+++ This bug was initially created as a clone of Bug #783249 +++ Description of problem: When assert_on_child_down is enabled, glusterfs process exits without providing an opportunity to do cleanup before exiting. This needs to be corrected. Version-Release number of selected component (if applicable): 3.2.5 How reproducible: Always Steps to Reproduce: 1. Enable assert_on_child_down in dht 2. Bring a brick down Actual results: A stale client mount is seen Expected results: Stale client mount should not be seen Additional info: --- Additional comment from aavati on 2012-01-24 11:36:41 EST --- CHANGE: http://review.gluster.com/2682 (cluster/dht: Exit clean when assert_no_child_down is enabled) merged in release-3.2 by Anand Avati (avati) --- Additional comment from shmohan on 2012-02-17 04:18:04 EST --- 1. Created a distribute volume with 3 bricks 2. mounted using the command "/opt/glusterfs/3.2.6qa3/sbin/glusterfs --log-level=INFO --xlator-option=dist*.assert-no-child-down=yes --volfile-id=/dist --volfile-server=10.1.11.117 /mnt" 3. cd /mnt 4. killed one of the brick 5. cd / 6. client was automatically unmounted Marking as verified --- Additional comment from aavati on 2012-02-19 04:42:54 EST --- CHANGE: http://review.gluster.com/2664 (cluster/dht: Exit clean when assert_no_child_down is enabled) merged in master by Vijay Bellur (vijay)
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. http://rhn.redhat.com/errata/RHBA-2012-0538.html