[lustre-discuss] Filesystem hanging....

Phill Harvey-Smith p.harvey-smith at warwick.ac.uk
Thu Aug 11 06:43:09 PDT 2016


Hi all,

I have a (fairly urgent) problem.

I have been updating our cluster to Ubuntu 16.04, which has on the whole 
gone well, however in the last part I have run across a rather serious 
error.

Our frontend node has an instance of samba that shares out the home 
directories, we have found that writing to a file on /home will cause 
the /home mount to timeout and become inacessable.

The errors that are reported in the journal are like so :

> Aug 11 10:06:34 buster-fe0 kernel: Lustre: Lustre: Build Version: 2.8.53_51_g3680fa1_dirty
> Aug 11 10:06:34 buster-fe0 kernel: Lustre: Server MGS version (2.1.0.0) is much older than client. Consider upgrading server (2.8.53_51_g3680fa1_dirty)
> Aug 11 10:06:34 buster-fe0 kernel: Lustre: Trying to mount a client with IR setting not compatible with current mgc. Force to use current mgc setting that is IR disabled.
> Aug 11 10:06:34 buster-fe0 kernel: Lustre: Mounted home-client
> Aug 11 10:06:34 buster-fe0 mount[4687]: mount.lustre: addmntent: Invalid argument:
> Aug 11 10:06:34 buster-fe0 mount[4691]: mount.lustre: addmntent: Invalid argument:
> Aug 11 10:06:34 buster-fe0 mount[4670]: mount.lustre: addmntent: Invalid argument:
> Aug 11 10:06:35 buster-fe0 systemd[1]: Started Lustre setup.
> Aug 11 10:06:37 buster-fe0 lustre.sh[4870]: llite.home-ffff881005c37800.create_no_open_optimization=0
> Aug 11 13:48:12 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919685/real 1470919685]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919692 ref 2 fl Rpc:X/0/ffffffff rc 0/-1
> Aug 11 13:48:12 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:48:12 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:48:19 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919692/real 1470919692]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919699 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:48:19 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:48:19 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:48:26 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919699/real 1470919699]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919706 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:48:26 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:48:26 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:48:33 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919706/real 1470919706]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919713 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:48:33 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:48:33 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:48:40 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919713/real 1470919713]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919720 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:48:40 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:48:40 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:48:54 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919727/real 1470919727]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919734 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:48:54 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 1 previous similar message
> Aug 11 13:48:54 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:48:54 buster-fe0 kernel: Lustre: Skipped 1 previous similar message
> Aug 11 13:48:54 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:48:54 buster-fe0 kernel: Lustre: Skipped 1 previous similar message
> Aug 11 13:49:15 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919748/real 1470919748]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919755 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:49:15 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 2 previous similar messages
> Aug 11 13:49:15 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:49:15 buster-fe0 kernel: Lustre: Skipped 2 previous similar messages
> Aug 11 13:49:15 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:49:15 buster-fe0 kernel: Lustre: Skipped 2 previous similar messages
> Aug 11 13:49:50 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919783/real 1470919783]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919790 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:49:50 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 4 previous similar messages
> Aug 11 13:49:50 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:49:50 buster-fe0 kernel: Lustre: Skipped 4 previous similar messages
> Aug 11 13:49:50 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:49:50 buster-fe0 kernel: Lustre: Skipped 4 previous similar messages
> Aug 11 13:51:00 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919853/real 1470919853]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919860 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:51:00 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 9 previous similar messages
