Dev tools / Git hosting and CI / GitHub

GitHub

OPERATIONAL

Since Aug 2026 · 50 incidents · www.githubstatus.com · checked 2m ago

CompareStatus page
GradeFLast 73 days
DPRODFSHIP

Overview

30d90d1y
Admitted uptimeTime not covered by a widespread Sev 1 the provider posted itself. A best case.
99.869%Grade F
IncidentsEverything posted on the status page except scheduled maintenance.
492 Sev 1
Sev 1 downtimeTotal minutes of Sev 1 incidents, as posted.
2h 50m
Median fixMedian time to resolve. Half of incidents were fixed faster.
1h 18m
LongestLongest single incident, start to resolved.
20h 07m
Since lastTime since the last counted incident was resolved.
2h 27m

By area

DProduction
13INCIDENTS
2d 15hAFFECTED
Your users are hit: apps, APIs, databases, inference
Worst, 90d
SEV 2Aug 26, 2026 · 20h 07m · some customersDisruption with GitHub Billing
SEV 2Aug 20, 2026 · 9h 54m · some customersIntermittent failures creating agent tasks
SEV 2Aug 17, 2026 · 7h 36m · regionalIncident with GitHub.com
Most hit
Webhooks5
Pull Requests5
Actions4
Issues4
Worse. 13 in the last 6 months, 0 in the 6 before.
FShipping
35INCIDENTS
4d 4hAFFECTED
You can't deploy, build or manage: CI, CLI, dashboard, auth, billing
Worst, 90d
SEV 2Oct 22, 2026 · 11h 02mProject updates delayed
SEV 2Oct 12, 2026 · 7h 54mIncident with Actions, Pull Requests and Issues
SEV 1Aug 26, 2026 · 2h 50mIncident with Actions
Most hit
Actions14
Pull Requests13
Issues8
Webhooks6
Worse. 35 in the last 6 months, 0 in the 6 before.
CleanSev 3–4Sev 1–2No data

Trend, last 12 months

Worse. 49 in the last 6 months, 0 in the 6 before.

Severity mix

Sev 1 · Lights outProd site, API or database down, or data loss2
Sev 2 · BleedingProd degraded or partly down21
Sev 3 · Can't shipCan't deploy, build or manage it16
Sev 4 · PapercutDocs, status page, slow dashboard10

When it breaks

Start times, last 12 months, UTC.

What breaks

Actions14 incidents
Pull Requests13 incidents
Issues8 incidents
Copilot AI Model Providers8 incidents
Webhooks6 incidents
API Requests5 incidents

Communication

Updates eachStatus page updates per incident, on average.
6.9Last 12 months
Update gapMedian time between status page updates during an incident.
12mLast 12 months
PostmortemsSev 1–2 incidents with a published write-up.
0 of 23Sev 1–2

