Skip to content
dbt Labs · Data and observabilityJul 21, 2026, 10:23 UTC

Job runs failing at the git clone step

CriticalDeploymentUpdated 15h ago
Jul 21, 10:23 UTCJul 21, 12:52 UTC
Duration
2h 29m
Impact
Critical
Root cause
Deployment
dbt Labs, 90 days
26 incidents
Affected
Not listed by the vendor.
Status page

Final update

This incident is resolved. Scheduled, CI, and API-triggered job runs that clone from GitHub were failing at the "Clone git repository" step due to an upstream GitHub incident affecting SSH deploy key authentication. GitHub has resolved the underlying issue. Git clone error rates on the dbt platform have returned to normal and remained stable. Job runs are operating as expected, and any runs that failed during the incident can be safely retried. We apologize for the disruption and thank you for your patience.

Timeline

  1. Resolved · Jul 21, 12:52 UTC
    This incident is resolved. Scheduled, CI, and API-triggered job runs that clone from GitHub were failing at the "Clone git repository" step due to an upstream GitHub incident affecting SSH deploy key authentication. GitHub has resolved the underlying issue. Git clone error rates on the dbt platform have returned to normal and remained stable. Job runs are operating as expected, and any runs that failed during the incident can be safely retried. We apologize for the disruption and thank you for your patience.
  2. Monitoring · Jul 21, 11:45 UTC
    GitHub has resolved the upstream incident affecting SSH deploy key authentication. As of approximately 7:45 AM ET (11:45 UTC), git clone error rates have returned to pre-incident levels and job runs are completing normally again. Any runs that failed during the incident can now be retried. We are continuing to monitor to confirm full and sustained recovery before marking this incident resolved.
  3. Identified · Jul 21, 10:31 UTC
    We have identified the cause as an upstream GitHub incident affecting SSH authentication for **deploy keys**. Because the dbt platform uses deploy keys to clone Git repositories, scheduled, CI, and API-triggered runs that clone from **GitHub** are failing at the "Clone git repository" step with a `Permission denied (publickey)` error. Runs that connect to other providers (for example GitLab or Azure DevOps) are **not** affected. This is being addressed on GitHub's side. You can follow their progress here: We are monitoring closely and will provide an update as soon as GitHub's mitigation takes effect or we have more information.
  4. Investigating · Jul 21, 07:45 UTC
    We're investigating an issue with **repository cloning** that is causing **scheduled, CI, and API-triggered job runs to fail at the "Clone git repository" step**. This is impacting **job runs across all regions (multi-tenant and single-tenant)** anytime after **approximately 3:40 AM ET (07:40 UTC)**. The team is working on a resolution and we will provide updates at approximately 15 minute intervals or as soon as new information becomes available.

More from dbt Labs

Full history
StartedIncidentDuration
Sep 2322:59 UTCIntermittent dbt deps failures4m
Sep 1823:31 UTCSchema Hydration Panic14h 58m
Sep 1014:28 UTCIntermittent errors on the Discovery API2h 17m
Sep 1013:46 UTCIncreased error rates associated with dbt MCP and dbt Wizard41m
Sep 1008:02 UTCState usage metrics not being ingested1h 13m
Sep 818:44 UTCUser Login Sessions Terminated4h 32m

Also caused by deployment or rollout

All
StartedIncidentDuration
Sep 1106:22 UTCIssues with Workers VPC hostname route resolution on 2026-09-11Cloudflare0m
Aug 2014:43 UTCIntermittent failures creating agent tasksGitHub9h 54m
Aug 2000:31 UTC[Medium] Issue with Microsoft Office Integration and Box EditBox1h 13m
Aug 622:22 UTCTrouble Using Search Bar For Some AdminsSlack2h 44m
Jul 200:17 UTCTraces, spans, logs inaccesible for query or ingestion in de and usSentry3h 39m
Jun 2915:28 UTCElevated error rates and latency for EAS Build, Workflows, and SubmitExpo14m

From vendors' own status pages and disclosures. Times as reported. Logos via logo.dev; trademarks belong to their owners.

Weekly: the week's major outages, postmortems and breaches, Saturday mornings.