> Aug 11 13:51:00 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:51:00 buster-fe0 kernel: Lustre: Skipped 9 previous similar messages
> Aug 11 13:51:00 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:51:00 buster-fe0 kernel: Lustre: Skipped 9 previous similar messages
> Aug 11 13:53:13 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470919986/real 1470919986]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470919993 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:53:13 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 18 previous similar messages
> Aug 11 13:53:13 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:53:13 buster-fe0 kernel: Lustre: Skipped 18 previous similar messages
> Aug 11 13:53:13 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:53:13 buster-fe0 kernel: Lustre: Skipped 18 previous similar messages
> Aug 11 13:55:38 buster-fe0 kernel:  [<ffffffffc0c588fc>] ? ll_lookup_finish_locks+0xfc/0x8a0 [lustre]
> Aug 11 13:57:32 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470920245/real 1470920245]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470920252 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 13:57:32 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 36 previous similar messages
> Aug 11 13:57:32 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 13:57:32 buster-fe0 kernel: Lustre: Skipped 36 previous similar messages
> Aug 11 13:57:32 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 13:57:32 buster-fe0 kernel: Lustre: Skipped 36 previous similar messages
> Aug 11 13:57:38 buster-fe0 kernel:  [<ffffffffc0c588fc>] ? ll_lookup_finish_locks+0xfc/0x8a0 [lustre]
> Aug 11 13:57:38 buster-fe0 kernel:  [<ffffffffc0c588fc>] ? ll_lookup_finish_locks+0xfc/0x8a0 [lustre]
> Aug 11 13:57:38 buster-fe0 kernel:  [<ffffffffc0c588fc>] ? ll_lookup_finish_locks+0xfc/0x8a0 [lustre]
> Aug 11 13:59:38 buster-fe0 kernel:  [<ffffffffc0c588fc>] ? ll_lookup_finish_locks+0xfc/0x8a0 [lustre]
> Aug 11 13:59:38 buster-fe0 kernel:  [<ffffffffc0c588fc>] ? ll_lookup_finish_locks+0xfc/0x8a0 [lustre]
> Aug 11 13:59:38 buster-fe0 kernel:  [<ffffffffc0c588fc>] ? ll_lookup_finish_locks+0xfc/0x8a0 [lustre]
> Aug 11 14:06:10 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470920763/real 1470920763]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470920770 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 14:06:10 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 73 previous similar messages
> Aug 11 14:06:10 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 14:06:10 buster-fe0 kernel: Lustre: Skipped 73 previous similar messages
> Aug 11 14:06:10 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 14:06:10 buster-fe0 kernel: Lustre: Skipped 73 previous similar messages
> Aug 11 14:16:12 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470921365/real 1470921365]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470921372 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 14:16:12 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 85 previous similar messages
> Aug 11 14:16:12 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 14:16:12 buster-fe0 kernel: Lustre: Skipped 85 previous similar messages
> Aug 11 14:16:12 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 14:16:12 buster-fe0 kernel: Lustre: Skipped 85 previous similar messages
> Aug 11 14:26:14 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470921967/real 1470921967]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470921974 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 14:26:14 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 85 previous similar messages
> Aug 11 14:26:14 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 14:26:14 buster-fe0 kernel: Lustre: Skipped 85 previous similar messages
> Aug 11 14:26:14 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 14:26:14 buster-fe0 kernel: Lustre: Skipped 85 previous similar messages
> Aug 11 14:36:16 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1470922569/real 1470922569]  req at ffff8807e6076c00 x1542357171466016/t0(0) o55->home-MDT0000-mdc-ffff881005c37800 at 192.168.0.4@tcp:12/10 lens 592/224 e 0 to 1 dl 1470922576 ref 2 fl Rpc:X/2/ffffffff rc 0/-1
> Aug 11 14:36:16 buster-fe0 kernel: Lustre: 20221:0:(client.c:2067:ptlrpc_expire_one_request()) Skipped 85 previous similar messages
> Aug 11 14:36:16 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection to home-MDT0000 (at 192.168.0.4 at tcp) was lost; in progress operations using this service will wait for recovery to complete
> Aug 11 14:36:16 buster-fe0 kernel: Lustre: Skipped 85 previous similar messages
> Aug 11 14:36:16 buster-fe0 kernel: Lustre: home-MDT0000-mdc-ffff881005c37800: Connection restored to 192.168.0.4 at tcp (at 192.168.0.4 at tcp)
> Aug 11 14:36:16 buster-fe0 kernel: Lustre: Skipped 85 previous similar messages

I also can't get the recently checked out sources to compile, but will 
post a seperate query about that. :)

Cheers.

Phill.


More information about the lustre-discuss mailing list