Incident history

Oct 22, 202611h 02mSEV 2
Project updates delayedIntermittent errors loading Issues, Pull Requests, search and Packages on github.com, plus degraded Actions and API requests, starting 2026-10-22. Capacity was added and the other services recovered by 18:26 UTC. GitHub Projects updates stayed delayed until mitigation at 00:55 UTC and resolution at 02:30 UTC on 2026-10-23. A root cause analysis is promised.AI-graded · API Requests +4
Why this grade
  1. RULES · SEV 3
    Sev 3 because API Requests had degraded performance
  2. CLAUDE · SEV 2 · GLOBAL · MEDIUM CONFIDENCE
    Issues, pull requests, search and Actions had intermittent errors for hours; Projects updates stayed delayed until about 10 hours in.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
18 updates
  1. 15:28 UTC · investigating
    We are investigating reports of degraded performance for Actions, Issues and Pull Requests
  2. 15:33 UTC · investigating
    Some users may experience intermittent errors loading Issues, Pull Requests and search results on github.com. <br /><br />We're investigating and working to restore normal service.
  3. 16:08 UTC · investigating
    Some users continue to experience intermittent errors loading Issues, Pull Requests and search results on github.com. <br /><br />We're investigating and evaluating mitigation options to restore normal service.
  4. 16:26 UTC · investigating
    API Requests is experiencing degraded performance. We are continuing to investigate.
  5. 16:28 UTC · investigating
    Packages is experiencing degraded performance. We are continuing to investigate.
  6. 16:35 UTC · investigating
    API Requests is operating normally.
  7. 16:41 UTC · investigating
    Some users continue to experience errors loading Issues, Pull Requests, package listings and search results on github.com. <br /><br />We're working through multiple mitigation efforts and continuing to investigate.
  8. 17:16 UTC · investigating
    We're seeing recovery after adding capacity, with errors decreasing. Some users may still experience errors loading Issues, Pull Requests and search results on github.com. <br /><br />We're continuing to work towards full recovery and monitoring for stability.
  9. 17:41 UTC · investigating
    Actions, Issues, Packages and Pull Requests are operating normally.
  10. 17:43 UTC · investigating
    We're seeing recovery across the affected services. GitHub Projects is still recovering as queued updates are processed, and we're continuing to monitor for stability.
  11. 18:26 UTC · investigating
    The other affected services have recovered. <br /><br />We're continuing to investigate GitHub Projects and verify that normal operation has been restored.
  12. 19:04 UTC · investigating
    Some updates in GitHub Projects are still delayed. We're continuing mitigation work and monitoring.
  13. 19:56 UTC · investigating
    Some GitHub Projects updates related to issue changes and merged pull requests are still delayed. We're investigating the remaining delays and working to restore normal processing.
  14. 20:46 UTC · investigating
    Some GitHub Projects updates, including newly added items and changes to issues and pull requests, remain delayed.<br /><br />We've identified interruptions in update processing and are evaluating mitigations to restore normal operation.
  15. 22:07 UTC · investigating
    Some GitHub Projects updates, including newly added items and changes to issues and pull requests, are still delayed.<br /><br />We have seen some recovery, and are working to add more capacity to accelerate the recovery.
  16. 23:48 UTC · investigating
    GitHub Projects update processing has improved, but newly added items may still take longer than expected to appear. <br /><br />We're continuing to work on the remaining delays and monitor recovery.
  17. 00:55 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  18. 02:30 UTC · resolved
    This incident has been resolved. Thank you for your patience and understanding as we addressed this issue. A detailed root cause analysis will be shared as soon as it is available.
Source
Oct 22, 20261h 01mSEV 3
Incident with Actions ResultsGitHub Actions workflow and step results were delayed for about an hour, along with some repository updates. Processing recovered after GitHub scaled processing capacity; no data loss was reported.AI-graded · Actions
Why this grade
  1. RULES · SEV 3
    Sev 3 because Actions had degraded performance
  2. CLAUDE · SEV 3 · UNCLEAR · MEDIUM CONFIDENCE
    Delays in processing Actions workflow and step results and some repository updates for about an hour; runs still completed, just reported late.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
7 updates
  1. 14:18 UTC · investigating
    We are investigating reports of degraded performance for Actions
  2. 14:19 UTC · investigating
    We’re investigating delays in processing GitHub Actions workflow and step results, along with some repository updates.
  3. 14:35 UTC · investigating
    Repository update processing has recovered. We’re continuing to work on delays affecting GitHub Actions workflow and step results. We’ll share another update as we make progress.
  4. 15:06 UTC · investigating
    We’re seeing recovery, and repository update processing has returned to normal. Delays affecting GitHub Actions workflow and step results are continuing to clear as we scale processing capacity; the remaining backlog is expected to take about 20 minutes to drain.
  5. 15:17 UTC · monitoring
    The degradation affecting Actions has been mitigated. We are monitoring to ensure stability.
  6. 15:18 UTC · monitoring
    GitHub Actions workflow and step result processing has recovered. Repository update processing has also returned to normal.
  7. 15:18 UTC · resolved
    This incident has been resolved. Thank you for your patience and understanding as we addressed this issue. A detailed root cause analysis will be shared as soon as it is available.
Source
Oct 21, 20264h 30mSEV 2
Delays for Pull Requests and Webhook deliveriesGitHub experienced delays to pull request updates, webhook deliveries, and Actions workflow status on October 21, caused by load on its event streaming platform. Pull requests and Actions recovered by about 17:00 UTC, and webhooks recovered soon after. Audit log events stayed delayed by up to 110 minutes before recovering by about 19:20 UTC. A root cause analysis is promised.AI-graded · Webhooks +2
Why this grade
  1. RULES · SEV 2
    Sev 2 because Webhooks had a major outage
  2. CLAUDE · SEV 2 · GLOBAL · MEDIUM CONFIDENCE
    Pull request updates, webhook deliveries and Actions status were delayed for about one to two hours; audit log delays lasted longer, with no data loss.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
15 updates
  1. 14:52 UTC · investigating
    We are investigating reports of degraded performance for Actions, Pull Requests and Webhooks
  2. 15:15 UTC · investigating
    Actions is experiencing degraded performance. We are continuing to investigate.
  3. 15:36 UTC · investigating
    We are investigating delays affecting pull request updates and webhook deliveries. Increased load on our event streaming platform is contributing to hotspots. We are shifting traffic away from affected areas and adding processing capacity; impact is ongoing while these mitigations take effect.
  4. 15:58 UTC · investigating
    We continue to see delays affecting pull request updates, webhook deliveries, and audit log events. The event streaming platform is showing signs of recovery, and we are monitoring as the impact continues.
  5. 16:17 UTC · investigating
    Audit log events and Actions workflow status updates continue to be delayed. Pull request updates and webhook deliveries are recovering as we restore event processing and work through backlogs. We’re monitoring the recovery and will share further updates as we learn more.
  6. 16:25 UTC · investigating
    Audit log events are delayed by up to 100 minutes, and Actions workflow status updates continue to experience extended delays. Pull request updates and webhook deliveries are recovering as we restore event processing and work through backlogs. We’re monitoring recovery and will share further updates as we learn more.
  7. 16:33 UTC · investigating
    Pull Requests is operating normally.
  8. 16:34 UTC · investigating
    Audit log events continue to be delayed by up to 110 minutes. Pull request updates have recovered. Actions is recovering, and users should no longer experience impact in Actions. Webhook deliveries continue to recover as we work through backlogs. We are monitoring recovery and will share further updates as we learn more.
  9. 16:57 UTC · investigating
    Actions and Webhooks are operating normally.
  10. 16:59 UTC · investigating
    Actions and webhook deliveries have recovered. Audit log events are still delayed by about 100 minutes. We are working through the backlog, and full recovery will take some time. We will share more updates as recovery continues.
  11. 17:09 UTC · investigating
    Pull request updates, Actions, and webhook deliveries have recovered. Audit log events are recovering and delays are decreasing, but we expect the remaining backlog to take approximately two to three hours to clear. We’re monitoring recovery and will share further updates.
  12. 18:18 UTC · investigating
    Audit log events are recovering, with delays now below 90 minutes as the remaining backlog is processed. We’ve reduced load on the event-streaming platform and are monitoring recovery. Further updates will follow.
  13. 18:52 UTC · investigating
    Audit log events continue to recover, with delays now below 70 minutes. We’ve applied mitigations to speed recovery and expect audit logs to fully recover in about 40 minutes. Pull request updates, Actions, and webhook deliveries remain recovered. We are monitoring progress and will share further updates.
  14. 19:22 UTC · investigating
    Pull request updates, Actions, and webhook deliveries have recovered. Audit log delays are now below 40 minutes, within the healthy threshold, and continue to decrease as the remaining backlog is processed.
  15. 19:22 UTC · resolved
    This incident has been resolved. Thank you for your patience and understanding as we addressed this issue. A detailed root cause analysis will be shared as soon as it is available.
Source
Oct 19, 20261h 08mSEV 2
Degraded availability for GitHub Actions, GitHub Pages and WebhooksGitHub Actions, Pages and Actions-generated webhooks were degraded from about 15:24 to 16:52 UTC on October 19, 2026. About 5% of GitHub-hosted runner jobs started late and 0.02% of runs failed, while 8.5% of Actions jobs API requests errored in the first 30 minutes. No data was lost. Cause was an internal Actions service overloading a shared network link after simultaneous reconnect retries; a connection limit is now enforced.AI-graded · some customers · Webhooks +2
Why this grade
  1. RULES · SEV 1
    Sev 1 because the provider reported data loss or exposure
  2. CLAUDE · SEV 2 · SUBSET · HIGH CONFIDENCE
    Actions job starts were delayed for about 5% of runners and 0.02% of runs failed; Pages deploys and Actions webhooks were delayed for about 90 minutes.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
6 updates
  1. 15:49 UTC · investigating
    We are investigating reports of degraded performance for Actions and Pages
  2. 15:50 UTC · investigating
    Webhooks is experiencing degraded performance. We are continuing to investigate.
  3. 15:56 UTC · investigating
    We’re are currently investigating degraded availability affecting GitHub Actions, GitHub Pages and Webhooks. <br /><br />Customers may experience delays in workflow processing, site builds and webhook delivery.<br />
  4. 16:30 UTC · investigating
    We’re seeing recovery following mitigation and are continuing to monitor. <br /><br />Customers may still experience delays in GitHub Actions workflow status updates, GitHub Pages deployments, and check and workflow webhook events generated by Actions runs.
  5. 16:56 UTC · monitoring
    Actions, Pages and Webhooks are operating normally.
  6. 16:56 UTC · resolved
    On October 19, 2026, between 15:24 and 16:52 UTC, GitHub Actions, GitHub Pages, and webhooks generated by Actions were degraded. Live updates on GitHub.com and code search were also degraded for part of this time. No data was lost. <br /><br />Impact <br /><br /> - Workflow status updates, logs, and checks were delayed. <br /> - 5.13% of jobs on GitHub-hosted runners were delayed in starting, and 0.02% of workflow runs failed. <br /> - During the first 30 minutes, 8.47% of requests to the Actions jobs REST API returned errors. <br /><br />Pages deployments and webhooks generated by Actions runs were delayed. <br /><br />An internal Actions service that stores logs and run status lost connections to some of its cache servers in one cloud region and immediately retried them from every server at once. The resulting surge of new connections overloaded a shared network link between our data center and that region, which also disrupted other services that use the same link. We mitigated the incident by moving the Actions service off the affected cache servers and reducing connections from other services on the link. We then processed the delayed Actions work. We are continuing to investigate the cause of the initial connection timeouts. <br /><br />We were already in the process of rolling out a limit on how quickly this service can open new connections, but it was still in shadow mode (logging only, not enforcing) at the time of the incident. We have since fully enforced it and capped the service's concurrent connection attempts. We are extending these limits to other Actions services, adding backoff to retries, limiting how much shared network capacity a single service can use, and improving alerting so we can find the source of connection surges faster.
Source
Oct 15, 202624mNOT COUNTED
CCA Automations are delayed when starting an AutomationGitHub first reported delays when users started CCA Automations and mitigated a degradation. It later resolved the incident, stating investigation found no delay or impact to CCA Automations.AI-graded
Why this grade
  1. RULES · SEV 2
    Sev 2 from the incident title (no components listed)
  2. CLAUDE · SEV 0 · UNCLEAR · MEDIUM CONFIDENCE
    Initially reported as delays starting CCA Automations, but GitHub later found no delay or impact and resolved it as a false alarm.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
4 updates
  1. 18:31 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 18:34 UTC · investigating
    CCA Automations are experiencing delays when users start an Automation
  3. 18:53 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  4. 18:56 UTC · resolved
    We suspected that CCA Automations were experiencing delays when users start an Automation, but further investigation found that there was no delay or impact to CCA Automations.<br />
Source
Oct 14, 20260mSEV 3
[Retroactive] Actions workflow run logs delayedSev 3 from the incident title (no components listed)
1 update
  1. 20:30 UTC · resolved
    On October 13, around 20:42 UTC, an infrastructure failure caused GitHub Actions to have delays showing logs for workflow runs. No logs were lost and the issue has now been resolved and log processing has caught up. The actual workflows producing the logs were not impacted and ran normally and no run delays or check delays occurred.
Source
Oct 12, 20267h 54mSEV 2
Incident with Actions, Pull Requests and IssuesElevated 500 errors hit GitHub Actions, Pull Requests, Issues and Projects from about 11:51 to 12:41 UTC, with runs queuing and PR reviews failing. A notifications pipeline issued an expensive query that exhausted shared database replica capacity. Web and email notifications stayed delayed until 20:16 UTC. The query path was fixed.AI-graded · Issues +2
Why this grade
  1. RULES · SEV 3
    Sev 3 because Pull Requests had degraded performance
  2. CLAUDE · SEV 2 · GLOBAL · MEDIUM CONFIDENCE
    Elevated 500s hit Actions runs, Pull Requests and Issues for about 40 minutes; notifications were delayed for hours.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
13 updates
  1. 12:21 UTC · investigating
    We are investigating reports of degraded performance for Actions, Issues and Pull Requests
  2. 12:33 UTC · investigating
    Users may experience issues loading the Actions tab UI, interacting with the Checks API, queueing new Actions runs, creating and updating issues, and requesting PR reviews, including via REST and GraphQL APIs.
  3. 12:57 UTC · investigating
    We have a mitigation in progress and are monitoring for recovery. Users may also experience delayed web notifications as we mitigate.
  4. 13:02 UTC · investigating
    Pull Requests is operating normally.
  5. 13:38 UTC · investigating
    Our initial mitigation was not fully successful, and users may still experience delays and errors with Actions, issues and projects, and notifications are delayed. We're continuing to investigate the root cause and work on additional mitigations.
  6. 14:35 UTC · investigating
    The degradation affecting Issues has been mitigated. We are monitoring to ensure stability.
  7. 14:36 UTC · investigating
    The degradation affecting Actions has been mitigated. We are monitoring to ensure stability.
  8. 14:37 UTC · investigating
    We have mitigated the issues with Actions, issues, and projects, and have a mitigation in progress to address the delay with notifications.
  9. 15:19 UTC · investigating
    We have successfully applied a mitigation for notifications, and notifications have resumed but will continue to be delayed as we work through a backlog.
  10. 15:50 UTC · investigating
    Notifications have resumed but will continue to be delayed as we work through a backlog.
  11. 17:34 UTC · investigating
    Web and email notifications continue to be delayed as we work through a backlog of notifications.
  12. 19:28 UTC · investigating
    Web and email notifications will be delayed for 1 more hour as we are close to completing the backlog of notifications.
  13. 20:16 UTC · resolved
