Summary
The Node Core Subset Radar nightly (.github/workflows/node-core-subset.yml) has failed on every scheduled run for at least the last six nights. It dies in the Run Node core subset radar step with exit code 143 (SIGTERM / timeout) after ~1h15m, and produces no report artifact.
2026-07-12 failure
2026-07-11 failure
2026-07-10 failure
2026-07-09 failure
2026-07-08 failure
2026-07-07 failure
gh run download <id> -n node-core-subset-report -> "no valid artifacts found".
Impact
There is currently no parity number for node:http/https. The most recent figure on record is the 28.0% in #4975's own comment thread — that issue's central metric has been unmeasurable for over a week, so we cannot tell whether the tail is improving or regressing.
Ask
Either raise the step timeout / shard the radar so it completes, or make the job fail loudly (and page) when it times out, rather than silently producing no artifact.
Summary
The Node Core Subset Radar nightly (
.github/workflows/node-core-subset.yml) has failed on every scheduled run for at least the last six nights. It dies in theRun Node core subset radarstep with exit code 143 (SIGTERM / timeout) after ~1h15m, and produces no report artifact.gh run download <id> -n node-core-subset-report-> "no valid artifacts found".Impact
There is currently no parity number for node:http/https. The most recent figure on record is the 28.0% in #4975's own comment thread — that issue's central metric has been unmeasurable for over a week, so we cannot tell whether the tail is improving or regressing.
Ask
Either raise the step timeout / shard the radar so it completes, or make the job fail loudly (and page) when it times out, rather than silently producing no artifact.