IBM IDS ENTRP E / 5724L2304 / B15 - IDS 11.50

IDS hang can occur during smart blob kaio request. 
 
When this occurs you will see a kaio thread continually running 
with 
stack similar to: 
 
09:33:20  Raw hex dump of stack located in 
/tmp/af.43f71b0.rawstk 
09:33:20  Stack for thread: 63 kaio 
 
 base: 0x000000004dfde000 
  len:   69632 
   pc: 0x0000000000fe3267 
  tos: 0x0000000001706bb0 
state: running 
   vp: 1 
 
0x0000000000fe3267 (oninit) afstack 
0x0000000000fe3745 (oninit) afstack_dump 
0x0000000000fb2770 (oninit) notifyvp_signal_handler 
0x000000361120de70 (Linux) <signal frame> 
0x0000000000fdf5e4 (oninit) kaiothread 
0x0000000000fb878b (oninit) startup 
 
and you will also see an sqlexec thread in ready state waiting 
in sb_lt_read()/sb_lt_wait() 
for kaio request similar to: 
 
Stack for thread: 67 sqlexec 
 base: 0x000000004d2e3000 
  len:   69632 
   pc: 0x0000000000fb9db0 
  tos: 0x000000004d2ee8e0 
state: ready 
   vp: 1 
 
0x0000000000fb9db0 (oninit) yield_processor_mvp 
0x0000000000fad7dc (oninit) mt_yield 
0x0000000000f0bf77 (oninit) sb_lt_wait 
0x0000000000f0e672 (oninit) sb_lt_read 
0x0000000000ef9329 (oninit) sb_readwithseek 
0x0000000000ed7898 (oninit) sbx_readwithseek 
0x00000000007e0bab (oninit) sq_lo_read 
0x0000000000551431 (oninit) mi_lo_read 
0x00002b11eef63593 (/prod/extend/ETX.1.31.FC1/ETX.bld) 
IfxFseRead 
0x00002b11eef3fc29 (/prod/extend/ETX.1.31.FC1/ETX.bld) fse_read 
0x00002b11eef402b0 (/prod/extend/ETX.1.31.FC1/ETX.bld) 
trl_read_bytes 
0x00002b11eef7de10 (/prod/extend/ETX.1.31.FC1/ETX.bld) 
load_sec_page 
0x00002b11eef77981 (/prod/extend/ETX.1.31.FC1/ETX.bld) 
trl_search_data 
0x00002b11eef742c5 (/prod/extend/ETX.1.31.FC1/ETX.bld) 
trl_pattern_search 
0x00002b11eef4914d (/prod/extend/ETX.1.31.FC1/ETX.bld) 
etxgl_pattern_search 
0x00002b11eef46c70 (/prod/extend/ETX.1.31.FC1/ETX.bld) 
etxgl_begin_scan 
0x00002b11eef365c9 (/prod/extend/ETX.1.31.FC1/ETX.bld) 
etx_begin_scan 
0x00000000008f8691 (oninit) udrlm_clang_execute_internal 
0x00000000008fa3b9 (oninit) udrlm_clang_execute 
0x0000000000897986 (oninit) udrlm_exec_routine 
0x0000000000624791 (oninit) udr_execute 
0x0000000001131802 (oninit) vti_vii_udr_execute 
0x000000000112e493 (oninit) vii_beginscan 
0x000000000112eb5f (oninit) vii_getnext 
0x000000000102d86c (oninit) fmread 
0x000000000070111d (oninit) gettupl 
0x0000000000707c29 (oninit) scan_next 
0x0000000000720faa (oninit) sort_open 
0x000000000070f2d5 (oninit) prepselect 
0x00000000007382f0 (oninit) subqprep 
0x0000000000707732 (oninit) scan_next 
0x000000000099838a (oninit) next_row 
0x0000000000998a61 (oninit) get_first_row_from_producer 
0x000000000099b68f (oninit) group_open 
0x000000000070f2d5 (oninit) prepselect 
0x0000000000984ce3 (oninit) excommand 
0x000000000085f9e7 (oninit) sq_exselect 
0x0000000000917626 (oninit) sqmain 
0x00000000010634b3 (oninit) spawn_thread 
0x0000000000fbaecb (oninit) startup

**************************************************************** 
* USERS AFFECTED:                                              * 
* Users of smart blobs with concurrent reads and updates of    * 
* the same blobs                                               * 
**************************************************************** 
* PROBLEM DESCRIPTION:                                         * 
* If a smart blob is being read via light scan and the blob is * 
* updated by another session during the scan, it is possible   * 
* for the IO subsystem of IDS to hang or crash.                * 
**************************************************************** 
* RECOMMENDATION:                                              * 
* Upgrade to IDS 11.50.xC5 and above.                          * 
****************************************************************

First fixed in IDS 11.50.xC5.

not known / see Local fix

Date  - problem reported    :
Date  - problem closed      :
Date  - last modified       :

18.06.2009
21.01.2010
21.01.2010

11.50.xC5