CoolBoy/submmit
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

fix: keep agent alive across worker crashes

Browse Source
This commit is contained in:
CoolBoy
2026-08-21 03:38:06 +08:00
parent 85f6cb5157
commit 6eb7ded984
7 changed files with 183 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
modelhub_submmit_api/state_sync.py
View File

@@ -42,6 +42,7 @@ STATE_ALLOWLIST = (
".modelhub_state/recovery_intents.jsonl",
".modelhub_state/routing_intelligence.json",
".modelhub_state/submission_exclusions.jsonl",
".modelhub_state/worker_crashes.jsonl",
"ledger/submissions.jsonl",
"outcomes/submissions.jsonl",
)
@@ -276,7 +277,15 @@ class StateGitSync:
destination = self.project_root / relative
destination.parent.mkdir(parents=True, exist_ok=True)
temporary = destination.with_name(f".{destination.name}.restore-{uuid.uuid4().hex}")
shutil.copy2(source, temporary)
if relative == ".modelhub_state/worker_crashes.jsonl":
merged: dict[str, dict[str, Any]] = {}
for row in [*read_jsonl(source), *read_jsonl(destination)]:
key = json.dumps(row, ensure_ascii=False, sort_keys=True)
merged[key] = row
rows = sorted(merged.values(), key=lambda row: str(row.get("at") or ""))[-200:]
write_jsonl(temporary, rows)
else:
shutil.copy2(source, temporary)
os.replace(temporary, destination)
self.log(
f"[state-recovery] generation={self.generation} source=remote branch={self.branch} status=ok"