[Lustre-discuss] lustre OSS IP change

[Brendon]

Since the last post, I realized the fstab didn't have an entry for the
OST to mount. It's not clear to me how this was working before,
because I don't recall seeing an OST mounted when running "mount"
before. Anyway, continuing on...

On oss0 I ran:
mount -t lustre /dev/sda5 /mnt/ost0

Logs on oss0 say:
Jan 11 14:33:33 compute-oss-0-0 kernel: LustreError:
4458:0:(ldlm_lib.c:806:target_handle_connect()) mylustre-OST0003:
denying connection for new client 10.1.1.1 at tcp (mylustre-mdtlov_UUID):
3 clients in recovery for 300s
Jan 11 14:33:33 compute-oss-0-0 kernel: LustreError:
4458:0:(ldlm_lib.c:1536:target_send_reply_msg()) @@@ processing error
(-16)  req at ffff81021e84da00 x72/t0 o8-><?>@<?>:0/0 lens 240/144 e 0 to
0 dl 1294785313 ref 1 fl Interpret:/0/0 rc -16/0
Jan 11 14:33:58 compute-oss-0-0 kernel: LustreError:
4459:0:(ldlm_lib.c:806:target_handle_connect()) mylustre-OST0003:
denying connection for new client 10.1.1.1 at tcp (mylustre-mdtlov_UUID):
3 clients in recovery for 275s
...
Jan 11 14:37:18 compute-oss-0-0 kernel: LustreError:
4468:0:(ldlm_lib.c:806:target_handle_connect()) mylustre-OST0003:
denying connection for new client 10.1.1.1 at tcp (mylustre-mdtlov_UUID):
3 clients in recovery for 74s
Jan 11 14:37:18 compute-oss-0-0 kernel: LustreError:
4468:0:(ldlm_lib.c:806:target_handle_connect()) Skipped 1 previous
similar message
Jan 11 14:37:18 compute-oss-0-0 kernel: LustreError:
4468:0:(ldlm_lib.c:1536:target_send_reply_msg()) @@@ processing error
(-16)  req at ffff81021e310800 x91/t0 o8-><?>@<?>:0/0 lens 240/144 e 0 to
0 dl 1294785538 ref 1 fl Interpret:/0/0 rc -16/0
Jan 11 14:37:18 compute-oss-0-0 kernel: LustreError:
4468:0:(ldlm_lib.c:1536:target_send_reply_msg()) Skipped1 previous
similar message
Jan 11 14:38:33 compute-oss-0-0 kernel: LustreError:
0:0:(ldlm_lib.c:1105:target_recovery_expired()) mylustre-OST0003:
recovery timed out, aborting
Jan 11 14:38:33 compute-oss-0-0 kernel: LustreError:
4471:0:(genops.c:1024:class_disconnect_stale_exports())
mylustre-OST0003: disconnecting 3 stale clients


logs on head server :
Jan 11 14:33:33 jupiter kernel: LustreError: 11-0: an error occurred
while communicating with 10.1.1.2 at tcp. The ost_connect operation
failed with -16
Jan 11 14:34:23 jupiter last message repeated 2 times
Jan 11 14:35:38 jupiter last message repeated 3 times
Jan 11 14:37:18 jupiter last message repeated 3 times
Jan 11 14:37:18 jupiter kernel: LustreError: Skipped 1 previous similar message

Any help is much appreciated.
-Brendon