I have a SYS-A21GE-NBRT-04-LL014 server and after running the Fieldiag utility, I see the following failing error message for ibstressmad:
DGX-000000000007 | ibstressmad | ibstressmad | | InfiniBand | mlx5_0 <-> mlx5_1 | Failed to execute '('bash', '-c', 'for i in {1..5000}; do smpquery -C mlx5_0 NI -D 0,1,74,73 > /dev/null; done')'
DGX-000000000002 | ibstressmad | ibstressmad | | Infiniband | | Timeout
How do I fix or troubleshoot this issue?
Please check to see which version of Fieldiag / PartnerDiag you are using.
The following version has a known fix to ibstressmad test timeout issue:
Partner Diag Version: 629-26287-0102-FLD-43576