Description of problem: ----------------------- 'remove-brick force' was used to forcefully remove the brick from the volume, knowing that there would be data lost in that case. There is no data migration involved with 'remove-brick force' But as of now, with the latest build, there is a warning thrown, suggesting the user to copy the data from the removed brick to the gluster mount, if the files are not migrated. This arises contradiction. Version-Release number of selected component (if applicable): -------------------------------------------------------------- glusterfs-3.6.0.28-1.el6rhs How reproducible: ------------------ Always Steps to Reproduce: ------------------- 1. Create a distributed volume 2. Try removing a brick forcefully Actual results: --------------- Warning message is thrown to copy back the data from the removed-brick to gluster mount ( if there are any data available on removed brick ) [root@rhss5 ~]# gluster volume remove-brick repvol NODE1:/rhs/brick1/b1 NODE2:/rhs/brick1/b1 force Removing brick(s) can result in data loss. Do you want to Continue? (y/n) y volume remove-brick commit force: success Check the removed bricks to ensure all files are migrated. If files with data are found on the brick path, copy them via a gluster mount point before re-purposing the removed brick. Expected results: ----------------- The warning message should not be thrown, as there is no data migration is involved with 'remove-brick force' Additional Info ---------------- The intent of 'remove-brick force' is to remove the brick forcefully ignoring the data in it. We confirm the same before executing 'remove-brick force' saying - "Removing brick(s) can result in data loss. Do you want to Continue? (y/n)" Again after executing 'remove-brick force' again warning message is thrown as - "Check the removed bricks to ensure all files are migrated. If files with data are found on the brick path, copy them via a gluster mount point before re-purposing the removed brick.", which would be contradicting.
Example : [root@rhss5 ~]# gluster volume remove-brick repvol NODE1:/rhs/brick1/b1 NODE2:/rhs/brick1/b1 force Removing brick(s) can result in data loss. Do you want to Continue? (y/n) y volume remove-brick commit force: success Check the removed bricks to ensure all files are migrated. If files with data are found on the brick path, copy them via a gluster mount point before re-purposing the removed brick.
Please add doc text for this known issue.
Canceling need_info as Nithya reviewed and signed-off the doc text.
Verified with glusterfs-3.6.0.34-1.el6rhs and 'remove-brick force' doesn't show warning message to copy the files back from fuse mount. Performed the following test, 1. Created a 2 node cluster 2. Created a distributed volume with 3 bricks 3. Added more files to the volume from fuse mount 4. Performed "remove-brick force" on the volume Observation - No warning message seen to copy back the files as shown in remove-brick with data migration Console logs ------------ [root@dhcp37-44 ~]# gluster volume remove-brick distvol dhcp37-201.lab.eng.blr.redhat.com:/rhs/brick1/bb1 force Removing brick(s) can result in data loss. Do you want to Continue? (y/n) y volume remove-brick commit force: success Marking this bug as VERIFIED
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://rhn.redhat.com/errata/RHBA-2015-0038.html