chore(ci): run memory soak for 4 hours#4512
Conversation
- Bump DURATION_SECONDS 3600 -> 14400 in both soak jobs - Bump soak job timeout-minutes 90 -> 300 to cover the longer sample (cherry picked from commit ee3d911)
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize the Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counterproductive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here. |
- Cut per-view container log volume that fills the runner disk and kills the host ~3.5h into the 4h soak - Set RUST_LOG=error in both memory-soak job env blocks
- Sample the soak in 1h chunks and re-upload (overwrite) after each so partial data survives a host-level OOM/kill, where if:always() steps never run - soak.py: add --append (SOAK_APPEND) to accumulate chunks into one JSONL set instead of truncating - Set ESPRESSO_NODE_TELEMETRY_LOG=error in the soak job; the OTLP log layer has its own filter (default warn) that RUST_LOG does not reach, and node-0 pushes those records to an unreachable endpoint - docker-compose: make ESPRESSO_NODE_TELEMETRY_LOG overridable, default unchanged (info)
1 participant
(cherry picked from commit ee3d911)