Description of problem: Currently, there are multiple ways to try to derive or obtain volume capacity utilization information, but they all seem to conflict in some way or other. With the new console (RHGS Web Admin), we are unable to reliably tell what is the proper capacity utilization for a volume as output differs when using various tools including gstatus, getstate (and aggregating the brick information), samba, etc. Ideally, this ought to be solved at the Gluster underlying infrastructure layer and gets bubbled up through the various tools/utilities with a consistent number. It would also be nice to have a "df" like mechanism in the CLI to quickly find out what the capacity utilization (total, free, available/free) for a volume is that might be part of the "gluster volume" CLI command. Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Just wanting to mention related BZ's here: [RFE] Gluster get state command should provide volume and cluster utilization related information - https://bugzilla.redhat.com/show_bug.cgi?id=1488018 FutureFeature: Capacity information provided in the volume view in the console -- https://bugzilla.redhat.com/show_bug.cgi?id=844321
We recommend to collect the 'volume utilization' on top of the mounted volume by 'df' (or with libgfapi ('gfapi_statfs()'). This is the right way because Gluster's components internally can consume extra space if they want it for xattrs or any other reason. Also, this will work regardless of the volume type for the consumer.
Closing the bug as the 2 bugs mentioned above bz1488018 and bz844321 are both in 'VERIFIED' or 'CLOSED' state. My comment (#3) still holds as an answer to the bug.