Degraded performance of API operations due to slow database operations

What to do

• Check whether you use Apify API (api.apify.com), Actors. 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 11h 49m it lasted.

  Anything that failed between 2026-07-21 14:49 UTC and 2026-07-22 02:38 UTC may need a replay.

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

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

Updates from Apify 10

1. identifiedJul 21, 2026, 2:49 PM UTC

   We are currently investigating timing out operations on our API caused by slow database operations due to cluster changes.

   provider

2. identifiedJul 21, 2026, 4:43 PM UTC

   Slow database operations continue to impact our systems. API performance has improved, but Actor runs are currently timing out due to reduced worker capacity.

   provider

3. identifiedJul 21, 2026, 5:09 PM UTC

   We are continuing to work on a fix for this issue.

   provider

4. monitoringJul 21, 2026, 5:11 PM UTC

   We are now monitoring the situation. Worker capacity improved and is now able to handle starting Actor runs.

   provider

5. investigatingJul 21, 2026, 5:47 PM UTC

   Our database cluster performance has further degraded, and the whole platform is facing a major outage. We will keep informing you as the situation progresses.

   provider

6. identifiedJul 21, 2026, 8:33 PM UTC

   We have identified the issue and are working towards recovery. The system is operational, but the capacity is capped, resulting in Actor runs being buffered in the "ready" state and a large portion of them timing out.

   provider

7. identifiedJul 21, 2026, 8:34 PM UTC

   We are continuing to work on a fix for this issue.

   provider

8. identifiedJul 21, 2026, 8:52 PM UTC

   Actors are now fully operational, and the system is processing all the runs in time.

   provider

9. monitoringJul 21, 2026, 9:28 PM UTC

   All systems are operational. We expect the database to fully recover in 4 hours (1 am UTC). We are monitoring the platform's performance. We apologize for any inconvenience this might have caused.

   provider

10. resolvedJul 22, 2026, 2:38 AM UTC

    The database is fully recovered, and all systems are operational. We apologize for any inconvenience this might have caused. We are currently implementing new measures to ensure this won't happen again.

    provider

Corrections

None so far.

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