search

ofi-cray / libfabric-cray

Open Fabric Interfaces
http://ofiwg.github.io/libfabric/
Other
16 stars 9 forks source link

Intermittent failure of self pingpong vc tests #1194

Open jswaro opened 7 years ago

jswaro commented 7 years ago

jswaro@jupiter:libfabric-cray(master**) $ for x in seq 1 1 20 ; do prov/gni/test/run_gnitest --filter 'vc_conn_ping_auto/ep_connect_self_pp' ; done [====] Synthesis: Tested: 1 | Passing: 1 | Failing: 0 | Crashing: 0 [====] Synthesis: Tested: 1 | Passing: 1 | Failing: 0 | Crashing: 0 [====] Synthesis: Tested: 1 | Passing: 1 | Failing: 0 | Crashing: 0 [----] prov/gni/test/vc.c:669: Assertion failed: CQE Context mismatch [FAIL] vc_conn_ping_auto::ep_connect_self_pp: (0.00s) [====] Synthesis: Tested: 1 | Passing: 0 | Failing: 1 | Crashing: 0 srun: error: nid00042: task 0: Exited with exit code 1 srun: Terminating job step 2081437.0 [====] Synthesis: Tested: 1 | Passing: 1 | Failing: 0 | Crashing: 0 [====] Synthesis: Tested: 1 | Passing: 1 | Failing: 0 | Crashing: 0 ...

@sungeunchoi believes this could be related to #1192 , and I suspect she is right. @a-abraham

sungeunchoi commented 7 years ago

I didn't say I thought it was related, I said that it could be a similar problem.