Created attachment 925971 [details] example showing the mismatch between peer/brick names and vol heal info Description of problem: I have a trusted pool that has a front end network for nfs and a backend network for gluster. The front end network uses the servers hostname for client connectivity and the gluster network uses a different subnet and node names for the trusted pool. vol info in this environment returns a brick name that references the name of the hosts on the client side network not the gluster network. Version-Release number of selected component (if applicable): testing on rhs3 How reproducible: this behavior is always seen Steps to Reproduce: 1. create a cluster as shown above 2. run the commands (example output attached) 3. Actual results: vol heal provides bricknames using the incorrect node names Expected results: vol heal info brick information should match the brick paths known by gluster and shown in the other commands like vol status/vol info. Additional info:
This was fixed upstream in BZ 1265633 and is also available in RHGS 3.1.1 as part of rebase. Hence closing this bug. Feel free to re-open if I have overlooked something,