Oso incident

Increased latency of write application for some customers

Minor Resolved View vendor source →

Oso experienced a minor incident on September 22, 2025 affecting Authorization APIs, lasting 5h 14m. The incident has been resolved; the full update timeline is below.

Started
Sep 22, 2025, 04:57 PM UTC
Resolved
Sep 22, 2025, 10:11 PM UTC
Duration
5h 14m
Detected by Pingoru
Sep 22, 2025, 04:57 PM UTC

Affected components

Authorization APIs

Update timeline

  1. monitoring Sep 22, 2025, 04:57 PM UTC

    We have identified an increase in delay in write processing for some customers. We have identified the source to be a backlog of writes in our processing queue, and a sequential bottleneck for some environments. The backlog of writes has completed processing and we are monitoring the replication lag as it returns to normal levels.

  2. monitoring Sep 22, 2025, 06:01 PM UTC

    We are continuing to monitor the replication recovery. Some affected customers should be seeing their workloads returning to normal, others should continue to see a reduction in write delay.

  3. monitoring Sep 22, 2025, 07:37 PM UTC

    Over the next hour environments should see gradual recovery in write delays, with full recovery estimated in 2 hours.

  4. resolved Sep 22, 2025, 10:11 PM UTC

    Write latencies to all affected Oso Cloud environments have recovered.

  5. postmortem Sep 24, 2025, 02:40 PM UTC

    Link to postmortem: [https://osohq.notion.site/2025-09-22-Write-delays-for-certain-customers-2769f1471f2b80e3b36fd94ada7f34ce](https://osohq.notion.site/2025-09-22-Write-delays-for-certain-customers-2769f1471f2b80e3b36fd94ada7f34ce)