Re: DAOS_test failed


anton.brekhov@...
 

I've installed MOFED 5.0.2 on both hosts, and openmpi. And it works a little further!
I've launched daos_test like 
mpirun --allow-run-as-root -np 1 daos_test.

There were passed and failed tests, but it ended with another error:

=================

DAOS rebuild tests..

=================

[  PASSED  ] 3 test(s).

setup: creating pool, SCM size=4 GB, NVMe size=8 GB

setup: created pool 88a9a5f1-8a21-4e45-a32d-ef87791c5f80

setup: connecting to pool

connected to pool, ntarget=4

setup: creating container 28aa2f9d-8559-41f3-907f-ec1b893ca90c

setup: opening container

REBUILD0: drop rebuild scan reply

No enough targets, skipping (4/0)

teardown: destroyed pool 88a9a5f1-8a21-4e45-a32d-ef87791c5f80

REBUILD1: retry rebuild for not ready

setup: creating pool, SCM size=0 GB, NVMe size=0 GB

daos_pool_create failed, rc: -1003

[sky08:21007:0:21007] Caught signal 11 (Segmentation fault: tkill(2) or tgkill(2) at address 0x520f)

==== backtrace (tid:  21007) ====

 0 0x000000000004cb95 ucs_debug_print_backtrace()  ???:0

 1 0x000000000045471e ???()  /usr/bin/daos_test:0

 2 0x000000000044ca21 ???()  /usr/bin/daos_test:0

 3 0x0000000000406db3 ???()  /usr/bin/daos_test:0

 4 0x0000000000022505 __libc_start_main()  ???:0

 5 0x00000000004079e2 ???()  /usr/bin/daos_test:0

=================================

[sky08:21007] *** Process received signal ***

[sky08:21007] Signal: Segmentation fault (11)

[sky08:21007] Signal code:  (-6)

[sky08:21007] Failing at address: 0x520f

[sky08:21007] [ 0] /lib64/libpthread.so.0(+0xf5f0)[0x7fdac52295f0]

[sky08:21007] [ 1] daos_test[0x45471e]

Do I need to set some env variables with scm and nvme size for this phase?

 
 
 
 
 

Join daos@daos.groups.io to automatically receive all group messages.