← board

origin/master has advanced 4 commit(s) since this sha. Re-verify at current HEAD before acting — the callback is tagged to the sha that was tested, which may no longer be the state of the tree.

regression: lib-test#src:test/lib_tls.pas red at 459e96f985d1 (auto-filed by twatch)

Repro

tools/testmgr.py --tier full --job 'lib-test#src:test/lib_tls.pas' at 459e96f985d1588fac20836b151341cf7e967a61

Range

bad 459e96f985d1, last good 137a182ad46a, 70 commit(s) in range — the watcher narrows this by idle bisect; check tstate/TSTATE.md for the current range.

Log tail

ok: /tmp/testmgr-scratch-1497482/lib_tls  [code=252676B  data=11228B  bss=43836B  procs=614]

Stub ticket: signal only. Track T agent (face 2) enriches or a dev track takes it from the repro line.

Triage 2026-08-16 — NOT REPRODUCIBLE natively at HEAD (Track A/P pass)

Built and ran at 341dbf99f on this box: compiles clean and prints all 14 =ok lines, exit 0. The watcher's own log tail shows a clean ok: compile line and no failing assertion, so the job failed at RUN time, not compile time.

lib_tls.pas carries a real loopback socket round-trip (bind/connect on a live port), which makes it the one test in this pair that can fail for reasons that are not the compiler — a busy port or a sandboxed network on the watcher host. Left open rather than resolved: needs a re-run on plexus (the failing host) to say transient vs host-specific. If it comes back green there, close it; if it stays red only on plexus, it is a Track T environment item, not a compiler bug.

Log