Description of problem: ======================= Gluster volume heal info lists all the entries of all the bricks instead of just the bricks which were down. Brought down 2 of the bricks in 1x(8+4) and created 2000 files of 1g size. Brought back the bricks and given gluster v heal info. whlie heal full throws below error: [root@vertigo ~]# gluster v heal vol1 full Commit failed on aff26539-85fc-47c2-8d76-854503066ccc. Please check log file for details. Version-Release number of selected component (if applicable): ============================================================= [root@vertigo ~]# gluster --version glusterfs 3.7.0beta2 built on May 11 2015 01:27:46 Repository revision: git://git.gluster.com/glusterfs.git Copyright (c) 2006-2011 Gluster Inc. <http://www.gluster.com> GlusterFS comes with ABSOLUTELY NO WARRANTY. You may redistribute copies of GlusterFS under the terms of the GNU General Public License. [root@vertigo ~]# How reproducible: ================ 100% Steps to Reproduce: 1. create a 1x(8+4) disperse volume 2. Fuse mount on the client. 3. Bring down 2 of the bricks and create 2000 files of 1g from client 4. Brick back up the bricks 5. Do heal with gluster v heal volname full and see the entries with gluster v heal volname info Actual results: =============== All the files in all the bricks are listed instead of just failed bricks. Expected results: ================= Additional info: ================ sosreport of the node will be attached.
This is working as per expected behaviour. If the bricks are taken down at the start of untar. It will show all the files as needing heal from other bricks to the ones that are down.