Keepalived in a VM

Apps and Virtualization
,
1

I am curious if there would be any reason I could not set up keepalived in a Ubuntu 24.04 VM on scale?

I am currently attempting to set up a backup pihole server with a vip using keepalived and am getting an error when attempting to start the service after configuring it stating

Aug 16 21:42:28 pihole01 systemd[1]: Starting keepalived.service - Keepalive Daemon (LVS and VRRP)...
Aug 16 21:42:28 pihole01 Keepalived[37731]: Starting Keepalived v2.2.8 (04/04,2023), git commit v2.2.7-154-g292b299e+
Aug 16 21:42:28 pihole01 Keepalived[37731]: Running on Linux 6.8.0-40-generic #40-Ubuntu SMP PREEMPT_DYNAMIC Fri Jul  5 10:34:03 UTC 2024 (built for>
Aug 16 21:42:28 pihole01 Keepalived[37731]: Command line: '/usr/sbin/keepalived' '--dont-fork'
Aug 16 21:42:28 pihole01 Keepalived[37731]: Configuration file /etc/keepalived/keepalived.conf
Aug 16 21:42:28 pihole01 Keepalived[37731]: daemon is already running
Aug 16 21:42:28 pihole01 systemd[1]: keepalived.service: Failed with result 'protocol'.
Aug 16 21:42:28 pihole01 systemd[1]: Failed to start keepalived.service - Keepalive Daemon (LVS and VRRP).

I have set this up using 2 VMs in proxmox and everything worked but if I try this in Scale I get the above. The only differences I can think of are that the VM in proxmox are using a cloud-init image and the one on scale is using the normal ubuntu image and the networking in scale is also handled differently. The networking in Scale I have setup with a bridge and attached the VM to the bridge.

2

Same problem here with the exact same error messages. Trying to run keepalived directly on the TrueNAS Scale 25.04.2.3.

Here is the journalctl -xe log

Sep 11 10:53:25 truenas systemd[1]: keepalived.service: Failed with result ‘protocol’.
Subject: Unit failed
Defined-By: systemd
Support:
The unit keepalived.service has entered the ‘failed’ state with result ‘protocol’.
Sep 11 10:53:25 truenas systemd[1]: Failed to start keepalived.service - Keepalive Daemon (LVS and VRRP).
Subject: A start job for unit keepalived.service has failed
Defined-By: systemd
Support:

A start job for unit keepalived.service has finished with a failure.
The job identifier is 12110 and the job result is failed.

Any help would be appreciated.

3

Allowing Mac addresses to be changed? (“trust” something in the VM network settings).

The protocol assigns a specific Mac address to the “active” system.

4

Thanks for the response. For context, I tried to setup a loadbalacer for my piholes which are running on docker like described here: gist dot github dot com/Drallas/e61e4ebc1eaa2e52f21597fc8f2f43b8

I restarted the truenas and the error disappeared resulting with an empty keepalived.conf. Then I wrote the config again with the same content and it seemed to work. To prevent the service failure after a reboot I created a backup of the conf file and created a cron job which copies the keepalived.conf backup to the original conf and restart the service afterwards. Seems to work flawlessly so far.