What I finally did was:
-commit force the replace-brick operation and the brick was changed.
-to force healing I ran find /mnt/gvol/ -print0 | xargs --null stat >/dev/null;
If the source disk dies I still have the old copy on the replaced brick, but going through a full directory listing is quite a time/resource consuming task.
2014-11-07 9:50 GMT-03:00 Juan José Pavlik Salles <jjpavlik@xxxxxxxxx>:
Hi guys, we added a new node to the cluster we have so I used replace-brick to move a brick from node 4 to the new node 5. Everything was going great but the operation was completed before moving all the files it was supposed to move. It barely moved 1TB from 2.3TB it should have moved.What should I do? Abort the replacement? Commit force it and let self heal move the rest of the files?Regards--Pavlik Salles Juan José
Pavlik Salles Juan José
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://supercolony.gluster.org/mailman/listinfo/gluster-users
