[LU-4103] interop 2.5/2.6 replay-dual test_21a: Input/output error

• Type: Bug
• Resolution: Duplicate
• Priority: Minor
• Affects Version/s: Lustre 2.5.0, Lustre 2.6.0
• Environment: server and client: lustre-b2_5 RHEL6 build #2
• Severity: 3
• Assignee: Mikhail Pershin
• Reporter: Maloo

Description

This issue was created by maloo for sarah <[e-mail]>

This issue relates to the following test suite run: http://maloo.whamcloud.com/test_sets/bd37a616-3458-11e3-9356-52540035b04c.

The sub-test test_21a failed with the following error:

test_21a failed with 2

Lustre: DEBUG MARKER: == replay-dual test 21a: commit on sharing == 23:14:14 (1381644854)
Lustre: DEBUG MARKER: local REPLAY BARRIER on lustre-MDT0000
Lustre: Unmounted lustre-client
LustreError: 11-0: MGC10.10.4.211@tcp: Communicating with 10.10.4.211@tcp, operation obd_ping failed with -107.
LustreError: Skipped 6 previous similar messages
Lustre: lustre-MDT0000-mdc-ffff88007a40c000: Connection to lustre-MDT0000 (at 10.10.4.211@tcp) was lost; in progress operations using this service will wait for recovery to complete
Lustre: Skipped 6 previous similar messages
Lustre: 8782:0:(client.c:2679:ptlrpc_replay_interpret()) @@@ Version mismatch during replay
  req@ffff88007ae3a000 x1448748395066316/t511101108230(511101108230) o36->lustre-MDT0000-mdc-ffff88007a40c000@10.10.4.211@tcp:12/10 lens 536/424 e 0 to 0 dl 1381644986 ref 2 fl Interpret:R/4/0 rc -75/-75
Lustre: 8782:0:(import.c:1209:completed_replay_interpret()) lustre-MDT0000-mdc-ffff88007a40c000: version recovery fails, reconnecting
LustreError: 167-0: lustre-MDT0000-mdc-ffff88007a40c000: This client was evicted by lustre-MDT0000; in progress operations using this service will fail.
LustreError: 5224:0:(mdc_locks.c:915:mdc_enqueue()) ldlm_cli_enqueue: -5

Issue Links

• duplicates: LU-4104 Failure on test suite replay-dual test_21b: Not all renames are replayed. COS=1
• is duplicated by: LU-4135 mdt_save_lock() is broken

Comments

[Andreas Dilger]

This patch was landed to b2_5 and will be included in 2.5.1.

[Sarah Liu]

still hit this error in the latest test between 2.5.0 server and 2.6 client:

https://maloo.whamcloud.com/test_sets/6a064740-5dfc-11e3-aed2-52540035b04c

[Sarah Liu]

also hit this issue in interop test between 2.5.0 server and 2.6 client:

https://maloo.whamcloud.com/test_sets/50ff5ec8-4eaf-11e3-b578-52540035b04c

[Mikhail Pershin]

I think this is result of LU-4135 bug

[Jodi Levi (Inactive)]

Mike,
could you please have a look and comment on this one?

[Sarah Liu]

Hit this error with SLES11 SP3 client:

https://maloo.whamcloud.com/test_sets/c6bc907a-381d-11e3-844f-52540035b04c