Metadata space for pool vm-pool is running out. Current usage: 95%

User Support General
1

Qubes Disk Space Monitor

WARNING!
You are running out of disk space.

Metadata space for pool vm-pool is running out. Current usage: 95%

varlibqubes is around 10 GiB out of 20 GiB full, vm-pool is roughly 300 GiB out of 700 GiB full. Clearly, there’s still a lot of space remaining.

What does this warning tell me and what should I do? Can I somehow increase the metadata share of the full disk?

VM's won't start - LVM Metadata reached 100% sounds scary — apparently that person couldn’t even delete VMs when the LVM metadata reached 100%.

Thin pool running out of disc space (SSD) sounds unrelated, the disk in that example is actually full.

2

Thats LVM - it has a small meta directory of which blocks map to which LV, that is what is running out of space.

lvextend --poolmetadatasize +100M vg_name/pool_name

That will give you probably enough headroom. Make it +1G if you never want to hear from it again.

3 
lvextend --poolmetadatasize +100M qubes_dom0/vm-pool

in dom0 has indeed fixed the probblem. Thank you very much!

Apparently, the volume had a size of 88 MiB before, and it now has 188 MiB.