hermes-admin/hermes-agent
1
0
Fork 0
Code Issues Pull Requests Actions 33 Packages Projects Releases Wiki Activity

fix(tests): compression stall-fallback tests stop racing the 0.2s ceiling

Browse Source 
Under a loaded runner the primary stall plus the fallback retry overran the
0.2s total ceiling, so the retry never started and attempts==1 failed
intermittently (seen once in a 40-worker tests/agent run). Idle stays 0.05s;
the ceiling moves to 2s per the >=2s wall-clock rule in AGENTS.md.
This commit is contained in:
teknium1
2026-09-14 07:12:18 -07:00
committed by Teknium
parent d28938d3da
commit d57c28a554
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tests/agent/test_compression_stall_fallback.py
View File

@@ -90,7 +90,7 @@ class _StalledSummaryWorker:
fence.finish_commit()
def _run(worker, *, chain, timeouts, messages, idle=0.05, ceiling=0.2):
def _run(worker, *, chain, timeouts, messages, idle=0.05, ceiling=2.0):
with _patch_chain(chain):
return run_compress_context_with_progress_timeout(
worker=worker,
@@ -152,7 +152,7 @@ def test_retry_runs_on_a_host_published_fence():
messages=original,
system_prompt_fallback="degraded-prompt",
idle_timeout_seconds=0.05,
total_ceiling_seconds=0.2,
total_ceiling_seconds=2.0,
new_fence=_new_fence,
)
finally:
@@ -182,7 +182,7 @@ def test_hard_interrupt_suppresses_the_fallback_attempt():
messages=original,
system_prompt_fallback="degraded-prompt",
idle_timeout_seconds=0.05,
total_ceiling_seconds=0.2,
total_ceiling_seconds=2.0,
on_timeout=lambda *args: timeouts.append(args),
telemetry_agent=agent,
)