    On October 12th 2026 11:51 UTC GitHub experienced elevated 500 responses across several application pages. This caused failures on Issues and Pull Requests as well as Projects and Actions. Customers also experienced delayed notifications in their web inbox and via email. <br /><br />The immediate impact to Issues, Pull Requests, Projects, and Actions was mitigated by 12:41 UTC on October 12th. Notifications deliveries to the web inbox and via email were recovering at 15:27 UTC and fully mitigated by 20:16 UTC. <br /><br />The incident was triggered through a latent condition in our notifications delivery pipeline that put a high volume of an expensive query on our sharded database replicas. This reduced available database capacity to serve requests for Issues, Pull Requests, Projects, and Actions.<br /><br />We have fixed the notifications pipeline’s code path that issued the expensive query. We are also improving resilience of the database layer to reduce the likelihood and impact of similar incidents.
Source
Oct 7, 202648mSEV 2
Incident with Git Operations, Issues, Actions and Pull RequestsGitHub had widespread degradation of Git operations, Actions, pull requests and issues from 16:52 to 17:01 UTC, a recurrence of an earlier outage. Mitigations were applied and all systems recovered. The shared root cause is described in the linked combined summary of both incidents.AI-graded · Git Operations +3
Why this grade
  1. RULES · SEV 2
    Sev 2 because Git Operations had a major outage
  2. CLAUDE · SEV 2 · GLOBAL · MEDIUM CONFIDENCE
    Git operations, Actions, pull requests and issues degraded across GitHub for about nine minutes during a write-operation outage.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
5 updates
  1. 17:17 UTC · investigating
    We are investigating reports of degraded performance for Actions, Git Operations, Issues and Pull Requests
  2. 17:25 UTC · investigating
    We observed widespread impact across GitHub services from 16:52 UTC to 17:01 UTC, and are investigating the root cause. All systems have recovered. This incident is a reoccurrence of the incident from earlier in the day https://www.githubstatus.com/incidents/djlmxz2zd0j7. Durable mitigation is in progress. We will provide another update in 30 minutes.
  3. 17:27 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  4. 18:00 UTC · monitoring
    All systems have recovered. We have applied mitigations and do not anticipate another recurrence.
  5. 18:04 UTC · resolved
    On October 7, 2026, from 16:52 to 17:01 UTC, GitHub experienced a recurrence of the write-operation outage from earlier that day. Both outages shared the same root cause. Please refer to the combined summary covering both incidents, their impact, and our prevention measures: https://www.githubstatus.com/incidents/djlmxz2zd0j7
Source
Oct 7, 20260mSEV 1
Incident with Git Operations, Pull Requests and ActionsSev 1 because the provider reported data loss or exposureGit Operations +4
20 updates
  1. 15:14 UTC · investigating
    We are investigating reports of degraded availability for Actions, Git Operations and Pull Requests
  2. 15:14 UTC · investigating
    Webhooks is experiencing degraded performance. We are continuing to investigate.
  3. 15:15 UTC · investigating
    Issues is experiencing degraded performance. We are continuing to investigate.
  4. 15:16 UTC · investigating
    Webhooks is experiencing degraded availability. We are continuing to investigate.
  5. 15:21 UTC · investigating
    Git Operations is experiencing degraded performance. We are continuing to investigate.
  6. 15:24 UTC · investigating
    We observed widespread impact across GitHub services between 15:06 UTC and 15:16 UTC, and are investigating the root cause. We are seeing all systems recover, and will provide another update shortly.
  7. 15:25 UTC · investigating
    Pull Requests is experiencing degraded performance. We are continuing to investigate.
  8. 15:27 UTC · investigating
    Git Operations is operating normally.
  9. 15:27 UTC · investigating
    Webhooks is experiencing degraded performance. We are continuing to investigate.
  10. 15:31 UTC · investigating
    The degradation affecting Pull Requests has been mitigated. We are monitoring to ensure stability.
  11. 15:32 UTC · investigating
    Actions is operating normally.
  12. 15:35 UTC · investigating
    Git Operations is experiencing degraded performance. We are continuing to investigate.
  13. 15:41 UTC · investigating
    We are observing partial recovery across all systems. We are continuing to monitor and investigate the root cause of the disruption.
  14. 15:47 UTC · investigating
    The degradation affecting Issues has been mitigated. We are monitoring to ensure stability.
  15. 15:49 UTC · investigating
    The degradation affecting Git Operations and Webhooks has been mitigated. We are monitoring to ensure stability.
  16. 15:49 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  17. 15:56 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  18. 15:58 UTC · monitoring
    We are seeing full recovery across all systems including Git Operations, Pull Requests and Actions. We are continuing to investigate the root cause of the disruption between 15:06 UTC and 15:16 UTC, and will share another update shortly.
  19. 16:23 UTC · monitoring
    Git Operations, Pull Requests and Actions are fully recovered and stable. The root cause has been identified and mitigations are being applied to prevent a reoccurrence.
  20. 16:25 UTC · resolved
    On October 7, 2026, GitHub had two outages of write operations with the same cause, totaling about 19 minutes. The first ran from 15:06 to 15:16 UTC (about 10 minutes) and the second from 16:52 to 17:01 UTC (about 9 minutes). In both windows, actions that save or change data failed across Git operations, pull requests, issues, GitHub Actions, the API, webhooks, sign-in and authentication, GitHub Apps and OAuth, code scanning, Codespaces, and Copilot. Reading data was largely unaffected. In the first outage, 5.6% of all web and API requests failed, peaking at 8.9%, and more than 12,000 organizations saw failed requests. Up to 26% of Git authentication requests failed, and 17% of Actions workflow runs failed with infrastructure errors. In the second outage, 6.2% of web and API requests failed, peaking at 9.3%, and more than 11,000 organizations were affected. Up to 25% of Git authentication requests failed, and 19% of Actions workflow runs failed. No databases experienced data loss, and operations retried after recovery succeeded. GitHub relies on an internal service that keeps track of which database servers should receive new and updated data. While replacing machines that run this service as part of regular maintenance, one of the new machines started with an outdated copy of its records, which still listed database servers we had already retired. When this node came online, it incorrectly modified routing configuration with the outdated records, routing write requests to non-existent database servers. Each outage ended when one of our established machines with correct configuration became the leader of the quorum and sent the correct configuration. After the first outage, we identified the new machine as the source and paused automated maintenance to stop it from being leader again. At 16:52 UTC, an automated maintenance task that did not honor the pause put it back in charge, and the outage happened again. We shut that machine down at 17:02 UTC and have since removed it, and the automation now honors the pause, which was a new regression due to an update of our puppet infrastructure. To prevent recurrence, we are clearing retired database servers from the service's records before any new machine is added, and we are keeping new machines from being elected leader until their records have been checked. We are also changing our systems to reject configuration that point to servers that don't exist and to keep using the last known-good settings instead. Finally, we are giving responders a faster way to take a misbehaving machine out of service, and adding alerts so changes like this are spotted within seconds. We will share a more detailed write-up of these incidents in our monthly availability report, along with the status of the repairs we are making to prevent issues like these in the future.
Source
Oct 6, 20263h 43mSEV 2
Several services are degradedA database connection pool saturated by long GitHub Enterprise Importer queries caused elevated error rates, about 1.30% of requests, across 12 services from 19:44 to 19:50 UTC. Recovery came when the migration calls stopped. Enterprise migrations stayed paused until a code fix merged and they resumed at 23:32 UTC.AI-graded · Webhooks +1
Why this grade
  1. RULES · SEV 2
    Sev 2 because Webhooks had a major outage
  2. CLAUDE · SEV 2 · GLOBAL · MEDIUM CONFIDENCE
    Pull requests, webhooks, OAuth, Git over SSH and other services saw elevated errors for about six minutes; GEI migrations stayed paused until 23:32 UTC.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
9 updates
  1. 19:57 UTC · investigating
    We are investigating reports of degraded performance for Pull Requests and Webhooks
  2. 20:04 UTC · investigating
    Webhooks is operating normally.
  3. 20:12 UTC · investigating
    Pull Requests is operating normally.
  4. 20:22 UTC · investigating
    A number of services including issues, pull requests, and webhooks, were briefly degraded, but have now recovered. Enterprise migrations are paused while we validate causes and mitigations.
  5. 20:26 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  6. 20:57 UTC · investigating
    All services are functioning normally except for enterprise migrations, which are still paused. They will resume soon.
  7. 21:29 UTC · investigating
    We are working towards a mitigation that will allow us to resume processing enterprise migrations.
  8. 22:40 UTC · investigating
    We are deploying a mitigation and plan to resume migrations shortly thereafter
  9. 23:40 UTC · resolved
    On October 6, 2026, from 19:44 to 19:50 UTC, GitHub experienced a service outage with elevated error rates peaking at 1.30% of all requests, totaling approximately 1.16 million failed requests. This impacted 12 services across GitHub, including Pull Requests, Webhooks, Repos, OAuth, Git over SSH, Codespaces, and Gists. The incident was triggered by GitHub Enterprise Importer (GEI) making multiple, long duration queries which saturated a database connection pool. Recovery was seen at approximately 19:50 UTC once the migration's calls stopped; GEI migrations were reenabled at 23:32 UTC after a code fix was merged.<br /><br />We have completed a detailed root-cause review and are implementing improved database-health safeguards. We will continue to review and add additional safeguards to prevent future recurrence.
Source
Oct 5, 20261h 45mSEV 3
Disruption with some GitHub servicesA networking failure between one on-premises datacenter region and some cloud-hosted regional database services caused slow pages, timeouts and errors on organization and enterprise billing, licensing and settings pages from 23:05 to 00:11 UTC. Average error rate was 1.9%, peaking at 13.8%. Traffic was shifted to a healthy region to mitigate it.AI-graded · some customers
Why this grade
  1. RULES · SEV 3
    Sev 3 from the incident title (no components listed)
  2. CLAUDE · SEV 3 · SUBSET · HIGH CONFIDENCE
    Organization and enterprise billing, licensing and settings pages showed slow loads and errors for about an hour; average error rate 1.9%.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
3 updates
  1. 23:47 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 00:05 UTC · investigating
    Users may experience issues accessing organization and enterprise billing and licensing pages. We are deploying a mitigation and monitoring for recovery.
  3. 01:32 UTC · resolved
    On October 5, 2026, between 23:05 UTC and 00:11 UTC on October 6, some users experienced slow page loads, timeouts, and errors when accessing organization and enterprise billing, licensing, and settings pages. On average, the error rate was 1.9% and peaked at 13.8% of requests to the service that powers these pages.<br /><br />This was caused by a networking failure, which disrupted connectivity between one of our on-premises datacenter regions and a subset of cloud-hosted regional database services. We mitigated the incident by shifting the database traffic to a healthy region.<br /><br />We are working to make regional database failover faster and more complete, reduce timeouts so these pages fail fast instead of hanging, and add latency-based alerting to reduce our time to detection and mitigation of issues like this one in the future. We continue to partner with our cloud provider to understand the underlying network failure.
Source
Oct 5, 20263h 38mSEV 2
Incident with ActionsGitHub-hosted Actions runners failed or delayed jobs from about 18:48 to 22:49 UTC on October 5, 2026. At peak 47% of runs failed. A network failure between an on-premises region and cloud database and storage services caused it. Traffic was shifted and capacity added. Some repository list, billing and licensing pages also had intermittent errors. Self-hosted runners were not affected.AI-graded · some customers · Actions +1
Why this grade
  1. RULES · SEV 1
    Sev 1 because Pages had a major outage
  2. CLAUDE · SEV 2 · SUBSET · HIGH CONFIDENCE
    GitHub-hosted Actions runs failed or stalled for about four hours; self-hosted runners and existing Pages sites were unaffected.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
12 updates
  1. 19:11 UTC · investigating
    We are investigating reports of degraded performance for Actions
  2. 19:15 UTC · investigating
    We’re investigating an issue causing delays when assigning GitHub-hosted runners to Actions jobs. Some workflows may take longer to start across runner configurations. Our teams are working to mitigate the issue, and we’ll provide an update as we learn more.
  3. 19:50 UTC · investigating
    We’re still investigating delays in assigning GitHub-hosted runners, affecting workflow start times across multiple runner configurations. Our teams are working to mitigate the impact; we’ll share another update as we learn more.
  4. 20:39 UTC · investigating
    We’re continuing to investigate job failures and delays affecting GitHub-hosted runner assignment and workflow start times. Our teams are actively working to mitigate the impact and will provide another update as we learn more.
  5. 20:47 UTC · investigating
    Actions is experiencing degraded availability. We are continuing to investigate.
  6. 21:09 UTC · investigating
    In addition to ongoing issues with Actions and Hosted Runners, some customers are also unable to access repository list, licensing and billing pages within their GitHub accounts. We continue to investigate and are working on mitigation.
  7. 21:22 UTC · investigating
    Pages is experiencing degraded performance. We are continuing to investigate.
  8. 21:31 UTC · investigating
    Actions is experiencing degraded performance. We are continuing to investigate.
  9. 21:32 UTC · investigating
    We have applied mitigations to address GitHub Actions failures. Queued jobs are clearing, new jobs are not delayed, and we expect full runner recovery shortly. We continue to monitor for full recovery and are working to mitigate ongoing issues affecting other services, including access to repository lists, licensing, and billing pages.
  10. 21:54 UTC · investigating
    Actions is operating normally.
  11. 22:40 UTC · investigating
    Pages is operating normally.
  12. 22:49 UTC · resolved
    On October 5, 2026, between 18:48 and 22:49 UTC, GitHub Actions and Hosted Runners experienced degraded performance. Over the full incident, 14.3% of workflow runs using GitHub-hosted runners and 26.5% of individual GitHub-hosted jobs did not start within five minutes. <br /><br /> <br />During the roughly 90-minute period of highest impact, 47.0% of workflow runs using GitHub-hosted runners failed and 71.9% of GitHub-hosted jobs did not start within five minutes, while existing Pages sites continued to serve normally. A portion of customers also encountered intermittent errors in repository lists, licensing and billing pages, and some Copilot features. Self-hosted runners were not affected. <br /><br /><br />The incident was caused by a partial networking failure that disrupted connectivity between one of our on-premises datacenter regions and a subset of cloud-hosted regional database and storage services. <br /><br /><br />We mitigated the incident by shifting database and storage traffic to healthy regions and endpoints and increasing capacity for affected compute services. GitHub Actions and Hosted Runners recovered by 21:54 UTC, and all affected services recovered by 22:49 UTC. <br /><br /> <br />We are partnering with our cloud provider to understand the network failure and identify improvements to network-path resilience, detection, and recovery for similar incidents.
Source
Oct 1, 20263h 09mSEV 3
Actions Job DelaysGitHub-hosted Actions runners had delayed or failed job starts on Oct 1, 2026, from about 14:15 to 17:56 UTC. About 0.68% of runs were delayed and 0.196% of jobs failed to start. A cloud provider routing change caused rate-limit errors, and retry behavior added load. Runner subscriptions were excluded from the inventory service and capacity was scaled, and the provider raised rate limits.AI-graded · some customers · Actions
Why this grade
  1. RULES · SEV 2
    Sev 3 because Actions had degraded performance, raised to Sev 2 for security
  2. CLAUDE · SEV 3 · SUBSET · HIGH CONFIDENCE
    Job starts on GitHub-hosted runners were delayed or failed for about 0.2% of jobs and 0.68% of runs over several hours; most of Actions worked.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
10 updates
  1. 14:47 UTC · investigating
    We are investigating reports of degraded performance for Actions
  2. 14:54 UTC · investigating
    We have identified the cause of increased Actions run start delays on Ubuntu runners and are actively deploying a fix. Customers may continue to experience intermittent delays while the mitigation rolls out and service metrics return to normal.
  3. 15:02 UTC · investigating
    Run-start delays on Ubuntu runners have been resolved. We are investigating run-start delays on Windows runners and will share more information as it becomes available.
  4. 15:10 UTC · monitoring
    The degradation affecting Actions has been mitigated. We are monitoring to ensure stability.
  5. 15:20 UTC · investigating
    Actions is experiencing degraded performance. We are continuing to investigate.
