rac 私网相关问题

2^#

发表于 2012-5-11 20:48:57

请做一个ocrdump 并上传

action plan:

RUN STATEMENT ON 2 NODES:

select INST_ID,PUB_KSXPIA,PICKED_KSXPIA,NAME_KSXPIA,IP_KSXPIA from x$ksxpia;

olsnodes -n -p

3^#

发表于 2012-5-12 09:22:09

first node:
SQL> select INST_ID,PUB_KSXPIA,PICKED_KSXPIA,NAME_KSXPIA,IP_KSXPIA from x$ksxpia;

INST_ID PUB_KSXPIA PICKED_KSXPIA                      NAME_KSXPIA
---------- ---------- ----------------------------------- ---------------
IP_KSXPIA
----------------
      1 N       CI                               aggr2
10.10.10.1

SQL> exit
Disconnected from Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP, Data Mining
and Real Application Testing options
-bash-3.00$ olsnodes -n -p
siddb01 0    siddb01_crs_hb
esbdb01 1    esbdb01_crs_hb
second node:

-bash-3.00$ rlogin esbdb01_crs_hb
Last login: Thu Apr 12 10:42:36 from 132.42.33.110
Oracle Corporation    SunOS 5.10    Generic Patch January 2005
-bash-3.00$ sqlplus "/ as sysdba"
SQL*Plus: Release 10.2.0.5.0 - Production on Sat May 12 09:20:50 2012
Copyright (c) 1982, 2010, Oracle.  All Rights Reserved.

Connected to:
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP, Data Mining
and Real Application Testing options
SQL> select INST_ID,PUB_KSXPIA,PICKED_KSXPIA,NAME_KSXPIA,IP_KSXPIA from x$ksxpia;
INST_ID PUB_KSXPIA PICKED_KSXPIA                      NAME_KSXPIA
---------- ---------- ----------------------------------- ---------------
IP_KSXPIA
----------------
      2 N       CI                               aggr2
10.10.10.2

SQL> exit
Disconnected from Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP, Data Mining
and Real Application Testing options
-bash-3.00$ olsnodes -n -p
siddb01 0    siddb01_crs_hb
esbdb01 1    esbdb01_crs_hb

4^#

发表于 2012-5-12 09:27:11

ocrdump is comming..

ocrdump.txt

36.84 KB, 下载次数: 828

5^#

发表于 2012-5-12 10:52:48

ODM FINDING:

[SYSTEM.css.interfaces.global]
UNDEF :
SECURITY : {USER_PERMISSION : PROCR_ALL_ACCESS, GROUP_PERMISSION : PROCR_ALL_ACCESS, OTHER_PERMISSION : PROCR_READ, USER_NAME : oracle, GROUP_NAME : dba}

[SYSTEM.css.interfaces.global.aggr1]
UNDEF :
SECURITY : {USER_PERMISSION : PROCR_ALL_ACCESS, GROUP_PERMISSION : PROCR_ALL_ACCESS, OTHER_PERMISSION : PROCR_READ, USER_NAME : oracle, GROUP_NAME : dba}

[SYSTEM.css.interfaces.global.aggr1.132|d42|d42|d0]
UNDEF :
SECURITY : {USER_PERMISSION : PROCR_ALL_ACCESS, GROUP_PERMISSION : PROCR_ALL_ACCESS, OTHER_PERMISSION : PROCR_READ, USER_NAME : oracle, GROUP_NAME : dba}

[SYSTEM.css.interfaces.global.aggr1.132|d42|d42|d0.1]
ORATEXT : public
SECURITY : {USER_PERMISSION : PROCR_ALL_ACCESS, GROUP_PERMISSION : PROCR_ALL_ACCESS, OTHER_PERMISSION : PROCR_READ, USER_NAME : oracle, GROUP_NAME : dba}

[SYSTEM.css.interfaces.global.aggr2]
UNDEF :
SECURITY : {USER_PERMISSION : PROCR_ALL_ACCESS, GROUP_PERMISSION : PROCR_ALL_ACCESS, OTHER_PERMISSION : PROCR_READ, USER_NAME : oracle, GROUP_NAME : dba}

[SYSTEM.css.interfaces.global.aggr2.10|d10|d10|d0]
UNDEF :
SECURITY : {USER_PERMISSION : PROCR_ALL_ACCESS, GROUP_PERMISSION : PROCR_ALL_ACCESS, OTHER_PERMISSION : PROCR_READ, USER_NAME : oracle, GROUP_NAME : dba}

[SYSTEM.css.interfaces.global.aggr2.10|d10|d10|d0.1]
ORATEXT : cluster_interconnect
SECURITY : {USER_PERMISSION : PROCR_ALL_ACCESS, GROUP_PERMISSION : PROCR_ALL_ACCESS, OTHER_PERMISSION : PROCR_READ, USER_NAME : oracle, GROUP_NAME : dba}

