Just got back to this after long delay. I have a slightly different question now. I have a working distributed/replicated setup with 4 bricks. They are called: gf2ib-1, gf2ib-2, gf2ib-1r, gf2ib-2r. Running the "gluster peer status" command from gf2ib-1 I get: Number of Peers: 3 Hostname: gf2ib-2 Uuid: ed599401-25e2-41bb-8e41-b00247a3f3a0 State: Peer in Cluster (Connected) Hostname: gf2ib-1r Uuid: 8ea05f85-b416-4047-9e65-9a8185534bb6 State: Peer in Cluster (Connected) Hostname: gf2ib-2r Uuid: 34c58361-b3ef-4201-998b-09ff5c73558e State: Peer in Cluster (Connected) That's what I expect. But from the other three peers, I get a variant of this (the IP address of the first): [root at gf2-2 rcadmin]# gluster peer status Number of Peers: 3 Hostname: 172.24.215.202 Uuid: f5edea20-9467-48ed-b4f1-dc566a9b6d02 State: Peer in Cluster (Connected) Hostname: gf2ib-1r Uuid: 8ea05f85-b416-4047-9e65-9a8185534bb6 State: Peer in Cluster (Connected) Hostname: gf2ib-2r Uuid: 34c58361-b3ef-4201-998b-09ff5c73558e State: Peer in Cluster (Connected) Sure enough, when I look at /var/lib/glusterd/peers/* I see something like this: uuid=f5edea20-9467-48ed-b4f1-dc566a9b6d02 state=3 hostname1=172.24.215.202 Maybe it doesn't matter, but can I stop the volume and change that hostname line? The machine I'm looking at does't have trouble with it's DNS entry: [root at gf2-2r rcadmin]# nslookup 172.24.215.202 Server: 155.52.45.100 Address: 155.52.45.100#53 Non-authoritative answer: 202.215.24.172.in-addr.arpa name = gf2ib-1.dfci.harvard.edu. The "volume info gf2" command gives the *names* of the 4 bricks, too. Can anyone explain? Matt Temple ------ Matt Temple Director, Research Computing Dana-Farber Cancer Institute. On Wed, Dec 19, 2012 at 1:30 AM, Shawn Heisey <gluster at elyograg.org> wrote: > On 12/18/2012 7:33 PM, Matthew Temple wrote: > >> From any of those nodes, peer status is wrong: >> For instance, if I ssh over to 2-ib, >> and ask for a peer status, it shows the peers to to be >> >> 1-ib-r >> 2-ib-r >> and *155.52.48.1* (the last is the ethernet side of, not the IB side) >> >> I guess my question is this: How do I get the first node to use the ib >> side when >> connecting to the other nodes. (both of these networks are public.) Do >> I have to >> do something interesting with routes? >> > > Just probe the first server by name from any of the other three. That > will fix the peers. > > Thanks, > Shawn > > ______________________________**_________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://supercolony.gluster.**org/mailman/listinfo/gluster-**users<http://supercolony.gluster.org/mailman/listinfo/gluster-users> > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20130118/afcdc43a/attachment.html>
