Datastore usage summary on vms has wrong value

Support
#1

Looking at the data I see in vcenter and SCVMM for actual used space on disk of a given vm doesn’t match what I see in MIQ. In both cases MIQ just reports the base size of the disk though it correctly reports them as thin.

When I check vcenter for a given vm it shows used space as 121gb (due to a bunch of snapshots) but MIQ reports only 50gigs (0 bytes for snapshots and total equals disks = 50GB)

Similarly for SCVMM, the disk allocated column is 131gigs in SCVMM but MIQ shows only 50GB for the disks (the size of the base disk even though it’s thin) and snapshots is 0 bytes and total datastore used space is 4MB.

Finally, for an azure vm, there doesn’t seem to be anything about disk size.

Is this a bug or am i doing something wrong?

1 Like
#2

Hi @jeff1

I have same issue from vcenter provider and I opened up a ticket here https://github.com/ManageIQ/manageiq-providers-vmware/issues/98

To me, presenting wrong disk space usage is not good for manageiq’s acceptance.
But I am not at the level of tracing the ruby code and try fixing the issue myself yet.

Not sure why this bug is not getting attention from miq guru.

#3

good question why there’s no traction. How am I supposed to implement any type of chargeback if it’s not accounting for space properly (you can argue about whether to charge for the full size of the disk or just the user portion if it’s thin but if it’s ignoring the size of all the snapshots which makes the size of the vm balloon, it’s a real problem which makes us have to find all kinds of workarounds)

#4

This issue is being discussed over at https://github.com/ManageIQ/manageiq-providers-scvmm/issues/17