(bugfix) Reconcile real rsync failures from worker logs

Record live rsync log paths for normal backup runs so the worker can
recover stale running state after terminal rsync errors.

Treat rsync vanished-file exit code 24 as a warning and keep the
completed snapshot instead of failing the run into incomplete state.

Closes #54
This commit is contained in:
2026-05-23 00:23:14 +02:00
parent 8633cbea26
commit 6eb1b4add3
4 changed files with 285 additions and 9 deletions

View File

@@ -8,7 +8,13 @@ from pathlib import Path
from django.db import transaction
from django.utils import timezone
from pobsync.commands.run_scheduled import DEFAULT_DRY_RUN_TIMEOUT_SECONDS, classify_rsync_failure, dry_run_log_path, run_scheduled
from pobsync.commands.run_scheduled import (
DEFAULT_DRY_RUN_TIMEOUT_SECONDS,
classify_rsync_failure,
classify_rsync_warning,
dry_run_log_path,
run_scheduled,
)
from pobsync_backend.config_source import DjangoConfigSource
from pobsync_backend.models import BackupRun, HostConfig
from pobsync_backend.retention import run_sql_retention_apply
@@ -66,6 +72,7 @@ def execute_backup_run(
run_id=run.id,
cancel_check=lambda: _run_cancel_requested(run.id),
verbose_output=bool(dry_run or verbose_output),
state_callback=lambda state: _record_running_state(run.id, state),
)
except Exception as exc:
run.refresh_from_db()
@@ -83,6 +90,8 @@ def execute_backup_run(
run.refresh_from_db()
if result.get("cancelled") or run.status == BackupRun.Status.CANCELLED:
run.status = BackupRun.Status.CANCELLED
elif result.get("status") == BackupRun.Status.WARNING:
run.status = BackupRun.Status.WARNING
else:
run.status = BackupRun.Status.SUCCESS if result.get("ok") else BackupRun.Status.FAILED
run.ended_at = timezone.now()
@@ -201,11 +210,71 @@ def _run_cancel_requested(run_id: int) -> bool:
return False
def _record_running_state(run_id: int, state: dict[str, object]) -> None:
try:
run = BackupRun.objects.only("id", "status", "result", "snapshot_path", "rsync_exit_code").get(id=run_id)
except BackupRun.DoesNotExist:
return
if run.status != BackupRun.Status.RUNNING:
return
result = run.result if isinstance(run.result, dict) else {}
execution = result.get("execution") if isinstance(result.get("execution"), dict) else {}
rsync = result.get("rsync") if isinstance(result.get("rsync"), dict) else {}
incoming_rsync = state.get("rsync") if isinstance(state.get("rsync"), dict) else {}
log_path = state.get("log")
snapshot_path = state.get("snapshot")
if isinstance(log_path, str) and log_path:
execution["log"] = log_path
if isinstance(snapshot_path, str) and snapshot_path:
execution["snapshot"] = snapshot_path
run.snapshot_path = snapshot_path
if incoming_rsync:
result["rsync"] = {**rsync, **incoming_rsync}
exit_code = incoming_rsync.get("exit_code")
if isinstance(exit_code, int):
run.rsync_exit_code = exit_code
result["execution"] = {
**execution,
"worker_pid": os.getpid(),
"worker_host": socket.gethostname(),
"heartbeat_at": timezone.now().isoformat(),
}
run.result = result
run.save(update_fields=["snapshot_path", "rsync_exit_code", "result"])
def _reconcile_running_run(*, run: BackupRun, grace_seconds: int, stale_worker_seconds: int) -> bool:
result = run.result if isinstance(run.result, dict) else {}
requested = result.get("requested") if isinstance(result.get("requested"), dict) else {}
log_path = _execution_log_path(result)
log_tail = _read_log_tail(log_path) if log_path is not None else []
terminal_log = _terminal_rsync_log(log_tail)
exit_code = _exit_code_from_log(log_tail)
stale_worker = _running_worker_timed_out(run=run, stale_worker_seconds=stale_worker_seconds)
if not requested.get("dry_run"):
if terminal_log:
failure = classify_rsync_failure(exit_code or 255, log_tail)
result.update(
{
"ok": False,
"host": run.host.host,
"log": str(log_path) if log_path else "",
"failure": failure,
"rsync": {
**(result.get("rsync") if isinstance(result.get("rsync"), dict) else {}),
"exit_code": exit_code or 255,
"log_tail": log_tail,
},
}
)
run.status = BackupRun.Status.FAILED
run.ended_at = timezone.now()
run.rsync_exit_code = exit_code or 255
run.result = result
run.save(update_fields=["status", "ended_at", "rsync_exit_code", "result"])
return True
if stale_worker:
result.update(
{
@@ -225,14 +294,11 @@ def _reconcile_running_run(*, run: BackupRun, grace_seconds: int, stale_worker_s
return True
return False
log_path = _execution_log_path(result)
log_tail = _read_log_tail(log_path) if log_path is not None else []
terminal_log = _terminal_rsync_log(log_tail)
timed_out = _running_dry_run_timed_out(run=run, grace_seconds=grace_seconds)
if not terminal_log and not timed_out and not stale_worker:
return False
exit_code = _exit_code_from_log(log_tail) or (124 if timed_out or stale_worker else 255)
exit_code = exit_code or (124 if timed_out or stale_worker else 255)
failure = classify_rsync_failure(exit_code, log_tail)
if stale_worker and not terminal_log:
failure = {
@@ -305,6 +371,9 @@ def _read_log_tail(log_path: Path | None, *, max_lines: int = 40) -> list[str]:
def _terminal_rsync_log(log_tail: list[str]) -> bool:
warning = classify_rsync_warning(_exit_code_from_log(log_tail), log_tail)
if warning is not None:
return False
return any(line.startswith("rsync error:") for line in log_tail)
@@ -312,6 +381,8 @@ def _exit_code_from_log(log_tail: list[str]) -> int | None:
for line in reversed(log_tail):
if "code 255" in line:
return 255
if "code 24" in line:
return 24
if "code 124" in line:
return 124
if "code 12" in line: