Description of problem: When user creates 256 snapshots from the CLI and tries to delete all of them by using "Delete All" button it displays an error which says "Operation cancelled : Error while executing action Delete all volume snapshots: Network error during communication with the Host." An event message gets displayed "Failed to delete all the gluster volume snapshots for the volume vol2." but all the snapshots gets deleted. Version-Release number of selected component (if applicable): ovirt-engine-3.6.0-0.0.master.20150317182251.git9682bc4.el6.noarch How reproducible: Always Steps to Reproduce: 1. Manage a cluster through rhsc. 2. Create a volume. 3. Run the snapshot create command in a loop by executing the command "for i in {1..256}; do gluster snapshot create snap$i vol2;done" 4. Once the snapshots are created and synced in the UI, click on the "Delete All" button in the UI. Actual results: A popup appears with title "Operation cancelled : Error while executing action Delete all volume snapshots: Network error during communication with the Host." and event message reads as "Failed to delete all the gluster volume snapshots for the volume vol2." All the snapshots in the system gets deleted. Expected results: UI should not display Operation cancelled Popups. Additional info:
I do not see any error/exceptions in the engine logs.
This is happening only with RHEL6 nodes, on RHEL7 nodes this issue is gone.
In rhel6 also I remember the snapshots actually get deleted Gluster side, only the engine command fails due to time out.
Target release should be placed once a package build is known to fix a issue. Since this bug is not modified, the target version has been reset. Please use target milestone to plan a fix for a oVirt release.
oVirt 3.6.0 has been released on November 4th, re-targeting to 4.0 since this bug has been marked with severity < high
This bug is flagged for 3.6, yet the milestone is for 4.0 version, therefore the milestone has been reset. Please set the correct milestone or add the flag.
This should be blocked on being able to set different timeout for different commands. Can you add that RFE to block this one?
This bug is dependent on the ability to set time out values for vdsm calls.
Moving target to 4.1.0 as dependent on an infra bug
Revisit when dependent infra bug is fixed