Codacy incident

Major outage

Minor Resolved View vendor source →

Codacy experienced a minor incident on July 1, 2026, lasting 1h 47m. The incident has been resolved; the full update timeline is below.

Started
Jul 01, 2026, 09:08 AM UTC
Resolved
Jul 01, 2026, 10:56 AM UTC
Duration
1h 47m
Detected by Pingoru
Jul 01, 2026, 09:08 AM UTC

Update timeline

  1. monitoring Jul 01, 2026, 09:08 AM UTC

    Jul 1, 2026, 09:08 UTC Investigating - We are investigating a problem in our databases that is causing Codacy to become non responsive.

  2. investigating Jul 01, 2026, 09:08 AM UTC

    We are investigating a problem in our databases that is causing Codacy to become non responsive.

  3. investigating Jul 01, 2026, 09:34 AM UTC

    We are seeing recovery in our services but are still working to fully identify the root cause of the issue and make sure our mitigation measures hold.

  4. monitoring Jul 01, 2026, 09:43 AM UTC

    We are observing positive signs as we bring all services back up. There is a backlog of analysis jobs that will take longer than usual to process but signs are positive. We will keep on updating as the recovery progresses. Right now all services should be available but still not up to full performance.

  5. monitoring Jul 01, 2026, 09:58 AM UTC

    Systems are back to normal capacity but delays in analysis are to be expected for a bit longer while we process backlog of analysis that were stopped during the duration of the outage.

  6. resolved Jul 01, 2026, 10:56 AM UTC

    All systems are now operating at normal performance, and the analysis queue backlog has been fully consumed. Root cause: This incident was caused by an unplanned DDL (schema) change applied in production, which triggered cascading lock contention on a database table that is central to our platform's operation. Because of this table's role across multiple services, the impact extended well beyond the original change and affected system-wide performance. Next steps: We are hardening our deployment processes to ensure DDL operations cannot be run against production databases in ways that risk service disruption, particularly during peak hours. This includes stricter guardrails around when and how schema changes are applied. We apologize for the disruption this caused and appreciate your patience while we resolved it.