  6. 15:29 UTC · investigating
    We are seeing a reoccurrence in run-start delays, and are continuing to investigate to issue. Customers will potentially experience delays of up to ten minutes.
  7. 16:10 UTC · investigating
    We have identified an issue with our upstream provider which is causing Actions requests to 429 which is creating the delays. We have escalated to the owning team and are investigating how to mitigate the 429s.
  8. 16:48 UTC · investigating
    We are currently applying a mitigation and anticipate recovery within thirty minutes.
  9. 17:50 UTC · investigating
    GitHub Actions experienced degraded performance for some hosted runners due to throttling within an upstream Azure dependency. Service capacity has recovered, and we are continuing to monitor while working with Azure on the underlying condition.
  10. 17:56 UTC · resolved
    On October 1, 2026, GitHub Actions experienced two periods of delayed and failed job starts on GitHub-hosted runners, beginning at approximately 14:15 UTC. Across the 14:15–17:56 UTC incident-response window, approximately 0.68% of workflow runs experienced start delays, and 0.196% of jobs failed to start. <br /><br />A routing change at our cloud provider concentrated API traffic on a single backend, causing an increase in rate-limit errors. Incorrect retry behavior in the provider’s inventory service added load to the affected APIs, compounding the rate limiting. This disrupted runner provisioning and reduced available capacity in affected regions. The resulting increase in demand in unaffected regions exposed a concurrency limit in our job-assignment system, delaying jobs even where runners were available. <br /><br />During the response, we increased job-assignment concurrency, scaled runner capacity, and worked with our cloud provider to reduce API load by excluding our runner subscriptions from its inventory service. By approximately 16:15 UTC, jobs were starting and executing as expected. We continued mitigation and monitoring until the incident was resolved at 17:56 UTC. <br /><br />Our cloud provider has since increased resource-management API rate limits for our workloads. To reduce the likelihood and impact of similar incidents, we are also adding more granular rate-limiting alerts and updating our job-assignment system to handle increased demand across fewer available regions.
Source
Oct 1, 202620mSEV 4
Elevated request latencyGitHub.com experienced two periods of slow page loads and intermittent request failures from 13:04 to 13:34 UTC on October 1, 2026. The error rate averaged 0.52% and peaked at 6.13%. An unusually high volume of incoming traffic delayed requests before they reached application servers. Performance recovered as traffic subsided, and additional traffic filtering was applied.AI-graded
Why this grade
  1. RULES · SEV 2
    Sev 2 from the incident title (no components listed)
  2. CLAUDE · SEV 4 · GLOBAL · MEDIUM CONFIDENCE
    GitHub.com web requests were slow and intermittently failed for about 30 minutes, averaging a 0.52% error rate and peaking at 6.13%.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
4 updates
  1. 13:37 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 13:39 UTC · investigating
    We are investigating recurrent periods of elevated latency affecting web requests. We’ll share updates as more information becomes available.
  3. 13:51 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  4. 13:57 UTC · resolved
    On October 1, 2026, between 13:04 UTC and 13:34 UTC, users experienced two periods of slow page loads and intermittent request failures on GitHub.com. On average, the error rate was 0.52% and peaked at 6.13%. An unusually high volume of incoming traffic placed additional load on our traffic-handling infrastructure, delaying requests before they reached our application servers. Performance returned to normal as the traffic subsided, and we have since applied additional traffic filtering to protect against similar traffic. We are improving alerting for delays in our traffic-handling infrastructure and strengthening our automated traffic protections to reduce our time to detection and mitigation.
Source
Oct 1, 20260mSEV 3
[Retroactive] Actions workflow run failures after deployment gate approvalsSev 3 from the incident title (no components listed)
1 update
  1. 02:00 UTC · resolved
    On October 1, around 02:00 UTC, an isolated infrastructure failure caused GitHub Actions to lose execution state for a small number of existing workflow runs. Affected runs may remain stuck, fail deployment approvals, or return errors when cancelled. Service connectivity has recovered, but the lost state cannot be restored by retrying an approval. If you're affected, you can trigger a new run or contact GitHub Support with links to your stuck runs so we can unblock them. Once cleared, select "Re-run all jobs." This preserves the workflow run ID but starts a new attempt, rebuilds artifacts, and requires fresh deployment approvals. Before rerunning, check whether any deployment steps already completed to avoid repeating changes.
Source
Sep 28, 202652mSEV 4
Copilot Code Review is unable to complete reviewsSev 4 from the incident title (no components listed)
4 updates
  1. 21:16 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 21:23 UTC · investigating
    Revert of the impacted change is in flight, expect full recovery once the deployment is done.
  3. 22:08 UTC · investigating
    Copilot Code review has recovered, for reviews that were previously not successful you can re-request review from your PR.
  4. 22:08 UTC · resolved
    On September 28, 2026, between 19:23 UTC and 22:08 UTC the Copilot Code Review service was degraded and pull request reviews did not complete in all environments. On average, approximately 70% of requested reviews did not complete. Reviews requested during this window were not automatically retried, and users needed to re-request a review from their pull request. This was due to a change in a dependent service that stopped delivering data Copilot Code Review needs to complete a review. Review completion is an inherently delayed signal, because each review job takes time to start and finish, so it took longer to detect the impact.<br /><br />We mitigated the incident by reverting the change.<br /><br />We are working to add a faster signal for reviews that fail to complete, add deployment validation that confirms a Copilot Code Review completes end to end, improve how we track internal dependencies between services, and make our deployment pipeline more reliable to reduce our time to detection and mitigation of issues like this one in the future.
Source
Sep 24, 20263h 50mSEV 3
Disruption with billing information updatesUsers could not reliably add or update billing address information because requests to the address validation service stalled or timed out. About 1,700 users were affected from September 22 to September 24, 2026. An OS upgrade exposed an HTTP client adapter incompatibility; the fix was switching the HTTP client used for address validation, and alerting was added.AI-graded · some customers
Why this grade
  1. RULES · SEV 2
    Sev 3 from the incident title (no components listed), raised to Sev 2 for security
  2. CLAUDE · SEV 3 · SUBSET · HIGH CONFIDENCE
    Billing address updates failed for about 1,700 users, at up to 100% of requests in the flow, for roughly two days.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
7 updates
  1. 16:51 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 16:51 UTC · investigating
    We are investigating failures on billing information updates. Customers may be unable to create or update their billing information in the meantime.
  3. 17:40 UTC · investigating
    We have identified the problem and are actively working on mitigation.
  4. 18:22 UTC · investigating
    We are continuing to work on mitigation. We will post another update in approximately one hour.
  5. 20:10 UTC · investigating
    We have applied the mitigation and are seeing signs of recovery. We are continuing to monitor the system closely.
  6. 20:24 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  7. 20:41 UTC · resolved
    Beginning on September 22, 2026 at 11:51 UTC, approximately 1,700 users experienced delays and errors when adding or updating billing address information because requests to the address validation service timed out or failed. Within the affected address validation flow, requests failed at an average rate of 69.3%, reaching 100% at peak. <br /><br />An operating system upgrade exposed a compatibility issue in an adapter used by our HTTP client library, causing requests to stall. A subsequent change to shorten request timeouts caused stalled requests to return connection errors and increased the failure rate. <br /><br />We mitigated the incident on September 24, 2026 at 20:24 UTC by changing the HTTP client used for address validation. We have added alerting to detect similar issues sooner and are extending the fix to other integrations.
Source
Sep 23, 202618h 44mSEV 2
Incident across several servicesA failed planned database maintenance triggered a failover that left replicas in Azure Central US unable to resume replication. GitHub API requests, org creation, GitHub App installs and membership changes failed or degraded from 07:57 UTC on September 23 until errors were mitigated at 10:58 UTC. Projects label updates and search stayed delayed until full service returned at 04:55 UTC on September 24.AI-graded · regional · API Requests
Why this grade
  1. RULES · SEV 3
    Sev 3 because API Requests had degraded performance
  2. CLAUDE · SEV 2 · REGIONAL · HIGH CONFIDENCE
    Elevated 500 and 404 errors hit the API and blocked org creation, GitHub App installs and membership changes in Azure Central US for about three hours; Projects label updates lagged until the next day.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
16 updates
  1. 10:11 UTC · investigating
    We are investigating reports of degraded performance for API Requests
  2. 10:20 UTC · investigating
    Database replicas have detached. We're working to restore the database replicas. Users may experience issues beyond creating organizations and a degraded experience with the GitHub API and Projects.
  3. 10:57 UTC · investigating
    Database replicas have been restored. Org creation and the GitHub API are no longer degraded.
  4. 10:58 UTC · investigating
    The degradation affecting API Requests has been mitigated. We are monitoring to ensure stability.
  5. 11:38 UTC · investigating
    We are seeing recovery for Projects. Users may experience stale search results for Projects while indexing catches up.
  6. 13:35 UTC · investigating
    Users may experience stale Project search results. We are working to increase indexing speed. All other services are available.
  7. 17:01 UTC · investigating
    Updates to issue labels may be delayed in being reflected in Projects by about ~10 minutes. We have added some capacity to work through the backlog more quickly, but it'll likely be a few hours to complete processing the full backlog of messages. All other services are available.
  8. 17:30 UTC · investigating
    We will post another update in approximately one hour to share our progress.
  9. 18:42 UTC · investigating
    Continuing to investigate the lag that may be experienced in issue labels being accurately reflected in Projects. We are working on alternate solutions to process the backlog of label updates.
  10. 20:26 UTC · investigating
    We are preparing to deploy a change that will mitigate the impact.
  11. 21:39 UTC · investigating
    Updates to issue labels may be delayed in being reflected in Projects. We are continuing to deploy a change that will accelerate processing of the backlog of label updates. All other services are available. We will provide another update within the next hour.
  12. 00:16 UTC · investigating
    We've deployed a change intended to accelerate processing of the backlog of issue label updates in Projects. A sizable backlog still remains and we continue working through it. All other services are operating normally. We will provide another update within the next hour.
  13. 02:00 UTC · investigating
    We are continuing to process the backlog of issue label updates for Projects. Users may still see delays before label changes are reflected in Projects. All other services are operating normally.
  14. 03:14 UTC · investigating
    We are continuing to process the backlog of issue label updates for Projects. Label changes may still be delayed. All other services are operating normally.
  15. 04:55 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  16. 04:55 UTC · resolved
    Starting at 07:57 UTC on September 23, GitHub experienced elevated 500 and 404 responses across several application pages. This caused failures when installing GitHub Apps, creating organizations, and making some organization membership changes. Customers also experienced delayed label updates and stale search results in Projects. <br /><br />The infrastructure failure was isolated to our Azure Central US region. The API errors were mitigated by 10:58 UTC on September 23. Projects' processing continued to recover while an accumulated backlog was drained, and full service was restored at 04:55 UTC on September 24. <br /><br />The incident was caused by a failed planned maintenance operation on a primary database. Automated recovery initiated an emergency database failover, after which several replicas in the affected region were unable to resume replication correctly. This reduced available database capacity and caused the API errors and downstream Projects processing delays. <br /><br />We have mitigated the immediate failure mode. We are also improving maintenance safety checks, database failover handling, post-failover replica validation, and downstream processing resilience to reduce the likelihood and impact of similar incidents.
Source
Sep 20, 20261h 09mSEV 3
Incident with Pull RequestsSev 3 because Pull Requests had degraded performancePull Requests
4 updates
  1. 22:13 UTC · investigating
    We are investigating reports of degraded performance for Pull Requests
  2. 22:27 UTC · monitoring
    The degradation affecting Pull Requests has been mitigated. We are monitoring to ensure stability.
  3. 22:32 UTC · monitoring
    A git fileserver issue caused a brief delay in creating some merge commits - we've isolated the underlying server and already observed recovery.
  4. 23:22 UTC · resolved
    On September 20, 2026, between 21:46 and 22:24 UTC the Pull Requests service was degraded and pull request merge and test-merge commits were created late, with delays reaching approximately four minutes at peak. Merge commits were delayed rather than lost. Because some Actions workflow runs start only after a pull request's merge commit is created, a subset of workflow runs for pull request events were also delayed. <br /><br />This was due to a routine repository maintenance job for an unusually large repository consuming nearly all of the memory on a single Git storage server, which left that server unable to serve the Git operations used to create merge commits. <br /><br />We mitigated the incident by removing the affected server from service at 22:18 UTC, after which the queued merge commits were created within six minutes. <br /><br />We have capped the memory a single repository maintenance job may consume so that one repository cannot exhaust a server, and we have improved monitoring and alerting on storage server health to reduce our time to detection and mitigation of issues like this one in the future.
Source
Sep 17, 202650mSEV 4
Elevated rate of errors for OpenAI models provided by CopilotSev 4 because Copilot AI Model Providers had degraded performanceCopilot AI Model Providers
3 updates
  1. 20:59 UTC · investigating
    We are investigating reports of degraded performance for Copilot AI Model Providers
  2. 21:39 UTC · monitoring
    We are experiencing degraded availability for GPT-5.6 Luna, GPT-5.6 Terra, GPT-5.6 Sol, GPT-6 Astra, GPT-5.3-Codex in Copilot products and IDE surfaces. This is due to an issue with an upstream model provider. The provider is working to mitigate the problem and we are monitoring recovery. We recommend choosing another model or selecting 'Auto' to continue using Copilot.
  3. 21:49 UTC · resolved
    Between 20:26 and 21:17 UTC on September 17, 2026, GitHub Copilot experienced degradation affecting several GPT models, including GPT-5.6 Luna, GPT-5.6 Terra, GPT-5.6 Sol, GPT-5.3-Codex, and GPT-6 Astra. Users encountered elevated error rates when using these models.<br /><br />The degradation was caused by an issue with an upstream model provider. GitHub engineers detected the issue through automated monitoring and coordinated with the provider. Our automated model-warning system activated in-product warnings for affected models during the incident. Service returned to normal after the provider implemented a mitigation.
Source
Sep 16, 202610h 28mSEV 2
Degradation with Gemini 3.8 FlashRequests to the Gemini 3.8 Flash model in GitHub Copilot failed at an average rate of 6.4% from about 04:40 to 11:45 UTC on September 16, 2026, worst at peak traffic. Other Copilot models were not affected. The cause was a capacity issue at an upstream model provider; failure rates returned to normal as peak traffic passed, and the incident was resolved at 17:48 UTC.AI-graded · one model or feature · Copilot AI Model Providers
Why this grade
  1. RULES · SEV 4
    Sev 4 because Copilot AI Model Providers had a partial outage
  2. CLAUDE · SEV 2 · SINGLE · HIGH CONFIDENCE
    Gemini 3.8 Flash in Copilot failed about 6.4% of requests for several hours; other Copilot models were unaffected.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
5 updates
  1. 07:21 UTC · investigating
    We are investigating reports of degraded availability for Copilot AI Model Providers
  2. 08:28 UTC · investigating
    Copilot AI Model Providers is experiencing degraded performance. We are continuing to investigate.
  3. 10:58 UTC · investigating
    Copilot AI Model Providers is experiencing degraded performance. We are continuing to investigate.
  4. 11:45 UTC · monitoring
    The degradation affecting Copilot AI Model Providers has been mitigated. We are monitoring to ensure stability.
  5. 17:48 UTC · resolved
