket timeout after 10 seconds https://wikitech.wikimedia.org/wiki/Wikidata_query_service/Runbook

2022-12-12 20:25:00 Rough end of incident (end of user impact)

2022-12-12 20:31:00 [WDQS] ryankemper@cumin2002:~$ sudo -E cumin -b 4 wdqs2* 'systemctl restart wdqs-blazegraph' By this point, the service had already recovered, but was restarted out of an abundance of caution.

Detection

The issue was rapidly detected by pybal monitoring of WDQS. A page was emitted by the monitoring system fairly quickly.

2022-12-12 20:13:18 <+jinxer-wm> (ProbeDown) firing: Service wdqs-ssl:443 has failed probes (http_wdqs-ssl_ip4) #page - https://wikitech.wikimedia.org/wiki/Runbook#wdqs-ssl:443 - https://grafana.wikimedia.org/d/O0nHhdhnz/network-probes-overview?var-job=probes/service&var-module=All - https://alerts.wikimedia.org/?q=alertname%3DProbeDown

2022-12-12 20:13:18 <+jinxer-wm> (ProbeDown) firing: Service wdqs-ssl:443 has failed probes (http_wdqs-ssl_ip4) - https://wikitech.wikimedia.org/wiki/Runbook#wdqs-ssl:443 - https://grafana.wikimedia.org/d/O0nHhdhnz/network-probes-overview?var-job=probes/service&var-module=All - https://alerts.wikimedia.org/?q=alertname%3DProbeDown

2022-12-12 20:14:17 <+icinga-wm> PROBLEM - PyBal backends health check on lvs2009 is CRITICAL: PYBAL CRITICAL - CRITICAL - wdqs-heavy-queries_8888: Servers wdqs2003.codfw.wmnet, wdqs2001.codfw.wmnet, wdqs2004.codfw.wmnet, wdqs2002.codfw.wmnet are marked down but pooled: wdqs-ssl_443: Servers wdqs2003.codfw.wmnet, wdqs2001.codfw.wmnet, wdqs2004.codfw.wmnet, wdqs2002.codfw.wmnet are marked down but pooled: wdqs_80: Servers wdqs2003.codfw.wmnet, wdqs2001.codfw.wmnet, wdqs20

2022-12-12 20:14:17 <+icinga-wm> .wmnet, wdqs2002.codfw.wmnet are marked down but pooled https://wikitech.wikimedia.org/wiki/PyBal

Conclusions

OPTIONAL: General conclusions (bullet points or narrative)

What went well?

OPTIONAL: (Use bullet points) for example: automated monitoring detected the incident, outage was root-caused quickly, etc

What went poorly?

OPTIONAL: (Use bullet points) for example: documentation on the affected service was unhelpful, communication difficulties, etc

Where did we get lucky?

OPTIONAL: (Use bullet points) for example: user's error report was exceptionally detailed, incident occurred when the most people were online to assist, etc

  • The wdqs codfw dashboard shows a large spike in queries for wdqs2*, with a corresponding increase in CPU load and the throttling filter state.

Actionables

task T325324

Scorecard

Incident Engagement ScoreCard
Question Answer

(yes/no)

Notes
People Were the people responding to this incident sufficiently different than the previous five incidents?
Were the people who responded prepared enough to respond effectively
Were fewer than five people paged?
Were pages routed to the correct sub-team(s)?
Were pages routed to online (business hours) engineers?  Answer “no” if engineers were paged after business hours.
Process Was the incident status section actively updated during the incident?
Was the public status page updated?
Is there a phabricator task for the incident?
Are the documented action items assigned?
Is this incident sufficiently different from earlier incidents so as not to be a repeat occurrence?
Tooling To the best of your knowledge was the open task queue free of any tasks that would have prevented this incident? Answer “no” if there are

open tasks that would prevent this incident or make mitigation easier if implemented.

Were the people responding able to communicate effectively during the incident with the existing tooling?
Did existing monitoring notify the initial responders?
Were the engineering tools that were to be used during the incident, available and in service?
Were the steps taken to mitigate guided by an existing runbook?
Total score (count of all “yes” answers above)