viktor/infra
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

monitoring: fix PrometheusBackupStale false-fire (32d->40d threshold)

Browse source 
The prometheus-backup sidecar runs monthly on the 1st SUNDAY 04:00 UTC.
Consecutive first-Sundays can be ~35 days apart (e.g. May 3 -> Jun 7), but
the alert threshold was 32d (2764800s) -> it false-fired every year for the
~3 days between day-32 and the next run. Raised to 40d (3456000s): clears
the max first-Sunday spacing with margin, still catches a genuinely missed
monthly backup. Backup itself is healthy (last May 3, next Jun 7). Verified:
live rule now > 3.456e6, alert state inactive.
This commit is contained in:
Viktor Barzin 2026-06-04 08:07:58 +00:00
parent c4bd64f88a
commit 63ee655c08
1 changed files with 8 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

10
stacks/monitoring/modules/monitoring/prometheus_chart_values.tpl
View file

@ -1554,12 +1554,18 @@ serverFiles:
annotations:
summary: "Redis backup CronJob has never completed successfully"
- alert: PrometheusBackupStale
expr: (time() - prometheus_backup_last_success_timestamp{job="prometheus-backup"}) > 2764800
# The backup sidecar runs monthly on the 1st SUNDAY 04:00 UTC.
# Consecutive first-Sundays can be up to ~35-37 days apart (e.g.
# May 3 → Jun 7 = 35d), so a 32d threshold false-fired every year
# in the gap before the next run. 40d (3456000s) clears the max
# first-Sunday spacing with margin while still catching a genuinely
# missed monthly backup.
expr: (time() - prometheus_backup_last_success_timestamp{job="prometheus-backup"}) > 3456000
for: 30m
labels:
severity: critical
annotations:
summary: "Prometheus backup is {{ $value | humanizeDuration }} old (threshold: 32d)"
summary: "Prometheus backup is {{ $value | humanizeDuration }} old (threshold: 40d)"
- alert: PrometheusBackupNeverRun
expr: absent(prometheus_backup_last_success_timestamp{job="prometheus-backup"})
for: 32d