    On September 16, 2026, between 04:40 and 11:45 UTC, the Gemini 3.8 Flash model in GitHub Copilot experienced degraded availability. Requests to this model failed at an average rate of 6.4%, and the impact was highest during peak traffic hours. Other Copilot models were not affected. Users could continue to work with a different model or with 'Auto'.<br /><br />The degradation was caused due to a capacity issue with an upstream model provider. Failure rates returned to normal as the daily traffic peak passed. We monitored the model until it was healthy and resolved the incident at 17:48 UTC. We are working to make our systems resilient to cover peak demand for all the Copilot models.
Source
Sep 15, 202649mSEV 3
Disruption with some GitHub servicesSome Copilot code reviews on pull requests failed to complete from about 15:30 to 20:00 UTC on September 15, 2026. Increased latency in an internal caching service caused lock acquisition timeouts that interrupted review jobs. GitHub reverted the change to that service and restored normal operation by 20:00 UTC.AI-graded · some customers
Why this grade
  1. RULES · SEV 3
    Sev 3 from the incident title (no components listed)
  2. CLAUDE · SEV 3 · SUBSET · MEDIUM CONFIDENCE
    Some Copilot code reviews on pull requests failed to complete for about four and a half hours; other GitHub services were reported fine.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
3 updates
  1. 19:11 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 19:48 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  3. 20:00 UTC · resolved
    On September 15, 2026 between 15:30 and 20:00 UTC, some Copilot code reviews on pull requests failed to complete. The cause was increased latency in an internal caching service that GitHub Copilot Code Review relies on to coordinate its review jobs. This caused a timeout in lock acquisition, which interrupted the job. We reverted the change to the internal caching service and restored normal operation by 20:00 UTC.<br /><br />We sincerely apologize for the disruption.
Source
Sep 15, 20261h 30mSEV 4
Disruption with some GitHub servicesSev 4 because Copilot AI Model Providers had degraded performanceCopilot AI Model Providers
7 updates
  1. 09:47 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 09:48 UTC · monitoring
    We are investigating degraded availability for Claude Fable 5.1, affecting some Copilot products and integrated development environments. The issue is caused by a problem with our upstream model provider. Customers can use another model or select Auto while we investigate.
  3. 10:20 UTC · investigating
    Copilot AI Model Providers is experiencing degraded performance. We are continuing to investigate.
  4. 10:32 UTC · investigating
    We continue to investigate intermittent errors affecting Claude Fable 5.1 in some Copilot products and integrated development environments. Customers can use another model or select Auto while we monitor the situation.
  5. 11:09 UTC · investigating
    We continue to monitor intermittent errors affecting Claude Fable 5.1 in some Copilot products and integrated development environments. Customer-facing metrics have recovered, and we are awaiting confirmation from our upstream provider that the issue will not recur. Customers can select another model or Auto in the meantime.
  6. 11:17 UTC · investigating
    The issues with our upstream model provider have been resolved, and Claude Fable 5.1 is once again available in Copilot products and IDE surfaces.<br /><br />We will continue monitoring to ensure stability, but mitigation is complete.
  7. 11:17 UTC · resolved
    On September 15, 2026, between 05:45 and 09:50 UTC, the Claude Fable 5.1 model in GitHub Copilot experienced intermittently degraded availability, with an average error rate of 2.8%. During brief recurring 15 minute periods that recurred every ~45 minutes, availability for Claude Fable 5.1 dropped to a maximum of ~40% before recovering completely. Other Copilot models were not affected. Users could continue to work with a different model or with 'Auto'.<br /><br />The cause was an issue with an upstream model provider that intermittently rejected requests while overloaded. GitHub worked with the provider, who acknowledged and then resolved the underlying issue at 9:50 UTC, after which the model returned to constant normal availability. Once recovery was guaranteed, we resolved the incident at 11:17 UTC.<br /><br />To reduce the chance of recurrence and customer impact, GitHub is reviewing per-model availability alerting and automatic in-product fallback so that requests to a degraded model can shift to a healthy alternative more quickly.
Source
Sep 14, 202655mSEV 3
Actions Larger Runner Jobs for some customers may be slow to startFor about three hours on September 14, 2026, 5.7% of GitHub Actions larger-runner jobs started slowly due to a provisioning bug in runner capacity selection exposed by a capacity expansion. Provisioning was restored by correcting capacity records, the bug was fixed, and alerts were added.AI-graded · some customers
Why this grade
  1. RULES · SEV 2
    Sev 3 from the incident title (no components listed), raised to Sev 2 for security
  2. CLAUDE · SEV 3 · SUBSET · HIGH CONFIDENCE
    Actions larger-runner jobs waited longer than normal to start for about 5.7% of such jobs over roughly three hours; other runners unaffected.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
3 updates
  1. 18:40 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 19:01 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  3. 19:35 UTC · resolved
    On September 14, 2026, between 16:10 and 19:01 UTC, some customers using GitHub Actions larger runners experienced longer-than-normal wait times for jobs to start. During this period, <b>5.7%</b> of larger-runner jobs were affected. <br /><br />A routine expansion of our compute capacity exposed a bug in how our provisioning system handled capacity records when selecting where to create runner virtual machines. This slowed the creation of new runners, leaving insufficient runner capacity to start affected jobs promptly. <br /><br />We restored normal provisioning by correcting the affected capacity records. We have fixed the underlying capacity-selection bug to prevent this failure from recurring. We have also added alerts for VM-record creation failures associated with this capacity issue.
Source
Sep 13, 20261h 29mSEV 2
Incident with several GitHub ServicesFrom 08:43 to 10:44 UTC, GitHub services degraded across about 28 services. A background cleanup job overloaded a shared database primary, causing failures in token issuance, issue creation, signup and Actions. Peak token failures were 8.8%. Pausing the job and shedding load restored service.AI-graded · API Requests +4
Why this grade
  1. RULES · SEV 1
    Sev 1 because Pages had a major outage
  2. CLAUDE · SEV 2 · GLOBAL · HIGH CONFIDENCE
    Across about 28 services, token creation, issue creation, signup, Actions and Pull Requests degraded for about two hours; Pages had degraded availability.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
6 updates
  1. 09:16 UTC · investigating
    We are investigating reports of degraded availability for API Requests, Issues, Pages and Pull Requests
  2. 09:25 UTC · investigating
    Actions is experiencing degraded performance. We are continuing to investigate.
  3. 09:36 UTC · investigating
    We're seeing increased database replication delays on collab which is causing increased error rates in authorization endpoints and follow-on increased error rates across the system - we are investigating
  4. 10:26 UTC · investigating
    We have reduced load on this cluster with internal load-shedding and are seeing signs of recovery but continue to monitor
  5. 10:28 UTC · investigating
    Pull Requests is experiencing degraded performance. We are continuing to investigate.
  6. 10:44 UTC · resolved
    On September 13, 2026, between 08:43 and 10:44 UTC, GitHub experienced degraded availability across approximately 28 services, including Issues, Pull Requests, Actions, Codespaces, Pages, Notifications, Code Scanning, Git LFS, and new account signup. At peak, 8.8% of requests to create GitHub App installation access tokens failed. Token issuance for Actions workflows was also affected, impacting approximately 4% of workflows during the incident time frame. Creating issues through the web interface failed for about 96% of attempts, and signup failures were above 90%. <br /> <br />The cause was an internal data-cleanup job that began writing to a shared database cluster at 07:33 UTC. That cluster stores permission data read on nearly every authenticated request. The safeguard that was pacing the background job watched only one health signal — how far the database replicas were lagging — and that signal stayed low the whole time. It did not account for the load building on the primary itself, so the job kept writing while the primary quietly ran toward its limit. <br /><br />When the primary ran out of available connections, requests that needed it could not complete. First, there was no quick timeout on these database calls, so request handlers waited on the stalled database instead of failing fast, and the shared request-handling capacity degraded into site-wide errors. Second, a retry loop around token creation kept re-sending the writes that were already failing, which held the database saturated rather than letting it recover. <br /><br />Monitoring declared the incident at 08:50 UTC, but due to the broad impact and amplification from token creation, it took time to identify the source of the load. First responders mitigated by shedding internal load and pausing the job, and all services recovered by 10:44 UTC. <br /><br />To prevent recurrence, we are rate-limiting background jobs against shared, customer-serving databases by default, and adding automatic pausing and paging on primary-server load rather than replication lag alone. We are also surfacing running background work directly alongside database health signals so responders can see and pause it without leaving those dashboards, bounding retries in the token-issuing path, and adding request-level timeouts so one unhealthy database cannot consume shared web server capacity. In addition, we are breaking apart this database cluster to remove the single point of failure. We will be moving various service-specific data, including the authorization data, out of this shared cluster in the next two weeks.
Source
Sep 4, 202621mSEV 3
Degradation in repos contents APISev 3 from the incident title (no components listed)
2 updates
  1. 22:02 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 22:23 UTC · resolved
    On September 4, 2026, between approximately 21:45 and 22:07 UTC, some users experienced errors and elevated latency for repository operations. The incident was fully resolved at 22:23 UTC.<br /><br />The cause was a capacity change that spread one of our clusters across additional availability zones; our zone-aware traffic routing kept sending requests to the original zone for performance, overloading a small set of servers while the new capacity sat idle. We resolved the incident by reverting the change and letting traffic rebalance.<br /><br />We are improving per-zone capacity guarantees, cross-zone load-shedding, and pre-production testing of multi-zone changes to prevent recurrence.
Source
Sep 4, 20261h 48mSEV 4
Disruption with Copilot Code ReviewSev 4 from the incident title (no components listed)
5 updates
  1. 20:39 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 20:57 UTC · investigating
    Some users may be experiencing failures when using Copilot code review. We have identified the root cause and are working on a mitigation.
  3. 21:54 UTC · investigating
    We are applying the mitigation and expect recovery within approximately 30 minutes.
  4. 22:25 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  5. 22:26 UTC · resolved
    On September 4, 2026, between 20:04 and 22:26 UTC, GitHub Copilot code review experienced an increased failure rate. Affected pull request reviews failed to complete or post review comments.<br /><br />The incident was caused by a change to the service’s authentication permissions that prevented it from submitting affected reviews to the GitHub API. We reverted the change and restored normal operation by 22:26 UTC.<br /><br />We apologize for the disruption.
Source
Sep 3, 20262h 54mSEV 2
Incident with Grok Copilot AI Model ProviderCopilot's Grok 4.5 and 4.6 models saw degraded availability and elevated error rates from about 13:22 to 17:11 UTC on September 3, 2026, caused by an upstream model provider issue. Other models were unaffected. The provider implemented a mitigation and service returned to normal.AI-graded · one model or feature · Copilot AI Model Providers
Why this grade
  1. RULES · SEV 4
    Sev 4 because Copilot AI Model Providers had degraded performance
  2. CLAUDE · SEV 2 · SINGLE · HIGH CONFIDENCE
    Grok 4.5 and 4.6 models in Copilot had elevated errors for about four hours; other models were unaffected.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
5 updates
  1. 14:17 UTC · investigating
    We are investigating reports of degraded performance for Copilot AI Model Providers
  2. 14:20 UTC · investigating
    We are experiencing degraded availability for the Grok 4.6 model in Copilot Chat, VS Code and other Copilot products. This is due to an issue with an upstream model provider. We are working with them to resolve the issue.<br />
  3. 14:22 UTC · investigating
    The Grok 4.5 model has degraded availability as well. We are working with the upstream provider to resolve the issue.
  4. 17:11 UTC · investigating
    The issues with our upstream model provider have been resolved, and Grok models are once again available in Copilot products and IDE surfaces.<br />We will continue monitoring to ensure stability, but mitigation is complete.
  5. 17:11 UTC · resolved
    Between 13:22 and 17:11 UTC on September 03, 2026, GitHub Copilot experienced degradation affecting several Grok models, including Grok 4.5 and Grok 4.6. Users encountered elevated error rates, but other models were not affected. The degradation was caused by an issue with an upstream model provider. GitHub engineers detected the issue through automated monitoring, displayed in-product warnings for the affected models, and coordinated with the provider. Service returned to normal after the provider implemented a mitigation.
Source
Sep 1, 20261h 01mSEV 4
Delays in commit processingFrom about 14:01 to 16:01 UTC, diff updates after pushes were delayed, with median refresh rising from about 3 seconds to over 2 minutes at peak and over 140,000 accounts seeing a delayed refresh. Pushing and opening PRs kept working. A surge in push volume saturated worker pools and autoscaling did not add capacity; manual scaling cleared the backlog.AI-graded · Pull Requests
Why this grade
  1. RULES · SEV 3
    Sev 3 because Pull Requests had degraded performance
  2. CLAUDE · SEV 4 · GLOBAL · MEDIUM CONFIDENCE
    Pull request diff refreshes after pushes were delayed for about two hours; pushes and PR creation kept working.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
4 updates
  1. 15:00 UTC · investigating
    We are investigating reports of degraded performance for Pull Requests
  2. 15:00 UTC · investigating
    Diffs in the PR view may be stale for several minutes. We are investigating and scaling up resources.
  3. 16:00 UTC · investigating
    Time to update pull request diffs have improved to normal thresholds.
  4. 16:01 UTC · resolved
    On September 1, 2026, between approximately 14:01 and 16:01 UTC, updates in response to pushes were delayed, temporarily showing stale diffs. The median time to refresh a diff after a push rose from the normal level of about 3 seconds to over 2 minutes at the peak, and more than 140,000 customer accounts had at least one delayed refresh during the most affected 75 minutes. Pushing commits and opening pull requests continued to work normally. The incident was caused by a sharp, concentrated surge in push volume that saturated worker pools and job queueing infrastructure. Autoscaling did not increase capacity as intended, so the backlog did not clear on its own. <br /><br />The incident was mitigated by manually scaling the affected worker pools and increasing push-processing capacity. This allowed the system to process the backlog, after which refresh times returned to normal. To reduce the likelihood and impact of similar incidents, we are adding quotas and throttling earlier in the push path so a single concentrated source of load cannot saturate shared capacity, improving worker-pool autoscaling so capacity is added automatically, and improving monitors for background job processing so on-call is paged before customers experience delayed pull request updates.
Source
Aug 31, 202642mSEV 4
Elevated rate of errors for OpenAI models provided by CopilotSev 4 because Copilot AI Model Providers had degraded performanceCopilot AI Model Providers
6 updates
  1. 09:15 UTC · investigating
    We are investigating reports of degraded performance for Copilot AI Model Providers
  2. 09:22 UTC · investigating
    Copilot is experiencing a higher rate of errors for OpenAI models, including gpt-5.2, gpt-5.3-codex, gpt-5.4, gpt-5.4, and the gpt-5.6 family of models. Other models are not impacted. <br />
  3. 09:48 UTC · investigating
    One of our model providers has confirmed an incident on their end. We have provided them details to help identify the issue. We are starting to see recovery.
  4. 09:51 UTC · monitoring
    The degradation affecting Copilot AI Model Providers has been mitigated. We are monitoring to ensure stability.
  5. 09:58 UTC · monitoring
    The issues with our upstream model provider have been resolved, and gpt-5.3-codex, gpt-5.4-mini, gpt-5.4-nano, gpt-5.5, and the gpt-5.6 family of models are once again available in Copilot products and IDE surfaces.<br /><br />We will continue monitoring to ensure stability, but mitigation is complete.
  6. 09:58 UTC · resolved
    Between 08:37 and 09:41 UTC on August 31, 2026, GitHub Copilot experienced degradation affecting several GPT models, including gpt-5.2, gpt-5.3-codex, gpt-5.4, gpt-5.4-mini, gpt-5.4-nano, and the gpt-5.6 family (Luna, Sol, and Terra). Users encountered elevated error rates and interrupted streaming responses. Other models were not affected.<br /><br />The degradation was caused by an issue with an upstream model provider. GitHub engineers detected the issue through automated monitoring, displayed in-product warnings for the affected models, and coordinated with the provider. Service returned to normal after the provider implemented a mitigation.
Source
Aug 27, 20262h 08mSEV 2
Incident with Copilot AI Model ProvidersKimi K3 model in Copilot had elevated error rates from about 09:20 to 12:14 UTC on August 27, 2026, caused by an upstream model provider issue. No other models were impacted. The provider's mitigation restored availability, and GitHub is working with the provider on resiliency.AI-graded · one model or feature · Copilot AI Model Providers
Why this grade
  1. RULES · SEV 4
    Sev 4 because Copilot AI Model Providers had a major outage
  2. CLAUDE · SEV 2 · SINGLE · HIGH CONFIDENCE
    Kimi K3 model in Copilot suffered elevated errors for about two and a half hours due to an upstream provider; other models unaffected.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
5 updates
  1. 10:04 UTC · investigating
    We are investigating reports of degraded availability for Copilot AI Model Providers
  2. 10:43 UTC · investigating
    We are experiencing degraded availability for the Kimi K3 model in Copilot products and IDE surfaces. This is due to an issue with an upstream model provider. While we work with them to resolve the issue, we recommend choosing another model or selecting 'Auto' to continue using Copilot.
  3. 11:58 UTC · investigating
    Copilot AI Model Providers is experiencing degraded performance. We are continuing to investigate.
  4. 12:12 UTC · investigating
    The issues with our upstream model provider have been mitigated, and Kimi K3 is once again available in Copilot products and IDE surfaces.<br />We will continue monitoring to ensure stability.
  5. 12:12 UTC · resolved
    On August 27th, 2026, between approximately 09:20 and 12:14 UTC, the Copilot service experienced a degradation of the Kimi K3 model due to an issue with our upstream provider. Users encountered elevated error rates when using Kimi K3. No other models were impacted. <br /><br />The issue was resolved by a mitigation put in place by our provider. GitHub is working with our provider to further improve the resiliency of the service to prevent similar incidents in the future.
Source
Aug 26, 202620h 07mSEV 2
Disruption with GitHub BillingGitHub Billing degraded from August 26 to August 27 2026. Budget pages failed to load and some Copilot CLI sessions failed to start or continue, affecting under 1% of customers. A concentrated workload delayed the data storage layer and automated retries prolonged it. Traffic rebalancing mitigated the incident.AI-graded · some customers
Why this grade
  1. RULES · SEV 4
    Sev 4 from the incident title (no components listed)
  2. CLAUDE · SEV 2 · SUBSET · MEDIUM CONFIDENCE
    Billing budget pages failed to load and Copilot CLI sessions failed to start or continue for under 1% of customers for several hours.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
8 updates
  1. 23:37 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 23:42 UTC · investigating
    We are currently investigating increased errors with billing services. Customers may observe failed billing budget page loads, and users of the Copilot CLI may observe failures starting or continuing sessions.
  3. 00:31 UTC · investigating
    We've applied a mitigation to unblock Copilot usage and have observed recovery for this particular impact. We're continuing to investigate and apply mitigations for the billing page disruption while monitoring to ensure Copilot remains recovered.
  4. 01:35 UTC · investigating
    We are continuing to monitor the mitigation that we have applied for the billing page disruption.
  5. 14:49 UTC · investigating
    Our mitigation is still holding as we continue to investigate to find the root cause.
  6. 16:20 UTC · investigating
    Our mitigation continues to hold, and service conditions remain stable. We are continuing to investigate the concentrated workload responsible for the issue and are preparing additional preventative improvements. We have not identified a material change in customer impact since the previous update. We will provide another update as the investigation progresses.
  7. 17:58 UTC · investigating
    No material change since the previous update. Service conditions remain stable following the mitigation, and we have not observed any further customer impact. We are actively monitoring the service while implementing targeted fixes to address the underlying root cause.
  8. 19:44 UTC · resolved
    On August 26, 2026, between 20:40 UTC and 00:51 UTC on August 27, GitHub Billing experienced degraded performance affecting billing budget pages and GitHub Copilot CLI sessions. Affected customers encountered failed budget page loads or failures when starting or continuing CLI sessions. We confirmed this impact for a small number of customers (<1%). <br /><br />This was caused by a concentrated workload that created processing delays in our data storage layer. Automated retries increased the load and prolonged the degradation. We mitigated the incident by rebalancing traffic within our infrastructure. <br /><br />We are improving workload isolation, retry behavior, and detection of concentrated load to reduce the likelihood of recurrence and shorten our time to detect and mitigate similar incidents.
Source
Aug 26, 20261h 30mSEV 3
Incident with Actions and Pull RequestsGitHub Actions runs triggered by pull requests were delayed, with up to 2.6% (peak 25%) delayed and some failing to trigger, from about 21:55 to 23:58 UTC on August 26, 2026. Some PR merge-commit generation and mergeability info was delayed. Cause was timeouts on one git data partition causing a merge-commit backlog; workload was reduced and traffic shifted to restore service.AI-graded · some customers · Pull Requests +1
Why this grade
  1. RULES · SEV 3
    Sev 3 because Pull Requests had degraded performance
  2. CLAUDE · SEV 3 · SUBSET · HIGH CONFIDENCE
    PR-triggered Actions runs were delayed or failed to start, and some merges and mergeability checks were delayed for about two hours; other work was fine.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
6 updates
  1. 22:56 UTC · investigating
    We are investigating reports of degraded performance for Actions and Pull Requests
  2. 22:57 UTC · investigating
    We are investigating elevated delays and timeouts affecting Actions workflow runs triggered by pull request events. 20% of actions runs have delayed starts of more than 5 minutes and up to 4% of runs failed to trigger. We are actively working on mitigation and will provide updates as we learn more.
  3. 00:01 UTC · investigating
    We've applied mitigations and are seeing recovery in Actions workflow runs and blocked pull request merges. We're continuing to monitor for sustained health of merge commit creates before resolving.
  4. 00:25 UTC · investigating
    We confirmed full recovery beginning at 23:58 UTC. Actions workflow runs and pull request merges are operating normally. We will now resolve the incident while continuing to monitor service health.
  5. 00:26 UTC · monitoring
    The degradation affecting Actions and Pull Requests has been mitigated. We are monitoring to ensure stability.
  6. 00:26 UTC · resolved
    On August 26, 2026, from 21:55 UTC to 23:58 UTC, 2.6% of workflow runs triggered by pull request events were delayed, with the impact rising as high as 25% at its peak. Some users also experienced delays in pull request merge-commit generation, mergeability information, and merge-button availability. Actions and Pull Requests fully recovered by 23:58 UTC; the incident was resolved at 00:26 UTC after normal operation was confirmed. <br /><br />Background jobs that process pull request updates and generate merge commits were impacted by timeouts reaching a single partition of git data. This resulted in a backlog in pull request merge-commit processing, delaying pull request-triggered GitHub Actions workflows and some mergeability information. <br /><br />We reduced workload, shifted traffic away from affected infrastructure, and restored the affected service component to a healthy state. Together, these actions helped drain the backlog and restore normal operations. <br /><br />We are working to improve resource saturation detection and to eliminate customer impact in this scenario by isolating impact, placing better bounds on retries, and strengthening backpressure to make our systems more resilient under load.
Source
Aug 26, 20262h 50mSEV 1
Incident with ActionsActions jobs failed to start from 15:02 to 15:45 UTC on August 26, 2026, and runs were delayed more than 5 minutes until 17:40 UTC. Cause was database primary write saturation during peak load plus an upstream event-processing issue. Throttling and restarts restored service, but about 3.7% of larger-runner jobs stayed stuck until force-revoked. Pages was degraded early and recovered by 16:49 UTC.AI-graded · Actions +1
Why this grade
  1. RULES · SEV 1
    Sev 1 because Pages had a major outage
  2. CLAUDE · SEV 1 · GLOBAL · HIGH CONFIDENCE
    Actions jobs failed to start for about 43 minutes, then runs were delayed over two hours; Pages had degraded performance early on.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
11 updates
  1. 15:11 UTC · investigating
    We are investigating reports of degraded availability for Actions
  2. 15:12 UTC · investigating
    Pages is experiencing degraded performance. We are continuing to investigate.
  3. 15:23 UTC · investigating
    We've identified an issue with a database primary and are failing over to a replica immediately
  4. 15:48 UTC · investigating
    primary failover briefly improved performance but did not fully mitigate, we've throttled inbound traffic and are investigating upstream Vitess issues
  5. 16:14 UTC · investigating
    We believe we've identified and addressed the issue and are ramping traffic back up slowly to ensure it doesn't recur. Some customers will continue to see delays as we ramp up.
  6. 16:49 UTC · investigating
    Pages is operating normally.
  7. 16:50 UTC · investigating
    We are continuing to observe recovery and delayed queues are burning down. Some customers will continue to see increased delays until all throttled work has been completed - we expect this within the next hour.
  8. 17:32 UTC · investigating
    We are continuing to observe recovery and expect actions inbound queues to be back to normal in <30min. Work will continue to flow through the system subject to per-customer concurrency limits.
  9. 17:54 UTC · monitoring
    The degradation affecting Actions has been mitigated. We are monitoring to ensure stability.
  10. 18:00 UTC · monitoring
    All inbound queues have recovered and Actions is operating as expected. 3.7% of jobs assigned to larger runners during the early stage of this incident are stuck waiting for runner assignment. Those will be canceled within the hour. Other runners are successfully processing all new jobs.
  11. 18:01 UTC · resolved
    On August 26, 2026 from 15:02 to 15:45 UTC, Actions jobs failed to start. The following 2 hours until 17:40 UTC, Actions runs were delayed starting by more than 5 minutes as the system caught up with delayed load. This impact was triggered by saturation of writes to the database primary used by the service processing triggers for Actions workflows. The primary was failed over, but the system did not fully recover. The saturation was caused by growing daily peak load combined with an upstream issue in GitHub’s event processing infrastructure, https://www.githubstatus.com/incidents/hcbtzksccj2f, which caused burst amplification of already-high load. Downstream throttles that were later used to recover were set ~10% too high to protect the system. <br /><br />At 15:45 UTC, throttling combined with service restarts recovered the service’s core health. Those throttles were gradually raised between 15:54 and 17:22 to restore full webhook processing for Actions runs. This ramp was deliberately slow to ensure we did not re-overwhelm the system given our original throttling was now known to be incorrectly set. The queue of webhook events was fully burned down at 17:40 UTC. <br /><br />3.7% of larger-runner jobs, along with some scale-set self-hosted jobs, remained stuck in queued or “waiting for runner” state. We deployed a change to force-revoke jobs in this state, and they transitioned to failed at 18:40 UTC, about 50 minutes after incident mitigation. Releasing these jobs also freed hosted concurrency for larger-runner jobs. <br /><br />Customers using concurrency groups saw longer impact due to a separate issue where runners assigned to a subset of jobs disconnected before the force-revoke mitigation was deployed, which prevented runner acquisition from progressing and left jobs in a waiting-for-runner state. This was resolved at 01:00 UTC on August 27. <br /><br />Some runs triggered during the 15:02-15:45 UTC incident window encountered a bug that left them showing as queued even after service recovery. In the backend, these runs had already failed and will automatically move to canceled state 24 hours after creation. As follow-up, we are fixing the root cause of this queued state and improving our ability to bulk-cancel affected runs. <br /><br />Several changes to improve the general scalability of this part of Actions were already complete and deploying to production. Rollout of those changes will be complete within the next 24 hours. Further work to improve scale, resiliency, and more graceful degradation of Actions workflows are in flight. We are also taking a repair item to accelerate clearing of stuck queued or waiting jobs in similar future cases.
Source
Aug 26, 202659mSEV 3
Disruption with some GitHub servicesSev 3 from the incident title (no components listed)
2 updates
  1. 15:09 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 16:07 UTC · resolved
    Please refer to the combined summary in this related incident: https://www.githubstatus.com/incidents/y1t7p9fzrlj2
Source
Aug 24, 202638mSEV 2
Actions delays in starting runsGitHub Actions runs were delayed or failed to start for a subset of runs from about 13:33 to 14:04 UTC on August 24, 2026. A disk failure on one node handling runner assignment events kept sending healthy signals, so its queued work was not moved. An automatic rebalance at 13:54 UTC redirected processing, and the backlog was cleared by 14:04 UTC.AI-graded · some customers · Actions
Why this grade
  1. RULES · SEV 3
    Sev 3 because Actions had degraded performance
  2. CLAUDE · SEV 2 · SUBSET · HIGH CONFIDENCE
    About 3.8% of Actions runs were delayed over five minutes and 1.25% failed outright for roughly 30 minutes.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
4 updates
  1. 13:56 UTC · investigating
    We are investigating reports of degraded performance for Actions
  2. 14:22 UTC · investigating
    Failures while queuing and running Actions jobs for a subset of customers are now resolving. We are monitoring for full recovery.
  3. 14:26 UTC · monitoring
    The degradation affecting Actions has been mitigated. We are monitoring to ensure stability.
  4. 14:34 UTC · resolved
    On August 24, 2026, between 13:33 UTC and 14:04 UTC, 3.8% of Actions runs experienced start delays over 5 minutes with 1.25% of Actions runs failing outright. <br /> <br />The incident was caused by a disk failure on a node hosting one of many service instances responsible for processing runner assignment events. Typically, pods on unhealthy nodes are removed and replaced automatically without impact. In this case, although the node was severely degraded and unable to perform disk operations, it continued sending healthy signals, preventing the system from immediately moving its work elsewhere. During this period, events assigned to the affected component accumulated until an automatic rebalance redirected processing to healthy components at 13:54 UTC. The queue backlog was cleared at 14:00 UTC, and processing returned to normal by 14:04 UTC. <br /><br />To prevent a recurrence, we are improving detection and automated remediation for unhealthy nodes that aren’t fully offline. We are also strengthening application-level resiliency, so stalled consumers are automatically removed quickly and their work reassigned without waiting for the affected node to recover.
Source
Aug 24, 20260mSEV 3
Degraded Git Operations over SSHSev 3 from the incident title (no components listed)
1 update
  1. 14:00 UTC · resolved
    On August 21, 2026, between 14:00 and 14:07 UTC, dotcom Git operations over SSH were degraded. Successful Git operations over SSH fell by more than 95% for during the peak impact window, making clone, fetch, or push over SSH effectively unavailable to most users for approximately four minutes. Git operations over HTTPS were not affected. The incident was caused by a software defect in our load-balancing infrastructure that was triggered by a configuration change. The defect only occurred when connections passed through multiple layers of load balancers running the new configuration, which meant it was not detected during canary testing. We mitigated the incident by rolling back the configuration change. We are adding regression coverage for multi-layer load-balancer configurations and improving monitoring and alerting for Git operations over SSH to reduce our time to detection and mitigation of similar issues in the future.
Source
Aug 24, 202647mSEV 4
Elevated errors on Fable 5 due to upstream providerSev 4 because Copilot AI Model Providers had a partial outageCopilot AI Model Providers
3 updates
  1. 07:12 UTC · investigating
    We are investigating reports of degraded availability for Copilot AI Model Providers
  2. 07:12 UTC · investigating
    We are experiencing degraded availability for the Fable model in Copilot products and IDE surfaces. This is due to an issue with the upstream model provider. While we work with them to resolve the issue, we recommend choosing another model or selecting 'Auto' to continue using Copilot.
  3. 07:58 UTC · resolved
