Description of problem: ------------------------ After the failure of remove-brick on a volume, due to brick processes being killed, all subsequent attempts to start further tasks like rebalance on the volume are failing. Version-Release number of selected component (if applicable): Red Hat Storage Console 2.1.2-0.27.beta.el6_5 How reproducible: Always Steps to Reproduce: 1. Start remove-brick on a volume. 2. While remove-brick is in progress, kill a brick process. The remove-brick icon gets updated as failed in the UI and the drop-down menu beside the icon has the button 'Status' enabled. 3. Now try to start rebalance on the same volume. Actual results: Starting rebalance fails with an events log message, that rebalance could not be started. Expected results: Rebalance start should have been successful as there are no running tasks on the volume. Additional info:
This behaviour should be documented in the Administration guide of RHSC
Along with the work around of going to the gluster CLI and running the stop command as given in the BZ 1042777
Please review the edited doc text and sign off.
Made a slight correction in the doc-text. It looks fine now.
So, the error returned by gluster is "A remove-brick task on volume remove-brick is not yet committed. Either commit or stop the remove-brick task." But frontend is not showing this error.So I think its better to show the same error in the UI rather than we taking a call whether to commit or stop remove-brick operation
(In reply to anmol babu from comment #5) > So, the error returned by gluster is "A remove-brick task on volume > remove-brick is not yet committed. Either commit or stop the remove-brick > task." > But frontend is not showing this error.So I think its better to show the > same error in the UI rather than we taking a call whether to commit or stop > remove-brick operation Sounds good to me.
Please provide the fixed in version.
This was wrongly moved to ON_QA and hence moved it back to Assigned. The fix is in master but needs to backported.
Removing FailedQA as QE din't verify this
I do not see the fix which is mentioned in comment 5 ? Has this been changed?
Moving this back since the expected results does not match with actual ones.
BTW Failed remove-brick task is not possible to stop, commit or retain from GUI as all those actions are disabled. So just to tell that rebalance cannot be run till stop or commit will be done doesn't help much.
Thank you for your report. This bug is filed against a component for which no further new development is being undertaken