cloudstack: Virtual Router Failing Health Check Due To Missing Entries in /etc/hosts
ISSUE TYPE
- Bug Report
COMPONENT NAME
UI
CLOUDSTACK VERSION
4.18 But have noticed the issue in previous versions already.
CONFIGURATION
I have a very basic private cloud setup with Advanced Networking. No funny additional load balancers, VPCs etc just a basic set-up.
OS / ENVIRONMENT
Cloudstack Management Server Runs On AlmaLinux 8 and KVM hosts also on Almalinux 8
SUMMARY
After creating a new VM with a custom name, such as “Example-VM”, terminating the VM and Recreating it again using the same name again, causing the virtual router to stop updating the /etc/host entries and causing failed health checks. And now, it’s to a point where when I create new VMs I MUST always manually enter entires into the Virtual Routers /etc/hosts/ file it now no longer enters it automatically.
I have tried rebooting the router and nothing, I want to try patching it and possibly destroying it so its re-created again. This will most likely solve it, but no one wants to do that in production.
STEPS TO REPRODUCE
Create a few VMs with a custom name. Destroy it (expunge also) and recreate VM again using the SAME name, repeat this step three or four times. Following the same VM name, resources, host etc. Automatically assign IPs to VM.
EXPECTED RESULTS
The virtual router should always update entries accordingly in the /etc/hosts file.
About this issue
- Original URL
- State: closed
- Created a year ago
- Comments: 32 (15 by maintainers)
Some Voodoo is happening with my instance. I stopped the additional VR after rebooting CS Management. Then proceeded to destroy it. Rebooted CS again, did a network restart with clean up again on the old existing VR, and this time it completed successfully. I now have a new VR and everything in
/etc/hostsis showing.Going to create a few new VMs and see what happens and update here again. Will try to document my steps this time so I can see what exactly causes this behaviour.