Symptom
- Very high latency in the inbound queue for non-warm standby connections.
- The 'procstack' output for the SRS process shows a lot of threads in the _global_lock_common() and _global_unlock_common() calls as memory is allocated or freed via malloc/free calls, for example:
....
---------- tid# 85983339 (pthread ID: 26475) ----------
0x09000000006e47b4 _global_lock_common(??, ??, ??) + 0x4d4
0x0900000000099b70 malloc_y(0x200, 0x0, 0x9001000a006ebb0, 0x11005f048, 0x1, 0x116829710, 0x1007f4b58, 0x116829710) + 0x7f0
0x0900000000010748 malloc_common@AF102_86(??) + 0x28
0x0000000100214550 mem_memAlloc(??, ??, ??) + 0x130
0x0000000100214d34 memSegGetSeg(??, ??, ??, ??, ??, ??) + 0x74
0x0000000100214a74 mem__Allocate(??, ??, ??, ??, ??, ??, ??, ??) + 0xd4
0x00000001005ecb78 md_deliver_message(??, ??, ??, ??, ??, ??, ??, ??) + 0x178
0x00000001006ee580 IPRA.$td_deliver_command(??, ??, ??, ??, ??, ??, ??) + 0xc40
0x00000001006ef2bc td_deliver_command(??, ??, ??, ??, ??, ??, ??) + 0x5c
0x00000001006e74c0 _dist_applied_cmd(0x6d0000006d, 0x116829710, 0xa1400000a14, 0x132175600, 0x13239b2c0, 0x132394020, 0x13250b520, 0x1325922c0) + 0x2ba0
0x00000001006dc448 _dist_service(??, ??) + 0x2a08
0x00000001006d8c04 dist_wrapper(??) + 0xfc4
0x0000000100009aac srv__start_function(??) + 0x2cc
0x09000000006e5d50 _pthread_body(??) + 0xf0
---------- tid# 77660339 (pthread ID: 33644) ----------
0x09000000006e47b4 _global_lock_common(??, ??, ??) + 0x4d4
0x0900000000097a5c free_y(??, ??) + 0x11c
0x090000000000fd18 free_common(??) + 0xb8
0x000000010021688c memFree@AF30_6(??) + 0x12c
0x0000000100216afc mem_free@AF31_11(??, ??) + 0x21c
0x000000010026cb80 _sqt_del_from_cache(??, ??, ??, ??) + 0x840
0x000000010026b790 _sqt_service(??, ??) + 0x470
0x00000001002697e8 _sqt_wrap(??) + 0x2e8
0x0000000100009aac srv__start_function(??) + 0x2cc
0x09000000006e5d50 _pthread_body(??) + 0xf0
---------- tid# 65798391 (pthread ID: 30165) ----------
0x09000000006e33f0 _global_unlock_common(??, ??) + 0x10
0x0900000000099b70 malloc_y(0x800, 0x0, 0x9001000a006ebb0, 0x11005f048, 0x1, 0x11a89d890, 0x1007f24a4, 0x11a89d890) + 0x7f0
0x0900000000010748 (??) + 0x28
0x0000000100214550 mem_memAlloc(??, ??, ??) + 0x130
0x0000000100214d34 memSegGetSeg(??, ??, ??, ??, ??, ??) + 0x74
0x0000000100214bf4 mem__Allocate(??, ??, ??, ??, ??, ??, ??, ??) + 0x254
0x00000001005c0f5c yyparse(??) + 0xfdc
0x00000001005bfe48 prs_parse_sql(??, ??) + 0x28
0x00000001005bed94 pak_unpack_command(??, ??, ??, ??, ??, ??) + 0xa74
0x000000010055fefc _dsi_parse_cmd(??, ??, ??, ??, ??, ??, ??) + 0x7c
0x00000001006866d4 dsie<...>(...)(??, ??, ??, ??) + 0x374
0x0000000100685a58 dsie__ProcessGroup(??, ??, ??) + 0x458
0x0000000100684b1c dsie__main(??, ??) + 0x27c
0x00000001006842e4 dsie__wrap(??) + 0x864
0x0000000100009aac srv__start_function(??) + 0x2cc
0x09000000006e5d50 _pthread_body(??) + 0xf0
... - This started after SRS was upgraded from 15.6 to 15.7.1.
Read more...
Environment
- SAP Replication Server (SRS) 15.7.1
- AIX 6.1 and 7.1
Product
Keywords
malloc, free, simultaneous, AIXTHREAD_MUTEX_FAST, MALLOCOPTIONS, YIELDLOOPTIME, SPINLOOPTIME, multiheap , KBA , BC-SYB-REP , Sybase Replication Server (standalone) , Problem
About this page
This is a preview of a SAP Knowledge Base Article. Click more to access the full version on SAP for Me (Login required).Search for additional results
Visit SAP Support Portal's SAP Notes and KBA Search.
SAP Knowledge Base Article - Preview