RPO Monitor Alert

[Matthias Radmüller]

Hello,

we're creating Backups from Several Hosts to the PBS:
- from Proxmox VE Containers
- from Standalone Clients

My goal is to get a notification when a client initiates no backup within a time-range or did not successfully finish a backup within a defined interval.

In Veeam we can set a RPO Monitor interval for a machine/host, that we get an alert when no backup has been created within a defined interval (like no backups since 2 days).
Also on backuppc we have a similar notification.

I think PBS has no builtin feature for this, so the only way to do this is to create a script what handle this type of notifications?

Or can someone suggest me a "nicer" way to do this? I would be happy also if I cloud setup a webhook on backup OK from the task-list.

 

[Onslow]

Not the exact solution, but you could set notifications for results other than OK.

In https://pve.proxmox.com/pve-docs/chapter-notifications.html see "Severity Matching Rules":

"A notification has a associated severity that can be matched.

• match-severity error: Only match errors
• match-severity warning, error: Match warnings and error

The following severities are in use:info, notice, warning, error, unknown."

- and examples below that.

 

[Matthias Radmüller]

Not the exact solution, but you could set notifications for results other than OK.

In https://pve.proxmox.com/pve-docs/chapter-notifications.html see "Severity Matching Rules":

"A notification has a associated severity that can be matched.

• match-severity error: Only match errors
• match-severity warning, error: Match warnings and error

The following severities are in use:info, notice, warning, error, unknown."

- and examples below that.

From the Backup-Client (PVE) I could do this (Mail or Post-Hook of vzdump).
Also from the Backup-Client on a Standalone Linux-Server I could verify the backup-script exit-code and trigger then a Mail / Message manually.

But I would like to have a solution to trigger these Events from the PBS to centrally manage them.

This only works when the Client what makes the Backup is healty, but If there is something wrong on the Client I would prefer that the backup server verify the RPO Interval.

My goal is to have something like I have with Veeam: A RPO Dashboard / Alert Tool where I can verify if a Backup Client is outside the RPO-Interval.

If PBS has the possibility to trigger a Hook / Mail after a Successful Backup (Task Backup OK) I could send this information to Nagios / Check_MK or healthchecks.io where I have an external RPO Monitor:



Currently this would be possible by periodically Parsing the Output of proxmox-backup-manager task list --all --output-format json-pretty, but i think there must be a cleaner / easier way?