Struck in red is what Bitbucket removed; blue is what they added. Muted text is unchanged context.
BeforeAfter
ChangedStatus#1
Before: resolved
After: postmortem
AddedUpdatepostmortem#2
Before: not present
After: SUMMARY
On October 15, 2026, between 9:38 and 10:14 UTC, Bitbucket Cloud customers were unable to access the web interface, API, and Pipelines. Git CLI operations over HTTPS or SSH were unaffected. The incident was triggered by a deployment that did not scale quickly enough to handle peak traffic, compounded by a capacity limit in our feature flag evaluation. Automated monitoring detected the issue within 2 minutes. We restored the web interface and API by rolling back to the previous deployment and scaling it up. Pipelines performance stayed degraded until 11:54 UTC, when all services were fully recovered.
ROOT CAUSE
The trigger for this incident was a hot fix designed to mitigate a previous performance issue that was deployed using an accelerated rollout. The website auto-scaling configuration had not been tuned for a recent infrastructure change and therefore was under provisioned following the hotfix deployment. During the incident the newly deployed web and API services could not absorb traffic, and began failing. After the rollback, a surge of queued events overloaded Pipelines, delaying the processing of Pipelines jobs.
REMEDIAL ACTIONS PLAN & NEXT STEPS
We know that outages impact your productivity. We are prioritizing the following improvement actions to avoid repeating this type of incident:
Update scaling strategy: Update scaling thresholds and concurrency settings so website capacity is added earlier under load during hotfix deployments.
Improve progressive rollouts: Add smaller rollout increments and soak periods, and pre-scale new deployments for expected traffic growth.
Improve feature flag caching mechanisms: Adjust caching timeouts and cold-cache behavior to improve feature flag cache hits.
We apologize to customers whose services were impacted during this incident; we are taking immediate steps to improve the platform’s performance and availability.
Thanks,
Atlassian Customer Support
Rewindex commits each version of an incident it captures to a daily checkpoint, chains it to the day before, and has the checkpoint timestamped by FreeTSA and DigiCert. That proves these records existed by then and have not been changed since, by anyone, Rewindex included.
Bitbucket Cloud reported a critical disruption affecting API, Pipelines and Website, with connection errors, timeouts and access failures. It started at 2026-10-15 10:07 UTC and resolved at 2026-10-15 12:08 UTC (2 h 0 min); a fix restored core services while Pipelines cleared backlogged builds.
Cite this
Incident updated: Bitbucket Cloud - Service Disruption on Bitbucket’s status page, changed 2026-10-28 04:33 UTC (change 2 of 2).
Title
Incident updated: Bitbucket Cloud - Service Disruption