Lokalise application and API are unavailable

What to do

• Check whether you use Lokalise Lokalise API, Lokalise App in us. Check your stack

  Paste a package.json, lockfile, Terraform or a list of domains at down.guide/stack to see every provider you depend on.

• Confirm your retries, queues and scheduled jobs caught up after the 48m it lasted.

  Anything that failed between 2026-02-04 16:18 UTC and 2026-02-04 17:07 UTC may need a replay.

• Subscribe to Lokalise's feed to hear about the next one first. RSS feed

  https://down.guide/feed?provider=lokalise works in Slack (/feed subscribe), Discord feed bots and any RSS reader.

Updates from Lokalise 5

1. investigatingFeb 4, 2026, 4:18 PM UTC

   We are currently investigating this issue.

   provider

2. identifiedFeb 4, 2026, 4:30 PM UTC

   The issue has been identified and a fix is being implemented. Application and API are now operational with degraded performance.

   provider

3. monitoringFeb 4, 2026, 4:59 PM UTC

   A fix has been implemented and we are monitoring the results.

   provider

4. resolvedFeb 4, 2026, 5:07 PM UTC

   This incident has been resolved.

   provider

5. postmortemFeb 13, 2026, 3:49 PM UTC

   On February 4, 2026, Lokalise experienced a brief service interruption where both the application and API were unavailable for approximately 10 minutes. The incident began at 16:18 UTC and was fully resolved within 12 minutes as our engineering team restored service functionality. What happened? The Cause: A configuration update intended to enable new application metrics unexpectedly triggered a series of intensive, unoptimized database scans. This resulted in an immediate exhaustion of our backend processing capacity, preventing the web server from fulfilling incoming requests. The Fix: Our team identified the faulty configuration release and promptly initiated a rollback to the last known stable state. Once service was restored, a permanent fix was deployed to remove the problematic configuration entirely. Timeline \(UTC\): * 16:18: Automated alerts reported service unavailability; investigation initiated. * 16:20: The root cause was identified as a recent configuration deployment. * 16:27: Rollback to a stable version was initiated. * 16:30: Application and API functionality were fully restored. * 17:07: System verified as stable after a final hotfix deployment. What we are doing to prevent this in the future * Enhancing pre-deployment validation: We are updating our automated review agents to flag potential performance issues in legacy code or dormant features before they are activated via configuration changes. * Optimizing metric design: We are transitioning to a system that pre-computes application metrics in isolated processes to ensure that monitoring activities never compete with core application resources. * Refactoring database interactions: Our team is reviewing and refactoring resource-intensive database queries to improve overall platform stability and response times. * Consolidating internal alerting: We are refining our monitoring workflows to consolidate overlapping alerts, allowing our engineers to focus even more quickly on resolution during critical events. We sincerely apologize for the disruption this incident caused to your work and your automated workflows. We recognize the trust you place in Lokalise for your daily operations, and we are committed to enhancing our processes to prevent similar occurrences. If you have any questions or require further assistance regarding this incident, please reach out to us at [hello@lokalise.com](mailto:hello@lokalise.com).

   provider

Corrections

None so far.

Something wrong on this page? Report an error or send us a note. Every fix is listed here.