|
Problem IC67553
|
Status: Closed |
DB2 CRASH WHEN CTRL+C IS PRESSED DURING MDC REFRESHING, WHEN NOT LOGGED INITIALLY IS ACTIVATED |
| product: |
DB2 FOR LUW / DB2FORLUW / 950 - DB2 |
| Problem description: |
DB2 might crash when interrupt is received during MDC
refreshing, when NOT LOGGED INITIALLY is activated.
One of the symptom is showing the following stack in db2diag.log
(or in trap file)
2010-03-24-13.46.03.808277-240 I2069921E1331 LEVEL: Error
PID : 20752 TID : 198499625312 PROC :
db2sysc 5
INSTANCE: nypxt400 NODE : 005 DB : C1RISK
APPHDL : 1-37312 APPID:
10.163.26.167.16280.10032417454
AUTHID : CPGDBPRD
EDUID : 267757 EDUNAME: db2agnta (C1RISK) 5
FUNCTION: DB2 UDB, sqldb, sqldb_purgeAllRolloutCBsFromTENTRY,
probe:548
MESSAGE : Rollout CB list is not empty at tx boundary!
DATA #1 : Pointer, 8 bytes
0x0000002df3433080
CALLSTCK:
[0] 0x0000002A966E4509 pdLog + 0xD7
[1] 0x0000002A97113363
_Z34sqldb_purgeAllRolloutCBsFromTENTRYP8sqeAgent + 0x16F
[2] 0x0000002A96AD7FAB
_Z17sqlpEndUowRuntimeP8sqeAgentP9SQLP_DBCB + 0x7E7
[3] 0x0000002A96AD3A22
/ms/dist/ibmdb2/PROJ/ds/9.5.4.10/.exec/x86_64.linux.2.6.glibc.2.
3/lib64/libdb2e.so.1
+ 0x1351A22
[4] 0x0000002A96AD202F
_Z8sqlpxrbkP8sqeAgentP15SQLXA_CALL_INFOPiP9SQLP_GXIDPP11sqlo_xla
tch
+ 0x497
[5] 0x0000002A96B3F6CC
_Z12sqlrrbck_dpsP8sqlrr_cbiiiP15SQLXA_CALL_INFOP9SQLP_GXID +
0x106
[6] 0x0000002A97F35F13 _Z17sqlrr_tran_routerP8sqlrr_cb + 0x85F
[7] 0x0000002A97F20F5C
_Z21sqlrr_subagent_routerP8sqeAgentP12SQLE_DB2RA_T + 0xB42
[8] 0x0000002A971B51D1 _Z20sqleSubRequestRouterP8sqeAgentPjS1_
+ 0xA83
[9] 0x0000002A971B4147 _Z21sqleProcessSubRequestP8sqeAgent +
0x5F
The problem is that sqldRolloutBlocksOfCell passes the master
TCB to sqldb_allocateAndLinkRolloutCB, which then uses it to
check if Not Logged Initially is activated for the table.
However the TCB_D_NOLOG is not stored in the master TCB, only in
the TCB for the partitions. |
| Problem Summary: |
Problem Description:
DB2 CRASH WHEN CTRL+C IS PRESSED DURING MDC REFRESHING, WHEN NOT
LOGGED INITIALLY IS ACTIVATED
Problem Summary:
DB2 might crash when interrupt is received during MDC
refreshing, when NOT LOGGED INITIALLY is activated.
One of the symptom is showing the following stack in db2diag.log
(or in trap file)
2010-03-24-13.46.03.808277-240 I2069921E1331 LEVEL: Error
PID : 20752 TID : 198499625312 PROC :
db2sysc 5
INSTANCE: nypxt400 NODE : 005 DB : C1RISK
APPHDL : 1-37312 APPID:
10.163.26.167.16280.10032417454
AUTHID : CPGDBPRD
EDUID : 267757 EDUNAME: db2agnta (C1RISK) 5
FUNCTION: DB2 UDB, sqldb, sqldb_purgeAllRolloutCBsFromTENTRY,
probe:548
MESSAGE : Rollout CB list is not empty at tx boundary!
DATA #1 : Pointer, 8 bytes
0x0000002df3433080
CALLSTCK:
[0] 0x0000002A966E4509 pdLog + 0xD7
[1] 0x0000002A97113363
_Z34sqldb_purgeAllRolloutCBsFromTENTRYP8sqeAgent + 0x16F
[2] 0x0000002A96AD7FAB
_Z17sqlpEndUowRuntimeP8sqeAgentP9SQLP_DBCB + 0x7E7
[3] 0x0000002A96AD3A22
/ms/dist/ibmdb2/PROJ/ds/9.5.4.10/.exec/x86_64.linux.2.6.glibc.2.
3/lib64/libdb2e.so.1
+ 0x1351A22
[4] 0x0000002A96AD202F
_Z8sqlpxrbkP8sqeAgentP15SQLXA_CALL_INFOPiP9SQLP_GXIDPP11sqlo_xla
tch
+ 0x497
[5] 0x0000002A96B3F6CC
_Z12sqlrrbck_dpsP8sqlrr_cbiiiP15SQLXA_CALL_INFOP9SQLP_GXID +
0x106
[6] 0x0000002A97F35F13 _Z17sqlrr_tran_routerP8sqlrr_cb + 0x85F
[7] 0x0000002A97F20F5C
_Z21sqlrr_subagent_routerP8sqeAgentP12SQLE_DB2RA_T + 0xB42
[8] 0x0000002A971B51D1 _Z20sqleSubRequestRouterP8sqeAgentPjS1
+ 0xA83
[9] 0x0000002A971B4147 _Z21sqleProcessSubRequestP8sqeAgent +
0x5F
The problem is that sqldRolloutBlocksOfCell passes the master
TCB to sqldb_allocateAndLinkRolloutCB, which then uses it to
check if Not Logged Initially is activated for the table.
However the TCB_D_NOLOG is not stored in the master TCB, only in
the TCB for the partitions. |
| Local Fix: |
n/a |
| available fix packs: |
DB2 Version 9.5 Fix Pack 6a for Linux, UNIX, and Windows
DB2 Version 9.5 Fix Pack 7 for Linux, UNIX, and Windows
DB2 Version 9.5 Fix Pack 8 for Linux, UNIX, and Windows
DB2 Version 9.5 Fix Pack 9 for Linux, UNIX, and Windows
DB2 Version 9.5 Fix Pack 10 for Linux, UNIX, and Windows
|
| Solution |
Problem is first fixed in Version 9.5 Fix Pack 6 |
| Workaround |
n/a |
| BUG-Tracking |
forerunner : APAR is sysrouted TO one or more of the following: IC67554 IC67634
follow-up : |
| Timestamps |
Date - problem reported :
Date - problem closed :
Date - last modified :
| 30.03.2010
13.05.2010
13.05.2010 |
| Problem solved at the following versions (IBM BugInfos) |
9.5.FP6
|
| Problem solved according to the fixlist(s) of the following version(s) |
