Jump to content

Incidents/2022-12-21 shellbox-syntaxhighlight

From Wikitech

document status: draft


Incident metadata (see Incident Scorecard)
Incident ID 2022-12-21 shellbox-syntaxhighlight Start 2022-12-21 09:00:00
Task T325890 End 2022-12-21 09:10:00
People paged 2 Responder count 3
Coordinators jayme Affected metrics/SLOs No relevant SLOs exist, shellbox latency/php-fpm worker saturation
Impact For about 10min syntaxhighlighting was slow or returning errors

A sudden spike in requests to shellbox-syntaxhighlight overloaded the service leading to slow response times and increased failures.

The majority of requests where originating from jobrunners and api_appservers.


All times in UTC.

  • 09:00 OUTAGE BEGINS non paging alerts of failing http-api probes and api_appservers running out of idle workers where observed in #mediawiki-operations; Investigation started
  • 09:03 paging alert: (ProbeDown) firing: Service shellbox-syntaxhighlight:4014 has failed probes (http_shellbox-syntaxhighlight_ip4)
  • 09:05 shellbox-syntaxhighlight req/s increased from ~5 to ~70req/s
  • 09:05 shellbox-syntaxhighlight scaled up from 12 to 40 replicas
  • 09:09 Alerts recovering
  • 09:10 OUTAGE ENDS
  • 09:11 (CirrusSearchJobQueueBacklogTooBig) firing: CirrusSearch job topic eqiad.mediawiki.job.cirrusSearchLinksUpdate is heavily backlogged with 209k messages
  • 09:49 cirrusSearchLinksUpdate backlog was cleared
  • 09:54 shellbox-syntaxhighlight scaled back to 12 replicas


shellbox-syntaxhighlight grafana dashboard

Initially detected by a human spotting alerts on IRC, closely followed by a page.


What went well?

  • Problem spotted early and countermeasures where taken quickly
  • envoy telemetry dashboard
    One of the SREs paged knew how to increase the number of syntaxhighlight runners (it's not mentioned on Shellbox)

What went poorly?

  • shellbox-syntaxhighlight was probably overloaded by our own services (jobrunners), but currently this is an assumption.

Where did we get lucky?

  • The burst in requests was rather short

Links to relevant documentation


  • Figure out what caused the burst (there's a suggestion a template was changed leading to a lot of pages needing re-rendering at once)
  • Document how to scale up shellbox runners? (or link to it from Shellbox)


Incident Engagement ScoreCard
Question Answer


People Were the people responding to this incident sufficiently different than the previous five incidents? Yes
Were the people who responded prepared enough to respond effectively Yes
Were fewer than five people paged? Yes
Were pages routed to the correct sub-team(s)? Yes
Were pages routed to online (business hours) engineers?  Answer “no” if engineers were paged after business hours. Yes
Process Was the "Incident status" section atop the Google Doc kept up-to-date during the incident? No no google doc was created, only 3 responders
Was a public wikimediastatus.net entry created? No
Is there a phabricator task for the incident? Yes
Are the documented action items assigned? No
Is this incident sufficiently different from earlier incidents so as not to be a repeat occurrence? No
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.

No While they might not have fully prevented this incident, there are (long) outstanding performance issues with syntaxhighlight: task T271751, task T316858
Were the people responding able to communicate effectively during the incident with the existing tooling? Yes
Did existing monitoring notify the initial responders? Yes
Were the engineering tools that were to be used during the incident, available and in service? Yes
Were the steps taken to mitigate guided by an existing runbook? No
Total score (count of all “yes” answers above) 9