[Gluster-users] Problem mounting on second NIC

[Sean Davis]

I'm asking a lot of questions--sorry about that.

I have a cluster setup with most of the nodes having two NICs, one for
a local network running over an internal GigE switch and the other to
our institutional network (also GigE).  I have configured four bricks
to run in unify on the internal network.  The resulting file system is
mounted on each node via the local network:

volume remote1
  type protocol/client
  option transport-type tcp/client
  option remote-host 192.168.8.104
  option remote-subvolume brick
end-volume

volume remote2
  type protocol/client
  option transport-type tcp/client
  option remote-host 192.168.8.102
  option remote-subvolume brick
end-volume

.....

volume unify0
  type cluster/unify
  option scheduler rr # round robin
  option namespace remote-ns
  #option block-size *:1MB
  subvolumes remote1 remote2 remote3 remote4
end-volume

I have a second set of machines that are not on the internal network
but communicate over the institutional network.  I have modified the
client configuration to use the IP addresses for the institutional
network for each of the bricks.  However, when I try to mount the file
system using this configuration, I get this in the glusterfs.log.

2008-12-10 13:22:32 W [client-protocol.c:332:client_protocol_xfer]
remote2: not connected at the moment to submit frame type(1) op(34)
2008-12-10 13:22:32 E [client-protocol.c:4430:client_lookup_cbk]
remote2: no proper reply from server, returning ENOTCONN
2008-12-10 13:22:32 W [client-protocol.c:332:client_protocol_xfer]
remote3: not connected at the moment to submit frame type(1) op(34)
2008-12-10 13:22:32 E [client-protocol.c:4430:client_lookup_cbk]
remote3: no proper reply from server, returning ENOTCONN
2008-12-10 13:22:32 W [client-protocol.c:332:client_protocol_xfer]
remote4: not connected at the moment to submit frame type(1) op(34)
2008-12-10 13:22:32 E [client-protocol.c:4430:client_lookup_cbk]
remote4: no proper reply from server, returning ENOTCONN
2008-12-10 13:22:32 W [client-protocol.c:332:client_protocol_xfer]
remote-ns: not connected at the moment to submit frame type(1) op(34)
2008-12-10 13:22:32 E [client-protocol.c:4430:client_lookup_cbk]
remote-ns: no proper reply from server, returning ENOTCONN
2008-12-10 13:22:32 E [fuse-bridge.c:468:fuse_entry_cbk]
glusterfs-fuse: 6: (34) / => -1 (2)
2008-12-10 13:22:32 W [client-protocol.c:332:client_protocol_xfer]
remote2: not connected at the moment to submit frame type(1) op(34)
2008-12-10 13:22:32 E [client-protocol.c:4430:client_lookup_cbk]
remote2: no proper reply from server, returning ENOTCONN
2008-12-10 13:22:32 W [client-protocol.c:332:client_protocol_xfer]
remote3: not connected at the moment to submit frame type(1) op(34)
2008-12-10 13:22:32 E [client-protocol.c:4430:client_lookup_cbk]
remote3: no proper reply from server, returning ENOTCONN
2008-12-10 13:22:32 W [client-protocol.c:332:client_protocol_xfer]
remote4: not connected at the moment to submit frame type(1) op(34)
2008-12-10 13:22:32 E [client-protocol.c:4430:client_lookup_cbk]
remote4: no proper reply from server, returning ENOTCONN
2008-12-10 13:22:32 W [client-protocol.c:332:client_protocol_xfer]
remote-ns: not connected at the moment to submit frame type(1) op(34)
2008-12-10 13:22:32 E [client-protocol.c:4430:client_lookup_cbk]
remote-ns: no proper reply from server, returning ENOTCONN
2008-12-10 13:22:32 E [fuse-bridge.c:468:fuse_entry_cbk]
glusterfs-fuse: 6: (34) / => -1 (2)


Any suggestions as to what is going on?  I didn't an answer to this
anywhere in the docs and google came up pretty dry, also.

Thanks again,
Sean