    On August 24th, 2026, between approximately 06:35 and 07:25 UTC, the Copilot service experienced a degradation of the Claude Fable 5 model due to an issue with our upstream provider. Users encountered elevated error rates when using Claude Fable 5, with requests sometimes failing mid-response. No other models were impacted.<br /><br />The issue was resolved by a mitigation put in place by our provider. GitHub is working with our provider to further improve the resiliency of the service to prevent similar incidents in the future.
Source
Aug 20, 20269h 54mSEV 2
Intermittent failures creating agent tasksFor about 11 hours, some Copilot Cloud Agent users saw agent task status and results delayed 60 to 90 minutes. Tasks kept running and completing. A regional outage at a third-party cloud database storing task status caused it; the database was failed over and backlog processed. No task data was lost.AI-graded · some customers
Why this grade
  1. RULES · SEV 1
    Sev 1 because the provider reported data loss or exposure
  2. CLAUDE · SEV 2 · SUBSET · MEDIUM CONFIDENCE
    Copilot Cloud Agent task status and session output were delayed 60 to 90 minutes for some users; tasks kept completing.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
12 updates
  1. 14:43 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 14:51 UTC · investigating
    Users may experience delays when starting tasks using Copilot Cloud Agent. We are actively investigating the issue and will provide updates as we learn more.
  3. 15:01 UTC · investigating
    We have identified the problematic component and are working to fail over to a healthy instance. Further updates will be provided as we perform mitigations.
  4. 15:41 UTC · investigating
    We are experiencing issues with Copilot Cloud Agent tasks, resulting in newly started tasks not properly displaying on-going progress. These Copilot Cloud Agent tasks are still being completed correctly but lack proper visibility. We are actively investigating the issue and will provide updates as we learn more.
  5. 16:14 UTC · investigating
    Users are experiencing delays when starting tasks using Copilot Cloud Agent and are not be able to see the status of these tasks. Copilot Cloud Agent tasks are still being completed. We have identified the cause of the issue and are putting mitigations in place to return service to normal levels. We will provide another update about the expected recovery time shortly.
  6. 17:05 UTC · investigating
    We are seeing signs of recovery for Copilot Cloud Agent task status visibility, but this recovery is slower than anticipated. We are pursuing additional mitigating measures to accelerate recovery.
  7. 17:32 UTC · investigating
    We are observing gradual recovery for Copilot Cloud Agent task status visibility, with session output delayed approximately 1 hour. We have taken additional steps to accelerate the recovery and are continuing to monitor the impact.
  8. 18:04 UTC · investigating
    We are continuing to observe gradual recovery for Copilot Cloud Agent task status visibility, with session output delayed by approximately 1 hour. We have taken additional steps to accelerate the recovery and are continuing to monitor the impact.
  9. 18:45 UTC · investigating
    We are continuing to observe gradual recovery for Copilot Cloud Agent task status visibility, with session output delayed by approximately 1 hour. We have taken additional steps to accelerate the recovery and expect this to take effect within the next hour.
  10. 19:35 UTC · investigating
    We are continuing to observe gradual recovery for Copilot Cloud Agent task status visibility. Session output continues to be delayed by approximately 1 hour as our remediation steps take effect.
  11. 20:37 UTC · investigating
    We are seeing gradual recovery in Copilot Cloud Agent task status visibility as we deploy a fix for the root cause. Session output remains delayed by approximately one hour while remediation continues.
  12. 00:37 UTC · resolved
    Between 13:57 UTC on August 20 and 00:37 UTC on August 21, 2026, some users of the Copilot Cloud Agent experienced delays of up to 60 to 90 minutes in seeing the status and results of their agent tasks. The agent tasks themselves continued to run and complete during this time; only the visibility of their status was delayed.<br /><br />The cause was a regional outage in a third-party cloud database service that Copilot uses to store agent task status. We failed over the affected database to a healthy region, added processing capacity to work through the backlog, and restored normal operation once the underlying service recovered. No task data was lost during the incident.<br /><br />To prevent repetition of similar incidents, we are removing the database configuration that made us vulnerable to this regional outage and improving our database failover procedures.
Source
Aug 18, 202647mSEV 2
Incident with ActionsAn expired authentication certificate for an Actions backend service blocked Larger Runner jobs from starting and blocked viewing or managing runners and runner groups from 05:02 to 11:30 UTC on August 18. Engineers completed enabling the rotated certificate to mitigate, and added monitoring for certificates.AI-graded · some customers
Why this grade
  1. RULES · SEV 3
    Sev 3 from the incident title (no components listed)
  2. CLAUDE · SEV 2 · SUBSET · HIGH CONFIDENCE
    Larger Runner jobs could not start and Actions runners and runner groups could not be viewed or managed for about six hours; standard runners were not described as affected.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
2 updates
  1. 09:36 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 10:23 UTC · resolved
    On August 18, 2026, between 05:02 UTC and 11:30 UTC, customers were unable to run jobs on Actions Larger Runners and were unable to view or manage Actions Runners and Runner Groups through the GitHub UI and API. <br /><br />These issues were caused by failures in backend requests resolving essential metadata for starting Larger Runner workflow runs and for reading runner and runner group data. The failures were caused by an expired authentication certificate unique to this service. The certificate had been rotated in KeyVault, but a step to enable use at runtime had been paused to prevent recurrence of previous incidents that had been triggered by this operation. <br /><br />We mitigated the issues by completing the enablement of the new certificate in the backend system. We have added additional monitoring to this and other certificates. The relevant service is also in the process of being replaced as part of our availability and scale work, bringing this authentication path and secret management in line with patterns across all GitHub services.
Source
Aug 18, 20264h 02mSEV 2
Intermittent failures in runner group and runner-related permissions pagesSev 2 from the incident title (no components listed)
5 updates
  1. 07:40 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 07:40 UTC · monitoring
    We are investigating reports of failure to load runner groups and runner-related permissions for customers using larger runners.
  3. 10:41 UTC · monitoring
    We have identified the source of a communication issue between Actions services and are working toward mitigation. Customers may experience failure to load runner groups and runner-related permissions issues when using Larger Runners.
  4. 11:24 UTC · monitoring
    We have applied a mitigation and are seeing recovery signals. We will continue monitoring recovery and providing updates.
  5. 11:42 UTC · resolved
    On August 18, 2026, between 05:02 UTC and 11:30 UTC, customers were unable to view or manage Actions Runners and Runner Groups through the GitHub UI and API. <br /><br />The issue was caused by failures in backend requests reading runner and runner group data. The failures were caused by an expired authentication certificate unique to this service. The certificate had been rotated in KeyVault, but a step to enable use at runtime had been paused to prevent recurrence of previous incidents triggered by this operation. <br /><br />The impact was mitigated by completing the enablement of the new certificate in the backend system. We have added additional monitoring to this and other certificates. This service is also in the process of being replaced as part of our availability and scale work, bringing this authentication path and secret management in line with patterns across all GitHub services.
Source
Aug 17, 20267h 36mSEV 2
Incident with GitHub.comGitHub.com had about 20% web and API error rates from 13:28 to 21:15 UTC, with archive downloads near 50%. Cause was network saturation on Central US load balancers, traced to an autoscaling misconfiguration and HAProxy flow-limit exhaustion. Traffic was shifted to Northern Virginia. SAML/OIDC, SCIM, Team Sync, Actions and Copilot auth were affected; Copilot recovered after retry-storm mitigation.AI-graded · regional · Git Operations +7
Why this grade
  1. RULES · SEV 1
    Sev 1 because Pages had a major outage
  2. CLAUDE · SEV 2 · REGIONAL · HIGH CONFIDENCE
    Elevated errors around 20% across web, API, Git and Actions for about eight hours, with archive downloads near 50%, traced to Central US load balancers.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
36 updates
  1. 13:40 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 13:41 UTC · investigating
    API Requests is experiencing degraded performance. We are continuing to investigate.
  3. 13:42 UTC · investigating
    Actions is experiencing degraded performance. We are continuing to investigate.
  4. 13:44 UTC · investigating
    Webhooks is experiencing degraded performance. We are continuing to investigate.
  5. 13:45 UTC · investigating
    We are seeing an approximate 20% error rate across numerous experiences including Pull Requests, Issues, and others. Investigations are currently under way and we will be posting updates as they become available
  6. 13:46 UTC · investigating
    Issues is experiencing degraded performance. We are continuing to investigate.
  7. 13:58 UTC · investigating
    Pull Requests is experiencing degraded performance. We are continuing to investigate.
  8. 14:04 UTC · investigating
    We are experiencing high error rates around 20% for web experiences and api traffic. Archive downloads and raw repository content downloads are experiencing an approximate 50% error rate. Investigations are on-going into the root cause, and updates will continue to be provided as we investigate.
  9. 14:24 UTC · investigating
    We are experiencing high error rates around 20% for web experiences and api traffic. Archive downloads and raw repository content downloads are experiencing an approximate 50% error rate. SAML and OIDC authentication, SCIM, and Team Sync are also impacted. Investigations are on-going and we will continue to provide updates as we discover more information.
  10. 14:31 UTC · investigating
    Copilot is experiencing degraded availability. We are continuing to investigate.
  11. 14:45 UTC · investigating
    Pull Requests is experiencing degraded availability. We are continuing to investigate.
  12. 14:49 UTC · investigating
    Issues is experiencing degraded availability. We are continuing to investigate.
  13. 14:54 UTC · investigating
    Pull Requests is experiencing degraded availability. We are continuing to investigate.
  14. 14:58 UTC · investigating
    Actions is experiencing degraded availability. We are continuing to investigate.
  15. 14:58 UTC · investigating
    We are experiencing high error rates around 20% for web experiences and api traffic. Archive downloads and raw repository content downloads are experiencing an approximate 50% error rate. SAML and OIDC authentication, SCIM, and Team Sync are also impacted. We are currently performing mitigations based on our investigation thus far and are monitoring for improvement.
  16. 14:58 UTC · investigating
    Webhooks is experiencing degraded availability. We are continuing to investigate.
  17. 15:01 UTC · investigating
    API Requests is experiencing degraded availability. We are continuing to investigate.
  18. 15:10 UTC · investigating
    Pages is experiencing degraded performance. We are continuing to investigate.
  19. 15:21 UTC · investigating
    Git Operations is experiencing degraded performance. We are continuing to investigate.
  20. 15:40 UTC · investigating
    Webhooks is experiencing degraded performance. We are continuing to investigate.
  21. 15:42 UTC · investigating
    We are experiencing high error rates around 20% for web experiences and api traffic. Archive downloads and raw repository content downloads are experiencing an approximate 50% error rate. SAML and OIDC authentication, SCIM, and Team Sync are also impacted. We are currently performing mitigations and will post updates as we progress.
  22. 16:16 UTC · investigating
    We are experiencing high error rates around 20% for web experiences and api traffic. Archive downloads and raw repository content downloads are experiencing an approximate 50% error rate. SAML and OIDC authentication, SCIM, and Team Sync are also impacted. We are still working to identify the root cause and will continue to post updates as we learn more and perform mitigation.
  23. 16:36 UTC · investigating
    We identified the problematic component and have taken corrective actions. There are strong signs of recovery but we are still working to completely restore service, with error rates still remaining slightly elevated. We will post further updates as recovery continues.
  24. 16:59 UTC · investigating
    The degradation affecting API Requests, Actions, Git Operations, Issues, Pages, Pull Requests and Webhooks has been mitigated. We are monitoring to ensure stability.
  25. 17:30 UTC · investigating
    Git Operations is experiencing degraded performance. We are continuing to investigate.
  26. 17:34 UTC · investigating
    We identified the problematic component and have taken corrective actions, but we are seeing residual impact across numerous services. We are continuing to apply additional mitigations and investigate the remaining impact.
  27. 17:36 UTC · investigating
    Issues is experiencing degraded performance. We are continuing to investigate.
  28. 18:11 UTC · investigating
    We identified the problematic component and have taken corrective actions, but we are seeing residual impact in the form of sporadic authentication failures. We are continuing to apply additional mitigations and investigate the remaining impact.
  29. 18:23 UTC · investigating
    The degradation affecting Git Operations has been mitigated. We are monitoring to ensure stability.
  30. 18:48 UTC · investigating
    API Requests is experiencing degraded availability. We are continuing to investigate.
  31. 19:01 UTC · investigating
    API Requests is operating normally.
  32. 19:13 UTC · investigating
    We are continuing to investigate sporadic authentication failures. We have partially disabled authentication token retries and have seen improvement, and we are monitoring impact before fully applying this mitigation.
  33. 20:08 UTC · investigating
    We are continuing to investigate sporadic failures affecting Copilot authentication in some applications. Copilot usage via the GitHub CLI and GitHub App are unaffected.
  34. 20:22 UTC · investigating
    Issues is operating normally.
  35. 20:45 UTC · investigating
    We are continuing to apply mitigations to address sporadic Copilot authentication failures in some applications. We expect full recovery within the next 30 minutes. Copilot usage via the GitHub CLI and GitHub App are unaffected.
  36. 21:15 UTC · resolved
    On August 17, 2026, from 13:28–21:15 UTC (7h 47m), GitHub.com experienced elevated errors and latency across Issues, Pull Requests, APIs, Actions, and Copilot. At peak, web/API error rates were approximately 20%, while archive and raw-content downloads reached approximately 50%. SAML/OIDC authentication, SCIM, and Team Sync were also affected, as well as Actions workflows in GHEC with Data Residency that depend on public workflow step definitions hosted on GitHub.com. Most services recovered by 16:36 UTC as our Central US datacenter recovered; Actions was degraded until approximately 18:03 UTC; and Copilot Token Service fully recovered by 21:02. <br /><br />Some of the failing traffic was moved from Central US to Northern Virginia where it was served successfully until the network failure in Central US was debugged and resolved. Delayed replies to a single internal endpoint triggered a latent retry bug in VS Code that amplified traffic by approximately 10x and caused delayed recovery for the Copilot Token Service. <br /><br />The immediate cause of the failure was network saturation on load balancers in Central US due to a new peak in traffic. Originally this was caused by an Istio sidecar pod reaching its concurrency limits and failing to auto scale correctly because of a misconfigured policy that watched host service but not sidecar limits. One failure cascaded to more and eventually four HAProxy nodes exhausted their flow limits, degrading the gateway auth path and causing widespread authentication latency and failures. The problem was worsened by optimistic retry logic which overloaded internal load balancers. Pausing HAProxy on those nodes simultaneously produced immediate broad recovery. <br /><br />The retry storm in Northern VA was fixed by 1) temporarily reducing gateway retry logic with a PR and 2) blocking inbound Copilot Token Service token requests at the load balancers with a 403, and then gradually ramping back up traffic per-site to allow callers to succeed. <br /><br />Residual Copilot authentication failures continued because client retry behavior amplified load: a failed token operation could generate many extra requests and enter a retry loop. Copilot Token Service traffic increased from a normal 7–9K RPS to 70–100K RPS. Reducing gateway authentication retries and blocking retry-triggering responses stabilized Copilot Token Service and completed recovery. <br /><br />Complicating factors that impeded recovery included a number of scraping attacks on codeload endpoints. <br /><br />To prevent recurrence, our follow-up actions include: <br /><br />- Correcting autoscaling policies to account for service-mesh sidecar concurrency and capacity. <br /><br />- Auditing Istio request, concurrency, and scaling limits across affected services. <br /><br />- Reviewing retry limits and backoff behavior across gateways and clients. <br /><br />- Addressing the VS Code retry behavior that amplified Copilot token traffic. <br /><br />- Improving load-balancer capacity monitoring and regional failover safeguards.
