1. 使用命令行方式分析namd应用,使用-d指定采样时间为20时,会出现以下报错,采样失败;同样的采样参数在web界面也会报错退出;
70] Stack Traceback:
[70:0] linux-vdso.so.1 0xffff9e04693c __kernel_rt_sigreturn
[70:1] libucp.so.0 0xffff986bc970 ucp_worker_progress
[70:2] libucg_planc_ucx.so 0xffff9030b920 ucg_planc_ucx_p2p_testall
[70:3] libucg_planc_ucx.so 0xffff90306708
[70:4] libucg.so.1 0xffff981c2e74
[70:5] libucg.so.1 0xffff981c5fbc ucg_request_test
[70:6] mca_coll_ucg.so 0xffff981f2088 mca_coll_ucg_request_execute
[70:7] mca_coll_ucg.so 0xffff981f21d0 mca_coll_ucg_request_execute_cache
[70:8] mca_coll_ucg.so 0xffff981f5114 mca_coll_ucg_barrier_cache
[70:9] libmpi.so.40 0xffff9ada050c MPI_Barrier
[70:10] libmpicollector_ompi.so 0xffff9bf84850 MPI_Barrier
[70:11] namd3 0xb8e5e4
[70:12] namd3 0xb932b4
[70:13] namd3 0xb8cf0c
[70:14] namd3 0x473194
[70:15] namd3 0x46f790
[70:16] libc.so.6 0xffff9a975fbc
[70:17] libc.so.6 0xffff9a976094 __libc_start_main
[70:18] namd3 0x40f8f0
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 70 in communicator MPI COMMUNICATOR 3 DUP FROM 0
with errorcode 1.
NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
[Hyper Tuner MPI 23.0.T30][INFO]enter _cluster_clean
[Hyper Tuner MPI 23.0.T30][ERROR]Task 1(statistics) fail.
[Hyper Tuner MPI 23.0.T30][INFO]********************************************************************************************************************************************************************
[Hyper Tuner MPI 23.0.T30][INFO]* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * PACK DATA * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
[Hyper Tuner MPI 23.0.T30][INFO]********************************************************************************************************************************************************************
[Hyper Tuner MPI 23.0.T30][INFO]Number of task: 1
[Hyper Tuner MPI 23.0.T30][ERROR]package failed!
2. 指定采样时间为-u时,出现以下warning信息,可以采样成功,但采样时间较长。
mpool.c:44 UCX WARN object 0x29fc0bc0 was not returned to mpool ucp_requests
mpool.c:44 UCX WARN object 0x29fc0d80 was not returned to mpool ucp_requests
mpool.c:44 UCX WARN object 0x29fc0f40 was not returned to mpool ucp_requests
mpool.c:44 UCX WARN object 0x29fc1100 was not returned to mpool ucp_requests
mpool.c:44 UCX WARN object 0x29fc12c0 was not returned to mpool ucp_requests
1. 使用命令行方式分析namd应用,使用-d指定采样时间为20时,会出现以下报错,采样失败;同样的采样参数在web界面也会报错退出;
70] Stack Traceback:
[70:0] linux-vdso.so.1 0xffff9e04693c __kernel_rt_sigreturn
[70:1] libucp.so.0 0xffff986bc970 ucp_worker_progress
[70:2] libucg_planc_ucx.so 0xffff9030b920 ucg_planc_ucx_p2p_testall
[70:3] libucg_planc_ucx.so 0xffff90306708
[70:4] libucg.so.1 0xffff981c2e74
[70:5] libucg.so.1 0xffff981c5fbc ucg_request_test
[70:6] mca_coll_ucg.so 0xffff981f2088 mca_coll_ucg_request_execute
[70:7] mca_coll_ucg.so 0xffff981f21d0 mca_coll_ucg_request_execute_cache
[70:8] mca_coll_ucg.so 0xffff981f5114 mca_coll_ucg_barrier_cache
[70:9] libmpi.so.40 0xffff9ada050c MPI_Barrier
[70:10] libmpicollector_ompi.so 0xffff9bf84850 MPI_Barrier
[70:11] namd3 0xb8e5e4
[70:12] namd3 0xb932b4
[70:13] namd3 0xb8cf0c
[70:14] namd3 0x473194
[70:15] namd3 0x46f790
[70:16] libc.so.6 0xffff9a975fbc
[70:17] libc.so.6 0xffff9a976094 __libc_start_main
[70:18] namd3 0x40f8f0
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 70 in communicator MPI COMMUNICATOR 3 DUP FROM 0
with errorcode 1.
NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
[Hyper Tuner MPI 23.0.T30][INFO]enter _cluster_clean
[Hyper Tuner MPI 23.0.T30][ERROR]Task 1(statistics) fail.
[Hyper Tuner MPI 23.0.T30][INFO]********************************************************************************************************************************************************************
[Hyper Tuner MPI 23.0.T30][INFO]* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * PACK DATA * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
[Hyper Tuner MPI 23.0.T30][INFO]********************************************************************************************************************************************************************
[Hyper Tuner MPI 23.0.T30][INFO]Number of task: 1
[Hyper Tuner MPI 23.0.T30][ERROR]package failed!
2. 指定采样时间为-u时,出现以下warning信息,可以采样成功,但采样时间较长。
mpool.c:44 UCX WARN object 0x29fc0bc0 was not returned to mpool ucp_requests
mpool.c:44 UCX WARN object 0x29fc0d80 was not returned to mpool ucp_requests
mpool.c:44 UCX WARN object 0x29fc0f40 was not returned to mpool ucp_requests
mpool.c:44 UCX WARN object 0x29fc1100 was not returned to mpool ucp_requests
mpool.c:44 UCX WARN object 0x29fc12c0 was not returned to mpool ucp_requests