[Gluster-users] gluster peer probe error (v3.6.2)

[Andreas Hollaus]

Hi,

I get a strange result when I execute 'gluster peer probe'. The command hangs and
seems to timeout without any message (I can ping the address):
# gluster peer probe 10.32.1.144
# echo $?
146

The status looks promising, but there's a differences between this output and what
you normally get from a successful call:
# gluster peer status
Number of Peers: 1

Hostname: 10.32.1.144
Uuid: 0b008d3e-c51b-4243-ad19-c79c869ba9f2
State: Probe Sent to Peer (Connected)

(instead of 'State: Peer in Cluster (Connected)')

Running the command again will tell you that it is connected:

# gluster peer probe 10.32.1.144
peer probe: success. Host 10.32.1.144 port 24007 already in peer list

But when you try to add a brick from that server it fails:

# gluster volume add-brick c_test replica 2 10.32.1.144:/opt/lvmdir/c2 force
volume add-brick: failed: Host 10.32.1.144 is not in 'Peer in Cluster' state

The volume was previously created using the following commands:
# gluster volume create c_test 10.32.0.48:/opt/lvmdir/c2 force
volume create: c_test: success: please start the volume to access data
# gluster volume start c_test
volume start: c_test: success

What could be the reason for this problem?


Regards
Andreas