Source
Aug 13, 20262h 06mSEV 3
Disruption with GHEC Team SyncGitHub Enterprise Cloud team sync with IdP groups was degraded for enterprises using personal accounts from about 15:31 to 18:27 UTC on August 13, 2026. Affected teams saw sync delays of 3 to 13 hours, delaying access grants and removals for 2.8% of teams. A temporary change left active after a prior issue caused the delays; it was removed and resources were added.AI-graded · some customers
Why this grade
  1. RULES · SEV 4
    Sev 4 from the incident title (no components listed)
  2. CLAUDE · SEV 3 · SUBSET · HIGH CONFIDENCE
    Team Sync to IdP groups was delayed for enterprises with personal accounts, leaving 2.8% of teams with delayed access changes for about 3 hours.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
4 updates
  1. 16:21 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 16:21 UTC · investigating
    GHEC Team Sync is currently degraded for enterprises with personal accounts, causing delays when syncing teams to IdP groups. We have identified the cause of the delays and are working on a mitigation. We will provide an update on our progress at 20:00 UTC.
  3. 18:27 UTC · investigating
    We have deployed a mitigation. At this time GHEC Team Sync has recovered for enterprises with personal accounts. Teams syncing to IdP groups have returned to their normal cadence.
  4. 18:27 UTC · resolved
    On August 13, 2026, from 15:31:21 UTC to 18:27:55 UTC, GitHub Enterprise Cloud team synchronization was degraded for enterprises using personal accounts. Organization teams experienced delays of up to 3 to 13 hours (median 8 hours) when syncing with IdP groups, resulting in delayed access grants or removals for enterprise users across 2.8% of teams. <br /><br />A temporary change introduced to address a previous issue due to increased usage of this feature remained active after it was intended to be removed, causing synchronization delays during periods of high volume. We removed the temporary change and provisioned additional resources to handle the increased volume.
Source
Aug 13, 202651mSEV 2
Incident with WebhooksA background organization-deletion job overloaded a shared database, causing brief elevated errors and slowness. Webhook management API was worst hit, with smaller impact to Git operations, pull requests, issues, packages, sign-in and Copilot. Failures peaked near 1% around 14:37 UTC; the job was disabled and impact cleared by 14:53 UTC.}AI-graded · some customers · Git Operations +4
Why this grade
  1. RULES · SEV 3
    Sev 3 because Git Operations had degraded performance
  2. CLAUDE · SEV 2 · SUBSET · HIGH CONFIDENCE
    Elevated errors and slow responses hit webhook management and smaller parts of Git, PRs, issues and packages for about an hour; failures peaked near 1%.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
9 updates
  1. 14:45 UTC · investigating
    We are investigating reports of degraded performance for Webhooks
  2. 14:46 UTC · investigating
    Pull Requests is experiencing degraded performance. We are continuing to investigate.
  3. 14:46 UTC · investigating
    Issues is experiencing degraded performance. We are continuing to investigate.
  4. 14:46 UTC · investigating
    Git Operations is experiencing degraded performance. We are continuing to investigate.
  5. 14:56 UTC · investigating
    Packages is experiencing degraded performance. We are continuing to investigate.
  6. 14:58 UTC · investigating
    We are currently investigating a brief degradation of service for Git operations (specifically pushes), issues, pull requests, package registry, and webhooks between 14:32 and 14:46 UTC. We have identified the source of the degradation and are investigating mitigation strategies to prevent recurrence.
  7. 15:33 UTC · monitoring
    The degradation affecting Git Operations, Issues, Packages, Pull Requests and Webhooks has been mitigated. We are monitoring to ensure stability.
  8. 15:33 UTC · monitoring
    We have temporarily disabled a background job which caused the impact. At this time the impact is fully mitigated.
  9. 15:36 UTC · resolved
    Between 14:24 and 14:53 UTC on 13 August 2026, a routine background job to delete an organization overwhelmed a key shared database, causing multiple GitHub services to briefly return elevated errors and slower responses. Most affected was the webhook management API, with smaller impact to Git operations, pull requests, issues, packages, sign-in, and Copilot. Impact cleared on its own at about 14:53 UTC once the job finished; we resolved the incident at 15:36 UTC. <br /><br />Affected users may have experienced a brief increase in errors and slower responses, primarily when creating, listing, or updating webhooks, with smaller impacts to pull requests, issues, packages, and Git operations. Failures peaked at about 1% for several minutes around 14:37 UTC. <br /><br />To prevent future incidents, we've already shipped an update that turns on the safer deletion path for organizations, along with caps on deletion holds on databases. Building on these changes, we're auditing all bulk deletion and cleanup jobs that write to shared databases to prevent similar issues in future.
Source
Aug 13, 20261h 04mSEV 4
Errors with the Fable 5 Model in CopilotSev 4 because Copilot AI Model Providers had degraded performanceCopilot AI Model Providers
5 updates
  1. 14:43 UTC · investigating
    We are investigating reports of degraded performance for Copilot AI Model Providers
  2. 14:50 UTC · investigating
    We are experiencing degraded availability for the Fable 5 model in Copilot products and IDE surfaces. This is due to an issue with an upstream model provider. While we work with them to resolve the issue, we recommend choosing another model or selecting 'Auto' to continue using Copilot.
  3. 15:23 UTC · investigating
    We are seeing modest recovery, but are still experiencing degraded availability for the Fable 5 model in Copilot products and IDE surfaces. This is due to an issue with an upstream model provider. While we work with them to resolve the issue, we recommend choosing another model or selecting 'Auto' to continue using Copilot.
  4. 15:47 UTC · investigating
    The issues with our upstream model provider have been resolved, and Fable 5 is once again available in Copilot products and IDE surfaces.<br /><br />We will continue monitoring to ensure stability, but mitigation is complete.
  5. 15:47 UTC · resolved
    On August 13th, 2026, between approximately 14:06 and 15:47 UTC, the Copilot service experienced a degradation of the Claude Fable 5 model due to an issue with our upstream provider. Users encountered elevated error rates, peaking at 43% and averaging 12%. Users who selected Auto or alternative models were unaffected.<br /><br />The issue was resolved by a mitigation put in place by our provider. GitHub is working with our provider to further improve the resiliency of the service to prevent similar incidents in the future.
Source
Aug 12, 20261h 18mSEV 2
Disruption with Login and Release Asset downloadsLogged-out requests to github.com, including the sign-in page and release asset downloads, returned HTTP 5xx errors across three windows on August 12 and 13, 2026, with error rates averaging 2-17%. Signed-in users were unaffected. An unusual traffic pattern overloaded part of the infrastructure; edge traffic controls mitigated it, and monitoring was tightened afterward.AI-graded · some customers
Why this grade
  1. RULES · SEV 3
    Sev 3 from the incident title (no components listed)
  2. CLAUDE · SEV 2 · SUBSET · HIGH CONFIDENCE
    Anonymous github.com sign-in pages and release asset downloads returned 5xx errors for hours at 2-29% error rates; signed-in users were unaffected.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
4 updates
  1. 21:39 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 21:43 UTC · investigating
    We are investigating issues with Login and when downloading Release Assets. We will continue to keep users updated on progress towards mitigation.
  3. 22:22 UTC · investigating
    We have identified the root cause and are working on mitigation. Errors on the login page and downloading release assets have decreased, but we are not fully mitigated. We will continue to provide updates.
  4. 22:56 UTC · resolved
    On August 12 and 13, 2026, some anonymous (logged-out) requests to github.com experienced HTTP 5xx errors when loading pages like the sign-in page, and when downloading release assets, due to an unusual traffic pattern that repeatedly overloaded a part of our infrastructure that serves these types of requests. There were three windows of impact: (1) August 12 from 16:34 to 18:34 UTC, with an average error rate of 16.16% that peaked at 28.6%; (2) August 12 from 19:00 to 22:56 UTC, with an average error rate of 16.55% that peaked at 24.18%; and (3) August 13 from 06:19 to 08:05 UTC, with an average error rate of 2.01% that peaked at 7.49%.<br />Requests from signed-in users were unaffected.<br /><br />We mitigated the incidents by applying traffic controls at our network edge that limited any requests matching the pattern identified previously, thereby preventing overload on our systems.<br /><br />Since these incidents occurred, we have tightened our monitoring systems to alert server-side errors that affect logged-out traffic. We are also working to further strengthen our edge protections and reduce the time to detect and mitigate similar incidents.
Source
Aug 12, 202625mSEV 3
Incident with Pull Requests and IssuesSev 3 because Pull Requests had degraded performanceIssues +1
5 updates
  1. 16:16 UTC · investigating
    We are investigating reports of degraded performance for Issues and Pull Requests
  2. 16:24 UTC · investigating
    We are investigating reports of errors affecting Pull Requests and Issues on GitHub.com. Some users may encounter 500 errors when loading pull request and issue pages. Our engineering teams are actively investigating the root cause, which appears to be related to a database infrastructure issue. We will provide an update as soon as we have more information.
  3. 16:35 UTC · monitoring
    The degradation affecting Issues and Pull Requests has been mitigated. We are monitoring to ensure stability.
  4. 16:38 UTC · monitoring
    We identified the source of errors affecting Pull Requests, Issues, and Search on GitHub.com and have applied a mitigation. A database index hint was referencing an index that had been removed by a recent migration, causing query failures for some users. We disabled the problematic configuration and are seeing recovery across affected services. We are continuing to monitor to confirm full resolution.
  5. 16:41 UTC · resolved
    Between 16:03 and 16:29 UTC on August 12, some users encountered errors when viewing pull requests, issues, and search results. During this period, about 1.9% of Pull Request requests and 0.9% of Issues requests failed. During a database migration, two indexes were removed while application settings still referenced them, causing affected requests to fail. We detected the issue after the migration reached one database shard and before it progressed to the remaining shards. We restored service by disabling both settings. We are improving safeguards around database migrations and application configuration to prevent similar mismatches from causing errors.
Source
Aug 11, 20265h 16mSEV 2
Incident with GraphQL API RequestsGitHub's GraphQL API had elevated timeouts from about 14:00 to 16:00 UTC on August 11, 2026, averaging 0.06% of requests and peaking at 0.14%. Cause was increased utilization at one site causing resource contention across dependencies. Fixed by increasing capacity; error rates returned to normal by 20:06 UTC.AI-graded · regional · API Requests
Why this grade
  1. RULES · SEV 3
    Sev 3 because API Requests had degraded performance
  2. CLAUDE · SEV 2 · REGIONAL · MEDIUM CONFIDENCE
    GraphQL API requests saw a small rise in timeouts, about 0.06% on average and 0.14% at peak, for roughly two hours, tied to one site.
  3. Claude Haiku 5.5 · prompt v7 · Oct 23, 2026 · Raw record
6 updates
  1. 14:50 UTC · investigating
    We are investigating reports of degraded performance for API Requests
  2. 15:28 UTC · investigating
    We are investigating reports of a small increase in error rates affecting GraphQL API requests. We are working on increasing capacity and continue to investigate the increased errors. We will provide another update when we have more information
  3. 16:49 UTC · investigating
    We have returned to a healthy baseline on GraphQL API requests. We will continue to work on investigations into the errors seen during this incident.
  4. 16:49 UTC · monitoring
    The degradation affecting API Requests has been mitigated. We are monitoring to ensure stability.
  5. 20:06 UTC · monitoring
    We have identified and mitigated increased error rates affecting GraphQL API requests. A fix to increase service capacity has been deployed and error rates have returned to normal levels. We are resolving this incident.
  6. 20:06 UTC · resolved
    On August 11, 2026, between 14:00 UTC and 16:00 UTC the GraphQL API service was degraded and customers in saw higher than normal timeouts. On average, the timeout rate was 0.06% and peaked at 0.14% of requests routing to the service. <br /><br />This was due to increased utilization at one of our sites which caused resource contention across our dependencies, leading to an increase in timeouts for GraphQL requests. We mitigated the incident by increasing capacity to alleviate the capacity bottleneck. <br /><br />We are working to improve our monitoring so that we can proactively reduce the impact of high consumption requests in addition to scaling up; Additionally, we will improve our time to detection and mitigation of issues like this one in the future.
Source
Aug 10, 20261h 23mSEV 4
Disruption with Copilot for access to some modelsSev 4 from the incident title (no components listed)
6 updates
  1. 20:27 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 20:39 UTC · monitoring
    We are currently investigating reports of some Copilot users experiencing issues accessing certain models. Affected users may see errors or degraded functionality when attempting to use specific models. We are actively working on a fix and will provide updates as we have more information.
  3. 20:49 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  4. 21:19 UTC · monitoring
    The issue has been mitigated across all affected environments. We are currently deploying on a fix to prevent reoccurrence. We will provide another update once the fix has been deployed.
  5. 21:50 UTC · monitoring
    We have deployed and validated the fix to prevent immediate reoccurrence. We will be performing additional work to limit these kinds of failures in the future.
  6. 21:50 UTC · resolved
    On August 10, 2026, between 19:48 UTC and 20:49 UTC, GitHub Copilot users saw an incomplete list of available models. During this window, the service could return as few as one model instead of the full catalog. Requests that tried to use a model missing from that shortened list failed with a "model not found" error. Copilot requests that used an available model were not affected. This did not affect customers on data-residency (Proxima) environments.<br /><br />The issue was caused by a change to how model data was published, which our systems could not read back correctly and fell back to a limited default list.<br /><br />We mitigated the incident by 20:49 UTC and deployed a fix to prevent immediate recurrence by 21:50 UTC. We are adding validation and retry safeguards so that model data is verified before it is served.<br /><br />We apologize for the disruption.
Source
Aug 10, 202644mSEV 3
Disruption with creation of fine grained personal access tokensSev 3 from the incident title (no components listed)
5 updates
  1. 18:02 UTC · investigating
    We are investigating reports of impacted performance for some GitHub services.
  2. 18:09 UTC · investigating
    We are investigating reports of users being unable to create fine-grained Personal Access Tokens. Attempting to create a new token redirects the user back to the token overview page without an error message, but the token was not created.
  3. 18:21 UTC · monitoring
    The degradation has been mitigated. We are monitoring to ensure stability.
  4. 18:22 UTC · monitoring
    We identified the source of the issue affecting creation of fine-grained personal access tokens and have applied a mitigation. Users should now be able to create new fine-grained tokens successfully. We are continuing to monitor to confirm full recovery.
  5. 18:46 UTC · resolved
    On August 10, 2026, between 17:16 and 18:21 UTC, users were unable to create new fine-grained personal access tokens (FG PAT) through the GitHub website. When a user submitted the FG PAT creation form, they were returned to the FG PAT list without an error message and no FG PAT was created. Creating classic personal access tokens, as well as editing or deleting existing FG PAT were not affected.<br /><br />The cause was a change to how the website loads certain front-end JavaScript that was enabled for all users at 17:15 UTC; the change interacted with an issue in the token creation form's confirmation step that prevented it from running, so the final submission that actually creates the token never completed. Because the page still loaded and the server returned a normal response, the failure produced no error message. GitHub mitigated the incident by disabling the change at 18:21 UTC, at which point token creation recovered immediately, and the incident was resolved at 18:46 UTC.<br /><br />To reduce the chance of recurrence, GitHub is adding monitoring and alerting for anomalies in the FG PAT creation success rate and is removing the issue in the FG PAT creation form that prevented the confirmation step from running. GitHub is also adding automated detection of the issue so other areas of the GitHub front end do not repeat the problem.
Source

Peers, 90 days

Compare
Docker HubGrade A+100.00% up4 incidentsMTTR 11mnpmGrade D100.00% up9 incidentsMTTR 2h 15mGitHubGrade F99.87% up49 incidentsMTTR 1h 18m

README badge

GitHub's live grade, for your README.

GitHub grade badge
[![github on downagain.dev](https://downagain.dev/badge/github.svg)](https://downagain.dev/p/github)