|
Problem IT38875
|
Status: Closed |
MEMORY CORRUPTION IN CDR MEMORY POOL WHEN CDR_MEM=2 USED
|
| product: |
INFORMIX SERVER / 5725A3900 / E10 - |
| Problem description: |
When ER is defined across multiple platforms and replication is
happening on multiple tables, then we can see memory
corruption errors occurring with different assert
failures/warnings to be seen on the target server within the CDR
Memory pool (CDRmp).
We see this memory corruption only when we are running with
CDR_MEM=2 configuration.
10/28/2021 13:09:15 Found during mt_shm_free 1, error seen 1
times since last print to the log
10/28/2021 13:09:15 Pool 'CDRmp_1' (0x4aa69040)
10/28/2021 13:09:15 Bad block header 0x4bb336f0
10/28/2021 13:09:15 Assert Failed: Memory block header
corruption detected in mt_shm_free 1
10/28/2021 13:09:15 IBM Informix Dynamic Server Version
14.10.FC6
10/28/2021 13:09:15 Who: Session(185, informix@darkwood, 0,
0x48f92178)
Thread(199, CDRD_5, 48f524a0, 1)
File: mtshpool.c Line: 2796
10/28/2021 13:09:15 Results: Unable to repair pool
10/28/2021 13:09:15 Action: Please notify IBM Informix
Technical Support.
10/28/2021 13:09:15 stack trace for pid 23306 written to
/chunks/ifx_1410/msg/af.4af84db
10/28/2021 13:09:15 See Also: /chunks/ifx_1410/msg/af.4af84db
10/28/2021 13:09:21 Memory block header corruption detected in
mt_shm_free 1
The following stack trace is common:
0x000000000183c8da (oninit) afstack
0x0000000001839f08 (oninit) afhandler
0x0000000001839325 (oninit) affail_interface
0x00000000018864d7 (oninit) mt_free
0x00000000013da383 (oninit) cdrMemFreeRtn
0x00000000014a7977 (oninit) dsiGetReplRow
0x000000000149b4aa (oninit) dsiProcessRow
0x0000000001497de9 (oninit) dsiResolverThread
0x000000000135e598 (oninit) cdrTrampolineThread
0x000000000181fdf8 (oninit) th_init_initgls
0x00000000018710bf (oninit) startup |
| Problem Summary: |
****************************************************************
* USERS AFFECTED: *
* 14.10.xC6 onwards *
****************************************************************
* PROBLEM DESCRIPTION: *
* With CDR_MEM=2, we are observing the memory corruption *
* errors in in CDRmp_ pool *
****************************************************************
* RECOMMENDATION: *
**************************************************************** |
| Local Fix: |
Change the CDR_MEM setting to 0/1. |
| Solution |
|
| Workaround |
****************************************************************
* USERS AFFECTED: *
* 14.10.xC6 onwards *
****************************************************************
* PROBLEM DESCRIPTION: *
* With CDR_MEM=2, we are observing the memory corruption *
* errors in in CDRmp_ pool *
****************************************************************
* RECOMMENDATION: *
**************************************************************** |
| Comment |
As per my previous remark, please suggest the customer to use
CDR_MEM=0. This is not required in the 14.10.xC6 release onwards
as the functionality is supported automatically. |
| Timestamps |
Date - problem reported :
Date - problem closed :
Date - last modified :
| 29.10.2021
25.11.2022
25.11.2022 |
| Problem solved at the following versions (IBM BugInfos) |
|
| Problem solved according to the fixlist(s) of the following version(s) |
