Description of problem ====================== There is a discrepancy between Bricks and Brick Status panel, one uses "up" and "down", the other uses "started" and "stopped" when reporting status. Version-Release number of selected component ============================================ tendrl-monitoring-integration-1.6.3-7.el7rhgs.noarch Steps to Reproduce ================== 1. Instal RHGS WA using tendrl-ansible 2. Import Trusted storage pool with at least one volume 3. Go to Host dashboard and check "Bricks" and "Brick Status" panels Actual results ============== Brick panel reports number for: * bricks in "up" state * bricks in "down" state * total number of bricks While description of "Brick Status" panel defines status codes in the table as: * 0 = Started * 8 = Stopped See screenshot 1. Expected result =============== Both panels uses the same set of brick states. Additional questions ==================== Is there any difference? Do we use the same terminology as RHGS core?
Created attachment 1474481 [details] screenshot 1: discrepancy in brick states on host dashboard
Volume dashboard has the same problem, updating the scope of this BZ to cover both Volume and Host dashboards. To verify this BZ, QE team will check this problem on: * Volume Dashboard * Host Dashboard
Created attachment 1474600 [details] screenshot 2: discrepancy in brick states on volume dashboard
Created attachment 1479434 [details] Volume dashboard brick status
Created attachment 1479435 [details] Host dashboard brick status
Martin, please check my screenshots, is this the expectation?
Created attachment 1481084 [details] Cluster_dashboard_brick_status
I have proposed changes in screenshots, please review
Gowtham, I see you are proposing to use "started"/"stopped" everywhere. This looks good. That said, I would like to ask gluster eng. if they also agree with this choice (I would rather prevent some subtle naming clash). Atin, could you please verify that it's ok for WA to report status of a brick as either "started" or "stopped".
Note that there is other BZ, which points out that the list of states is not complete (BZ 1618770 demonstrates that there is at least 1 other state which the description doesn't talk about). To be able to make progress here, I'm highlighting that this BZ is solely about making sure the same names of brick states are used across WA dashboard, and it's not concerned with the fact that we have some brick states missing (which is tracked in another, already mentioned BZ 1618770).
(In reply to Martin Bukatovic from comment #10) > Gowtham, I see you are proposing to use "started"/"stopped" everywhere. This > looks good. That said, I would like to ask gluster eng. if they also agree > with > this choice (I would rather prevent some subtle naming clash). > > Atin, could you please verify that it's ok for WA to report status of a brick > as either "started" or "stopped". I'd definitely vote for the same as in glusterd the brick's status is represented as 'started' or 'stopped' .
PR is under review: https://github.com/Tendrl/monitoring-integration/pull/546
Brick status is Started or Stopped (not Up or Down) everywhere on Volume and Host dashboards.
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/RHBA-2018:3427