一次GI補丁安裝不完整導致的RAC心跳流量異常

一、系統環境

伺服器：IBM3850 X86兩節點RAC

作業系統：redhat6.8

資料庫版本：11.2.0.4

二、故障介紹

巡檢中發現資料庫未安裝任何補丁，停機安裝DB PSU後，出現 gc cr block lost/gc current block lost 等待事件，原本私網網路卡的大幀配置失效。

三、原因分析

檢查資料庫的私網資訊

SQL> select * from x$ksxpia;

ADDR INDX INST_ID PUB_KSXPIA PICKED_KSXPIA NAME_KSXPIA IP_KSXPIA

---------------- ---------- ---------- ---------- ----------------------------------- --------------- ----------------

00007F528A46E768 0 1 N OSD bond0 10.16.9.1

SQL> select * from v$configured_interconnects;

NAME IP_ADDRESS IS_ SOURCE

--------------- ---------------- --- -------------------------------

bond0 10.16.9.1 NO OS dependent software

[grid@dbsvr1 ~]$ oifcfg getif

bond0 10.16.9.0 global public

bond1 192.168.10.0 global cluster_interconnect

CRS-2302:Cannot get GPnP profile. Error CLSGPNP_NO_DAEMON (GPNPD daemon is not running).

[USER(18925)]CRS-2316:Fatal error: cannot initialize GPnP, CLSGPNP_ERR (Generic GPnP error).

kggpnpInit: failed to init gpnp

WARNING: No cluster interconnect has been specified. Depending on the communication driver configured Oracle cluster traffic may be directed to the public interface of this machine.

Oracle recommends that RAC clustered databases be configured with a private interconnect for enhanced security and performance.

檢查叢集profile.xml檔案

[grid@dbsvr1 peer]$ cat profile.xml

<?xml version="1.0" encoding="UTF-8"?><gpnp:GPnP-Profile Version="1.0" xmlns=" xmlns:gpnp=" xmlns:orcl=" xmlns:xsi=" xsi:schemaLocation=" gpnp-profile.xsd" ProfileSequence="4" ClusterUId="8de8f5467541dfa1bf17fd038b8aec7c" ClusterName="dbsvr-cluster" PALocation="">

<gpnp:Network-Profile>

<gpnp:HostNetwork id="gen" HostName="*">

<gpnp:Network id="net1" IP="10.16.9.0" Adapter="bond0" Use="public"/>

<gpnp:Network id="net2" IP="192.168.10.0" Adapter="bond1" Use="cluster_interconnect"/>

</gpnp:HostNetwork>

</gpnp:Network-Profile>

<orcl:CSS-Profile id="css" DiscoveryString="+asm" LeaseDuration="400"/><orcl:ASM-Profile id="asm" DiscoveryString="/dev/mapper" SPFile="+OCR_VOTE/dbsvr-cluster/asmparameterfile/registry.253.973884379"/><ds:Signature xmlns:ds=" Algorithm=" Algorithm=" URI=""><ds:Transforms><ds:Transform Algorithm=" Algorithm=" <InclusiveNamespaces xmlns=" PrefixList="gpnp orcl xsi"/></ds:Transform></ds:Transforms><ds:DigestMethod Algorithm="

檢查發現群集的public網段為10.16.9.0，private網段為192.168.10.0

但實際資料庫中卻使用10.16.9.0作為private網路使用，造成原本私網網路卡的大幀配置失效。

四、解決方案

方案一：

登入例項並修改 cluster_interconnects 引數後重啟所有例項

alter system set cluster_interconnects='169.254.232.49' scope=spfile sid='orcl1';

alter system set cluster_interconnects='169.254.144.102' scope=spfile sid='orcl2';

方案二：

同時應用GI PSU避免此問題