Backups not starting automatically

[permport]

Hello ProxMox fanatics ;-)

2 weeks ago we updated 1 node from 6.xx to 7.xx. SHTF in a way that we don't want to update our 2 other nodes at this moment. We only have a prod cluster and don't take that risk anymore. So we have 1 node on 7.4 and 2 nodes on 6.xx
In the future we are going to buy new hardware and a enterprise subscription for ProxMox. This project is already started. Until then we leave our VE like it is.

Our problem: since this update our backup-tasks are dead. It doesn't start by the scheduler. However I can start the BU manually and then it runs.

I've found several topics concerning a similar issue but I can't solve it.

Can you analyze it with me?

On node 1 I have this:

 systemctl status pvescheduler.service
● pvescheduler.service - Proxmox VE scheduler
     Loaded: loaded (/lib/systemd/system/pvescheduler.service; enabled; vendor preset>
     Active: active (running) since Tue 2024-09-17 17:21:24 CEST; 1 weeks 1 days ago
    Process: 3796183 ExecReload=/usr/bin/pvescheduler restart (code=exited, status=0/>
   Main PID: 35690 (pvescheduler)
      Tasks: 1 (limit: 153963)
     Memory: 101.5M
        CPU: 5min 11.792s
     CGroup: /system.slice/pvescheduler.service
             └─35690 pvescheduler

On node 2 I have this:

root@pmnode2:~# systemctl status pvescheduler.service
Unit pvescheduler.service could not be found.

We don't have VM's on node 1 but it seems that the BU would be OK.

Thank you.

 

[Pifouney]

Hey,
if backp task are launched localy on your new node, delete and recreate the backup task.
If she's on pve 6.xx node, create separates task between pve 6 and 7 for your backups tasks

 

[permport]

Thank you for the reply.

I already had created a new task for node 2 but it doesn't run either

 

[permport]

I could solve it by creating the backup jobs on the the nodes. So connecting to the several nodes and then creating the jobs.

 

[VictorSTS]

So we have 1 node on 7.4 and 2 nodes on 6.xx
In the future we are going to buy new hardware and a enterprise subscription for ProxMox. This project is already started. Until then we leave our VE like it is.

This is a very bad idea and not supported at all. You should upgrade all nodes to 7.4 asap and keep all nodes with the same version to have a reliable cluster. It's a "riskier" to keep different versions than to upgrade all nodes.