RunBook AcademyProxmox VE

← All checklists in Proxmox VE

WeeklyMaintenance

Weekly operational review

13 items ·5 critical ·5 warn ·3 info

Weekly operational review

A 30-60 minute review of cluster health and operational metrics. Performed by the on-call engineer or a designated reviewer.

Goals

Catch small problems before they become incidents
Validate backup and DR procedures
Identify trends that need attention (capacity, performance)

What to look for

Backup success rate: anything below 95% needs investigation
Disk space: anything above 80% needs a forecast; above 90% is urgent
Pending updates: prioritise security, schedule restarts
Failed login patterns: spikes may indicate attack or misconfiguration

Critical5 items

Backup success rate over the past 7 days (> 95% expected)

pvesh get /cluster/tasks --typefilter vzdump --since "7 days ago" --output-format json | jq -r ".data[] | \\"\\(.id) \\(.status)\\"" | tail -20

Offsite PBS sync completed successfully each day
```
proxmox-backup-manager sync job list
```
ZFS pool health (all pools ONLINE, no resilvers)
```
zpool list && zpool status
```
Ceph HEALTH_OK (if applicable)
```
ceph -s
```
Cluster quorum healthy (all nodes online)
```
pvecm status
```

Warning5 items

At least one restore test performed in the past 30 days

Disk space warnings (any storage > 80% used)

pvesh get /cluster/resources --type storage --output-format json | jq ".[] | {storage, used: .used, total: .total}"

Pending security updates reviewed
```
apt list --upgradable
```

TLS certificates valid for > 30 days

openssl x509 -in /etc/pve/local/pveproxy-ssl.pem -noout -dates

Replication lag within SLA
```
pvesr status
```

Info3 items

HA-managed VMs in expected state
```
ha-manager status
```
Review failed SSH login attempts (any unusual pattern)
```
lastb | head -20
```
Backup size trend compared to previous week (catch runaway growth)