Bug 1674611

The dependent bug is ON_QA, and this bug has all acks.
Moving this bug to ON_QA

Tested with the following:-
1.rhvh-4.3.0.5-0.20190221
2.glusterfs-server-3.12.2-45.el7rhgs
3.cockpit-ovirt-dashboard-0.12.2-1.el7ev.

Didnot increase the size of /var/log to 15GB and the deployment failed .

The error encountered was :-
 TASK [gluster.features/roles/gluster_hci : Check if /var/log has enough disk space] ***
fatal: [rhsqa-grafton1-nic2.lab.eng.blr.redhat.com]: FAILED! => {"changed": true, "cmd": "df -m /var/log | awk '/[0-9]%/ {print $4}'", "delta": "0:00:00.007996", "end": "2019-02-22 15:36:55.049497", "failed_when_result": true, "rc": 0, "start": "2019-02-22 15:36:55.041501", "stderr": "", "stderr_lines": [], "stdout": "7470", "stdout_lines": ["7470"]}
fatal: [rhsqa-grafton2-nic2.lab.eng.blr.redhat.com]: FAILED! => {"changed": true, "cmd": "df -m /var/log | awk '/[0-9]%/ {print $4}'", "delta": "0:00:00.007380", "end": "2019-02-22 15:36:55.285707", "failed_when_result": true, "rc": 0, "start": "2019-02-22 15:36:55.278327", "stderr": "", "stderr_lines": [], "stdout": "7470", "stdout_lines": ["7470"]}
fatal: [rhsqa-grafton3-nic2.lab.eng.blr.redhat.com]: FAILED! => {"changed": true, "cmd": "df -m /var/log | awk '/[0-9]%/ {print $4}'", "delta": "0:00:00.008420", "end": "2019-02-22 15:36:55.292038", "failed_when_result": true, "rc": 0, "start": "2019-02-22 15:36:55.283618", "stderr": "", "stderr_lines": [], "stdout": "7470", "stdout_lines": ["7470"]}


Hence, moving the bug to verified state.

I think there is a problem here.  The doc-bug https://bugzilla.redhat.com/show_bug.cgi?id=1565155#c9 requested that the size of the logical volume be created as 15GB, but the code doesn't check for that.  Instead the code checks for available space in the logical volume.  In my case, I installed the 4.3beta and because I failed to successfully install a few times, I had some content created in /var/log which reduced my available space.

My I suggest that the code [1] be changed from:
 # df -m /var/log | awk '/[0-9]%/ {print $4}'

to something like this:
 # lvs -o lv_name,lv_size --unit g | awk '/log/ {print $2}'


[1] https://github.com/gluster/gluster-ansible-features/blob/master/roles/gluster_hci/tasks/prerequisites.yml#L52

(In reply to John Call from comment #4)
> I think there is a problem here.  The doc-bug
> https://bugzilla.redhat.com/show_bug.cgi?id=1565155#c9 

Hello John,
This doc-bug is old ( for RHHI-V 1.5 )
We have updated the doc as part of bug https://bugzilla.redhat.com/show_bug.cgi?id=1683075

Laura has already filed a bug to help users ( with RHVH base ) to
expand /var/log using cockpit
BZ 1692228 - Changing /var/log size during deployment using Cockpit

Nevertheless, it would be a great idea to have RHVH default partition
gives importance to /var/log and makes it to atleast 30-40GB.
Currently /var/log LV is created with 8GB space, which is too small.
I remember lot of customer cases came up, related to low /var/log space.

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHEA-2019:1121