Skip to content

2022-08-12: Web service latency and error rate SLO exceeded

Incident Roles

The DRI for this incident is the incident issue assignee, see roles and responsibilities.

Roles when the incident was declared:

Current Status

The canary Gitaly host is repeatedly being driven to memory starvation by a git fsck process consuming a large amount of anonymous memory. During each of these events, there is a phase where memory saturation erodes the filesystem cache, causing disk IO wait and memory reclaim overhead. This makes Gitaly unresponsive, which manifests as user-facing slowness and timeout-related errors in Workhorse.

The scope of impact should be limited to projects hosted on the canary stage Gitaly host (i.e. Gitlab's open source projects and their forks).

The triggering condition for each of the 9 memory saturation events was a DisconnectGitAlternates gRPC call that runs as part of the project deletion clean-up task. When deleting a fork of a project with a large object pool, the anonymous memory overhead from traversing those objects can be large, leading to memory pressure. This overhead problem is being discussed in gitlab-org/gitaly#4365 (closed) as a corrective action.

Summary for CMOC notice / Exec summary:

  1. Customer Impact: Intermittent slowness and errors when accessing any project stored on the canary gitaly node. This node mainly stores forks of GitLab projects such as gitlab-org/gitlab. Non-GitLab projects were not affected. Each performance regression lasted a few minutes.
  2. Service Impact: ServiceWeb ServiceGitaly
  3. Impact Duration: intermittently between 2022-08-12 13:00 - 19:00 UTC, lasting roughly 2 minutes during each of the 9 spikes.
  4. Root cause: RootCauseSaturation

Timeline

Recent Events (available internally only):

All times UTC.

2022-08-12

  • 15:08 - @msmiley declares incident in Slack.

Create related issues

Use the following links to create related issues to this incident if additional work needs to be completed after it is resolved:

Takeaways

  • ...

Corrective Actions

Corrective actions should be put here as soon as an incident is mitigated, ensure that all corrective actions mentioned in the notes below are included.

  • gitlab-org/gitaly#4365 (closed) - The DisconnectGitAlternates gRPC call (which runs as part of the project deletion) can incur a dangerously large anonymous memory overhead while git fsck is traversing a large number of objects to compose a connectivity graph. This issue discusses the overhead problem and potential solutions.

Note: In some cases we need to redact information from public view. We only do this in a limited number of documented cases. This might include the summary, timeline or any other bits of information, laid out in out handbook page. Any of this confidential data will be in a linked issue, only visible internally. By default, all information we can share, will be public, in accordance to our transparency value.


Click to expand or collapse the Incident Review section.

Incident Review
  • Ensure that the exec summary is completed at the top of the incident issue, the timeline is updated and relevant graphs are included in the summary
  • If there are any corrective action items mentioned in the notes on the incident, ensure they are listed in the "Corrective Action" section
  • Fill out relevant sections below or link to the meeting review notes that cover these topics

Customer Impact

  1. Who was impacted by this incident? (i.e. external customers, internal customers)
    1. ...
  2. What was the customer experience during the incident? (i.e. preventing them from doing X, incorrect display of Y, ...)
    1. ...
  3. How many customers were affected?
    1. ...
  4. If a precise customer impact number is unknown, what is the estimated impact (number and ratio of failed requests, amount of traffic drop, ...)?
    1. ...

What were the root causes?

  • ...

Incident Response Analysis

  1. How was the incident detected?
    1. ...
  2. How could detection time be improved?
    1. ...
  3. How was the root cause diagnosed?
    1. ...
  4. How could time to diagnosis be improved?
    1. ...
  5. How did we reach the point where we knew how to mitigate the impact?
    1. ...
  6. How could time to mitigation be improved?
    1. ...
  7. What went well?
    1. ...

Post Incident Analysis

  1. Did we have other events in the past with the same root cause?
    1. ...
  2. Do we have existing backlog items that would've prevented or greatly reduced the impact of this incident?
    1. ...
  3. Was this incident triggered by a change (deployment of code or change to infrastructure)? If yes, link the issue.
    1. ...

What went well?

  • ...

Guidelines

Resources

  1. If the Situation Zoom room was utilised, recording will be automatically uploaded to Incident room Google Drive folder (private)
Edited by Matt Smiley