[LU-3027] Failure on test suite parallel-scale test_write_disjoint: invalid file size 140329 instead of 160376 = 20047 * 8

• Type: Bug
• Resolution: Fixed
• Priority: Blocker
• Fix Version/s: Lustre 2.4.2
• Affects Version/s: Lustre 2.4.0, Lustre 2.4.1, Lustre 2.5.0
• Severity: 3
• Assignee: Oleg Drokin
• Reporter: Maloo

Description

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

This issue relates to the following test suite run: https://maloo.whamcloud.com/test_sets/2ed1fef2-94bd-11e2-93c6-52540035b04c.

The sub-test test_write_disjoint failed with the following error:

write_disjoint failed! 1

test log shows:

librdmacm: Fatal: no RDMA devices found
librdmacm: Fatal: no RDMA devices found
librdmacm: Fatal: no RDMA devices found
librdmacm: Fatal: no RDMA devices found
librdmacm: Fatal: no RDMA devices found
librdmacm: Fatal: no RDMA devices found
librdmacm: Fatal: no RDMA devices found
loop 0: chunk_size 103399
[client-27vm6.lab.whamcloud.com:00935] 7 more processes have sent help message help-mpi-btl-base.txt / btl:no-nics
[client-27vm6.lab.whamcloud.com:00935] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages
loop 79: chunk_size 71702, file size was 573616
rank 2, loop 80: invalid file size 140329 instead of 160376 = 20047 * 8
loop 79: chunk_size 71702, file size was 573616
rank 4, loop 80: invalid file size 140329 instead of 160376 = 20047 * 8
loop 79: chunk_size 71702, file size was 573616
rank 6, loop 80: invalid file size 140329 instead of 160376 = 20047 * 8
loop 79: chunk_size 71702, file size was 573616
rank 0, loop 80: invalid file size 140329 instead of 160376 = 20047 * 8
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 4 in communicator MPI_COMM_WORLD 
with errorcode -1.

Looks like LU-2453 is a similar issue seen in b2_1 branch

Issue Links

• is duplicated by: LU-701 parallel-scale test_write_disjoint fails due to invalid file size
• is related to: LU-3889 LBUG: (osc_lock.c:497:osc_lock_upcall()) ASSERTION( lock->cll_state >= CLS_QUEUING )
• is related to: LU-3874 parallel-scale-nfsv3 test_iorfpp: ERROR in aiori-POSIX.c (line 235): seek failed
• is related to: LU-3889 LBUG: (osc_lock.c:497:osc_lock_upcall()) ASSERTION( lock->cll_state >= CLS_QUEUING )