Hi Team,
I am facing an issue while trying to upgrade the System Offering for my
Console Proxy and Secondary Storage VMs in Apache CloudStack.
**Environment Details:**
* **CloudStack Version:** 4.22.0.0 (Latest)
* **Hypervisor:** KVM
* **Storage:** NFS(for both Primary and Secondary)
**Problem Description:**
I created a new System Service Offering with higher resources (4 CPU
Cores, 4000 MHz, 4096 MB RAM). I have mapped the UUID of this new
offering in the Global Settings under `consoleproxy.service.offering`
and `secstorage.service.offering`.
However, when I destroy the existing System VMs to trigger a rebuild,
they are being recreated using the old/inactive default offering instead
of the new one.
**Steps I have already taken:**
1. Created new System Offerings with the correct "System VM Type"
(Console proxy / Secondary storage).
2. Updated Global Settings with the new Offering UUIDs.
3. Restarted the `cloudstack-management` service.
4. Destroyed and Expunged the old System VMs.
When I try to use the new offering, I sometimes get the following error:
`"Secondary Storage Vm creation failure in zone [ZONE]. Error details:
Unable to allocate capacity on zone [1] due to [null]."`
Interestingly, if I revert the Global Settings to default, the VMs
deploy successfully with the original small offering. It seems the
system is unable to "see" or "allocate" the resources for the new custom
offering despite the host having enough free capacity.
Could anyone please guide me on why CloudStack is ignoring the new UUID
in Global Settings or why it fails to allocate capacity for a larger
System Offering? I would prefer a solution that doesn't involve manual
database manipulation if possible.
BR/
Titu
+8801714945632
[email protected]
