Hi all!!
I’d like a mechanism of more intuitively identifying from within proxmox’s webui
(and possibly other places as well?)
the relationship between a block device and its vm-scoped-use
by this, i mean a label-construct that would allow for hinting to future-me ‘what-the-hell-this-device-is-for’ when looking at a VM.
the alignment of proxmox’s deviceid to blockdev-within-vm isn’t always terribly easy to deduce
on my gitlab vm, for example, I have a half dozen different blockdevs for the different heavy-io things…. ie each db has its own blockdev, elasticsearch has its own blockdev, logs, container storage, git-dir, git-lfs dir, artifacts, etc…
then there’s systemic blockdevs for docker, etc…
the problem that manifests is when looking at the VM in proxmox’ webui, I don’t have a good way to clearly identify which blockdev is which…
this has resulted in me needing to use blockdev size to inform me of which is which…
this feels….. suboptimal, and like there’s clearly a better way to facilitate this….
serial would work fine, if that were exposed in the UI….
a custom comment could work; but it would be possible for that to become stale and thus actively UNHELPFUL, so it wouldn’t be my first choice
if the qemu-agent exposes it, showing the vm-scoped active mount point for each blockdev in the webui would be stellar.
curious what others are doing here
I’d like a mechanism of more intuitively identifying from within proxmox’s webui
(and possibly other places as well?)
the relationship between a block device and its vm-scoped-use
by this, i mean a label-construct that would allow for hinting to future-me ‘what-the-hell-this-device-is-for’ when looking at a VM.
the alignment of proxmox’s deviceid to blockdev-within-vm isn’t always terribly easy to deduce
on my gitlab vm, for example, I have a half dozen different blockdevs for the different heavy-io things…. ie each db has its own blockdev, elasticsearch has its own blockdev, logs, container storage, git-dir, git-lfs dir, artifacts, etc…
then there’s systemic blockdevs for docker, etc…
the problem that manifests is when looking at the VM in proxmox’ webui, I don’t have a good way to clearly identify which blockdev is which…
this has resulted in me needing to use blockdev size to inform me of which is which…
this feels….. suboptimal, and like there’s clearly a better way to facilitate this….
serial would work fine, if that were exposed in the UI….
a custom comment could work; but it would be possible for that to become stale and thus actively UNHELPFUL, so it wouldn’t be my first choice
if the qemu-agent exposes it, showing the vm-scoped active mount point for each blockdev in the webui would be stellar.
curious what others are doing here