VR suddenly was created with DHCP Service Failing. Can we Implement an auto-recovery?

[btzq]

problem

One of our users created a VPC and was complaining about how they were not able to get a private IP.

When we checked, turns out dhcp service was down. This surprised us because usually the VR is created without issues, and suddenly this happens.

Attached screenshot below:

Questions:

• Why is it that suddenly a service fails when a new VR is created, when we usually dont have issues with VR creation all this while?
• When VR detects a service is down, can it also restart that specific service in the VR without having to reboot or cleanup the VR? This would be more automatic and less downtime/issues.

versions

Cloudstack Version: 4.19.1.1
Storage: Linstor DRBD
Architecture: Disaggregated
Hypervisor: Ubuntu

The steps to reproduce the bug

Cant give a step. Just happens randomly when a new VPC is created.

What to do about it?

When VR detects a service is down, auto estart that specific service in the VR without having to reboot or cleanup the VR? This would be more automatic and less downtime/issues.

[DaanHoogland]

@btzq A monitorring service could catch that event and issue a restart with cleanup command. I don’t think you need this implemented in cloudstack, do you?