[SYSTEM.css.interfaces.global.aggr1.132|d42|d42|d0.1]
ORATEXT : public

[SYSTEM.css.interfaces.global.aggr2.10|d10|d10|d0.1]
ORATEXT : cluster_interconnect

ocr 中的记录显示正常

x$ksxpia 中的信息也正常

6^#

发表于 2012-5-12 11:30:39

有点不明白为什么：
SQL> select * from gv$cluster_interconnects;
INST_ID NAME          IP_ADDRESS    IS_ SOURCE
---------- --------------- ---------------- --- -------------------------------
      2 aggr1          132.42.42.6    YES OS dependent software
      1 aggr2          10.10.10.1    NO  Oracle Cluster Repository
显示不正常了呢？？
麻烦您帮忙看看。

7^#

发表于 2012-5-12 13:19:29

GV$CLUSTER_INTERCONNECTS

is

SELECT INST_ID,
   NAME_SKGXPIA,
   IP_SKGXPIA,
   decode(PUB_SKGXPIA, 'Y', 'YES', 'N', 'NO'),
   decode(PICKED_SKGXPIA,
            'OSD',
            'OS dependent software',
            'OCR',
            'Oracle Cluster Repository',
            'CI',
            'cluster_interconnects parameter')
  FROM X$SKGXPIA

CI  代表 cluster_interconnects parameter

set linesize 100 pagesize 1000

select * from  X$SKGXPIA;

oradebug setmypid;
oradebug ipc;
oradebug tracefile_name

upload  the trace file

8^#

发表于 2012-5-12 21:19:35

oradebug信息见附件。

clusterinter.txt

23.31 KB, 下载次数: 634

9^#

发表于 2012-5-12 21:29:36

ODM DATA:

ADDR                INDX INST_ID PUB_SKGXPI PICKED_SKGXPIA
---------------- ---------- ---------- ---------- -----------------------------------
NAME_SKGXPIA IP_SKGXPIA
--------------- ----------------
0000000A1FEAEB48       0       1 N       CI
aggr2          10.10.10.1

SQL> select * from  X$SKGXPIA;

ADDR                INDX INST_ID PUB_SKGXPI PICKED_SKGXPIA
---------------- ---------- ---------- ---------- -----------------------------------
NAME_SKGXPIA IP_SKGXPIA
--------------- ----------------
0000000A1FEAEB48       0       2 N       CI
aggr2          10.10.10.2

SSKGXPT 0x6c10000 flags SSKGXPT_READPENDING    socket no 8    IP 10.10.10.2 UDP 47879
last bytes received: 32824
context timestamp 0
done Queue
      no completed requests
port Queue
      no ports
connection Queue
      no pending connect disconnect operations
sends waiting to be transmitted
      no sends waiting to be transmitted
pending ack Queue
      no send requests pending ack
Mapped regions
      Region[0] Id 342948099 Base Address 381000000 Size -1627389952 key 764646201
      rgnport 106c0f2f4 lbuf 0 nrgns 1 flags 1
      SSKGXPT 0x6c0f2f4 flags       socket no 14 IP 10.10.10.2 UDP 47880

SSKGXPT 0x6c10000 flags SSKGXPT_READPENDING    socket no 8    IP 10.10.10.1 UDP 65094
last bytes received: 32824
context timestamp 0
done Queue
      no completed requests
port Queue
      no ports
connection Queue
      no pending connect disconnect operations
sends waiting to be transmitted
      no sends waiting to be transmitted
pending ack Queue
      no send requests pending ack
Mapped regions
      Region[0] Id 1852261323 Base Address 381000000 Size -1627389952 key 587274849
      rgnport 106c0f2f4 lbuf 0 nrgns 1 flags 1
      SSKGXPT 0x6c0f2f4 flags       socket no 16 IP 10.10.10.1 UDP 65095

就IPC信息看使用的interconnect interface 和 ip 也是正常的。

存在一些bug 可能使 gv$cluster_interconnects 的信息显示不正确，但是目前没有找到 10.2.0.5 上已知的bug 。

10^#

发表于 2012-5-12 22:41:01

搞不懂了。我现在没搞明白为啥interconnect显示不正常。。。ocrdump的信息，看起来应该没问题的。

11^#

发表于 2012-5-12 22:47:38

可以从 MOS 上找到的 V$CLUSTER_INTERCONNECTS 相关的bug

Bug 10004875 : WHY V$CLUSTER_INTERCONNECTS AND GV$CLUSTER_INTERCONNECTS ARE EMPTY? 10.2.0.3
Bug 7445346 : RAC: V$CLUSTER_INTERCONNECTS SHOWING WRONG INFORMATION 10.2.0.3

As Maclean think:

这应当是一个可以ignore的bug 或者 behavior ，建议你提交一个SR确认。