Imgix incident: Delay in Purge requests

Affected components

• Purging

1. resolvedDec 16, 2024, 13:03 UTC

   The purge queue and purge time has been stable since the last provided update.

2. monitoringDec 16, 2024, 13:02 UTC

   We are continuing to monitor for any further issues.

3. monitoringDec 16, 2024, 09:02 UTC

   A fix has been implemented and new purge requests should again complete within expected times. We are continuing to monitor the situation.

4. identifiedDec 16, 2024, 07:01 UTC

   Purge requests for images that use any of the below AI-parameters have been restored, but we are still observing longer purge times with a high queue. - bg-replace - compositions (any two or more ai features composed together, like bg-remove=true&upscale=true) We are looking at different approaches and will update accordingly.

5. identifiedDec 16, 2024, 06:42 UTC

   The issue has been identified and we are working on a fix. To speed up processing of queued purge requests, we will have to temporarily disable purge requests for images that use any of the following AI-parameters: - bg-replace - compositions (any two or more ai features composed together, like bg-remove=true&upscale=true)

6. investigatingDec 16, 2024, 06:21 UTC

   We have identified an issue with our purging service, causing new purge requests to be delayed in execution. (10mins+) Rendering services remain unaffected.

What APIStatus.watch saw

Our 5-minute checks didn't record a change in Imgix's overall status around this incident. Smaller or regional incidents often leave a provider's overall status green.

Other services with incidents at the same time

Overlapping incidents aren't necessarily related.

• Mapbox: Delayed delivery of Mapbox Traffic DataMinor · Dec 10, 2024, 17:37 UTC

How often does Imgix have incidents?

Imgix reported 3 incidents in the last 90 days, 2 of them major or critical. A typical incident lasted 51 min. See Imgix's uptime and incident history