Skip to content

遭遇ora-600[kjblpkeydrmqscchk:pkey]

平台为 linux x86_64
db版本 11.2.0.3

现象节点2,3,4自动重启.有session挂住集中出现等待事件gc buffer busy acquire.
诊断为bug 14409183

具体如下:

Fixed:

The fix for 14409183 is first included in
12.1.0.1 (Base Release)
11.2.0.4 (Server Patch Set)
11.2.0.3.4 Database Patch Set Update
11.2.0.3 Bundle Patch 10 for Exadata Database
11.2.0.3 Patch 10 on Windows Platforms

Interim patches may be available for earlier versions - click here to check.
Symptoms:

Related To:

Instance May Crash
Process May Dump (ORA-7445) / Abend / Abort
Hang (Process Hang)
Internal Error May Occur (ORA-600)
ORA-600 [kjblpkeydrmqscchk:pkey]
ORA-600 [kjbldrmrpst:pkey]
ORA-600 [kjbrchkpkeywait:timeout]
ORA-600 [kclcls_8]
ORA-600 [kjbmmchkinteg:from]
ORA-600 [kclantilock_5]
ORA-600 [kclchkinteg_8]
ORA-600 [kclexpand_9]
Waits for "gc buffer busy acquire"
RAC (Real Application Clusters) / OPS
_gc_read_mostly_locking
_gc_read_mostly_locking
Description

This bug is only relevant when using Real Application Clusters (RAC)
The fix for bug 13397104 addressed some potential ORA-600 issues in RAC
*BUT* the fix was faulty and can cause a hang scenario in RAC with sessions waiting on
"gc buffer busy acquire" READING buffer's pin is not cleaned up properly during a
previous exception(ORA-8103).

This fix supersedes the 13397104 and should be used instead of that fix
for interim patches. It addresses both the symptoms of bug 13397104 and the
follow on hang issue.

Rediscovery Notes:
If you have the fix for bug 13397104 present and see the following symptoms then
this bug should be suspected:
1. Fix 13397104 is present on the system,
2. RAC env,
3. Sessions wait for a READING buffer with a dangling pin.

If you do not have the fix for bug 13397104 but see the symptoms of that bug#
then this fix should be used. Symptoms of that issue include:
- Various ora-600's such as:
ora-600 [kjblpkeydrmqscchk:pkey]
ora-600 [kjbldrmrpst:pkey]
ora-600 [kjbrchkpkeywait:timeout]
in 11.2.0.3 where fix for bug 10170926 is present
- the LE flag includes KCLL_F_ANTI (0x080)
ie. flags: 0xa1
- ORA-600 [kclcls_8] due to a mismatch from a guess-dba

Workaround
The original problem in bug 13397104 can be avoided by
disabling read mostly locking via: _gc_read_mostly_locking=false

Getting a Fix
Use one of the "Fixed" versions listed above
(for Patch Sets / bundles use the latest version available as
contents are cumulative - the "Fixed" version listed above is
the first version where the fix is included)
or
Click here for suggestions on how to get a fix for this issue

Please note: The above is a summary description only. Actual symptoms can vary. Matching to any symptoms here does not confirm that you are encountering this problem. For questions about this bug please consult Oracle Support.