Document URL: Section Number and Name: Describe the issue: The order in which glusterd generates the volfiles during add-brick can cause data inconsistency when we change from a replica 2 to replica 3 or replica 2 + arbiter. Suggestions for improvement: It is recommended to stop the self heal process and then then add-brick. Once the graph changes have happened its fine to start the self heal. Additional information: This bug is to address the bug: https://bugzilla.redhat.com/show_bug.cgi?id=1400092
Based on comment#0 It is required that the vol file is updated on all the clients before we enable all heals and issue heal. Ideally waiting for 5 minutes is a practice followed so that vol file gets updated Could you add a note after the below doc text ------------- If you have an existing two-way distributed-replicated volume, you need a new brick for each sub-volume in order to convert it to an arbitrated distributed-replicated volume, for example: # gluster volume add-brick testvol replica 3 arbiter 1 server1:/bricks/arbiter_brick1 server2:/bricks/arbiter_brick2 ----Please add the below Note: It is advised to wait for 5 minutes and then turn on the self-heal and the Self-heal-daemon after the bricks are added.
Verified the updated doc, Looks good to me. Hence setting this to verified
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 1000 days