#03:32:56+00:00 warn agent/embedded {"subsystem":"agent/embedded"} [timeout-compaction] LLM timed out

1 messages · Page 1 of 1 (latest)

warm kiln
#

03:32:56+00:00 warn agent/embedded {"subsystem":"agent/embedded"} [timeout-compaction] LLM timed out with high prompt token usage (84%); attempting compaction before retry (attempt 1/2) diagId=ovf-moi2mafp-CeM1iA

gusty thicketBOT
#

To help others find answers, you can mark your question as solved via Right click solution message -> Apps -> ✅ Mark Solution

next ridgeBOT
#

That warning by itself isn’t usually fatal — it means the model request timed out while the prompt was already huge (84% of context), so OpenClaw is trying an automatic compaction + retry.

Please share:

openclaw --version
openclaw status --all

And the follow-up log lines after that diagId, especially whether you see:

  • compaction succeeded ... retrying prompt
  • compaction did not reduce context
  • another timeout/failover error

You can grab recent logs with: