[SOLVED] 2nd node backup in 4.3 cluster are not executed

[nowrap]

Hello,
i've 2 proxmox 4.3-9 cluster setups with 2 servers each.
Everything is working as expected.

Only the via web gui configured backups on the 2nd node are not executed on each cluster.
I could not pinpoint the problem yet.

The /etc/pve/vzdump.cron files look similar. Here is one example:

# cluster wide vzdump cron schedule
# Automatically generated file - do not edit

PATH="/usr/sbin:/usr/bin:/sbin:/bin"

0 1 * * 1,3,5,7     root vzdump --storage backup --compress lzo --all 1 --mode snapshot --mailnotification always --mailto ***@***.de --quiet 1 --node host1
0 1 * * 2,4,6       root vzdump --storage backup --compress lzo --all 1 --mode snapshot --mailnotification always --mailto ***@***.de --quiet 1 --node host2

The /etc/pve/storage.cfg looks like:

lvm: vg0
        vgname vg0
        content images

dir: local
        path /var/lib/vz
        maxfiles 0
        content iso,images,rootdir,vztmpl

dir: backup
        path /mnt/backup_***
        maxfiles 4
        shared 0
        content backup

/mnt/backup_*** exists on every node on both clusters.

Running the the command for the 2nd node directly in the shell, just finishes quite fast without an output:

root@host1:/root > vzdump --compress lzo --storage backup --all 1 --mode snapshot --mailnotification always --mailto ***@***.de --quiet 1 --node host2
root@host1:/root >

What is wrong with my config? Or how can i debug this?

Regards
nowrap

 

[fabian]

Running the the command for the 2nd node directly in the shell, just finishes quite fast without an output:

root@host1:/root > vzdump --compress lzo --storage backup --all 1 --mode snapshot --mailnotification always --mailto ***@***.de --quiet 1 --node host2
root@host1:/root >

What is wrong with my config? Or how can i debug this?

this is to be expected, the "-node" option tells vzdump to only run if executed on that node (this is needed because the cron jobs are defined on all nodes and synchronized via /etc/pve, but should only be executed on the correct node).

 

[nowrap]

Ok! I guess, this was too obvious for me ...

I started the command now on host2 and it is still running. So far so good.

But why is the cron not executed automatically?
And how can i debug this?

On host2 the vzdump.cron is identically. So it is correctly synchronised.

 

[fabian]

are you sure it is not executed? you have different weekdays configured for both nodes

you can also double check that the symlink in /etc/cron.d exists on node2 ("/etc/cron.d/vzdump" should point to "/etc/pve/vzdump.cron")

 

[nowrap]

I guess, you are right again!

The symlinks exists:
vzdump -> /etc/pve/vzdump.cron

Scrolling down and searching the task grid in the web gui reveals the execution.
My fault ;(
So i've to check, my the reporting mail wasn't send or received.
And why i thought the content wasn't rising.

Sorry, for bothering you