Skip to main content
DigitalOcean

Serverless Inference - High error rates for open source models ( Qwen 3 32B)

Degraded Lasted 3h 1m · started · DigitalOcean's page

Level
confirmed
Worst impact
Degraded
Components
Agent Runtime
Down Index
20of 100

What to do

  • Check whether you use DigitalOcean Agent Runtime. 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 3h 1m it lasted.

    Anything that failed between 2026-04-07 12:49 UTC and 2026-04-07 15:50 UTC may need a replay.
  • Subscribe to DigitalOcean's feed to hear about the next one first. RSS feed

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

Updates from DigitalOcean 3

  1. investigating

    Serverless inference for alibaba-qwen3-32b (Qwen 3 32B) in tor1 is experiencing high error rates starting at 10:46 UTC.

    provider

  2. identified

    We are currently investigating reports of elevated latency affecting requests to this model when using Serverless Inference and Agents. Earlier observations indicated increased error rates for the open-source Qwen 3 32B model. The Ray dashboard also showed multiple workers in a pending state, suggesting capacity constraints. Our analysis determined that the model was experiencing higher-than-expected request volume without sufficient resources to scale accordingly. To address this, the node pool size has been increased to improve available capacity. However, there are still insufficient nodes to fully support the desired number of model replicas. Following the node pool expansion, a new pod-related error has been identified. Our Engineering team is actively working to resolve this issue and restore full service performance.

    provider

  3. resolved

    Service has been fully restored, and the model is now operating normally. We have implemented improvements to enhance stability and reduce the likelihood of similar issues in the future.

    provider

Reports count in aggregate; nothing about you is shown.

Corrections

None so far.

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

Questions about DWN-b2ab12c268

Is DigitalOcean still down?

No, this incident was resolved 2026-04-07 15:50 UTC. For DigitalOcean's live state, see https://down.guide/digitalocean.

How long did it last?

3h 1m, from 2026-04-07 12:49 UTC to 2026-04-07 15:50 UTC, by the times DigitalOcean posted.

What was affected?

DigitalOcean named Agent Runtime.

What caused it?

DigitalOcean hasn't published a root cause. If it does, it shows here.

Where does this come from?

DigitalOcean's official status page, https://status.digitalocean.com, quoted verbatim. down.guide's own probes, reports and social counts are labeled as ours.