Unclear paragraph in 8.2. In-Service Software Upgrade from Red Hat Gluster Storage 3.2 to Red Hat Gluster Storage 3.3

Latest response

8.2.3. In-Service Software Upgrade
11. Perform the following operations to change the Automatic File Replication extended attributes so that the heal process happens from a brick in the replica subvolume to the thin provisioned brick.

The whole paragraph is not clear.
e.g. "Create a FUSE mount point from any server to edit the extended attributes. Extended attributes cannot be edited using the NFS and CIFS mount points." - what FUSE mount? which server? what directory ?

should this be done for all bricks or just one or some?


Q. Which server ? A. At the beginning of section 8.2.3, It is clearly given that : "The following steps have to be performed on each node of the replica pair: " and then it is followed by the sequence of steps and then in step 20 it says : "Repeat the above steps on the other node of the replica pair. In the case of a distributed-replicate setup, repeat the above steps on all the replica pairs. "

Q. What FUSE mount? A. Am administrator upgrading the gluster server is expected that he is aware of gluster a bit. FUSE is native protocol which can be used to access gluster volumes. In the doc, it has been suggested to use FUSE mount and not to use NFS and CIFS. But I understand your concern, sentences are not very clear.And even it is not clear if this is to be done for all volume or any one

Q. What Directory? A. It is being asked to create and delete a dummy directory via the temporary FUSE mount. Again, things are not clear here if this is to be done for one volume or all.

I understand your concern and will get this sorted via Bugzilla. Will update you about the bug once I open it.

Thanks for posting, Tomas - this is definitely not clear. I'll speak with Bipin today if he's available and see how quickly we can get this corrected and updated in the doc.

In the meantime, this section should help you create a FUSE mount point and mount your volume:


I have filed a bug for it https://bugzilla.redhat.com/show_bug.cgi?id=1618716. Let's continue on the bug.