My Oracle Support Banner

rootupgrade.sh で PROC-26: [Insufficient quorum to open OCR devices] [0] が発生する (Doc ID 2266468.1)

Last updated on FEBRUARY 03, 2019

適用範囲:

Oracle Database - Enterprise Edition - バージョン 12.1.0.2 から 12.1.0.2 [リリース 12.1]
この文書の内容はすべてのプラットフォームに適用されます。
本文書利用上のご注意
  本文書は英語の文書 <Document 2227363.1> (最終メジャー更新日: 2017年02月02日) の日本語翻訳版です。
  英語の文書の最終メジャー更新日が新しい場合は本文書に適宜反映いたします。

現象

Grid Infrastructure を 11.2 から 12.1.0.2 にアップグレードする際、rootupgrade.sh が node1 で失敗します。 

CRS-4123: Oracle High Availability Services has been started.
2016/12/11 12:05:12 CLSRSC-472: Attempting to export the OCR

2016/12/11 12:05:12 CLSRSC-482: Running command: 'ocrconfig -upgrade oracle oradba'

PROT-1: Failed to initialize ocrconfig
PROC-26: Error while accessing the physical storage Storage layer error [Insufficient quorum to open OCR devices] [0]
2016/12/11 12:05:13 CLSRSC-157: Failed to create Oracle Cluster Registry configuration

2016/12/11 12:05:13 CLSRSC-248: Failed to upgrade Oracle Clusterware Repository

Died at /orab/app/oracle/1210/gridinf/crs/install/crsupgrade.pm line 906.

 

node1 の ASM alert.log には、次のような出力があります。 

GMON dismounting group 37 at 174 for pid 23, osid 9985
NOTE: Disk DG_NADCLPDBSCCMC2A_VOTE__0000 in mode 0x7f marked for de-assignment
NOTE: Disk DG_NADCLPDBSCCMC2A_VOTE__0001 in mode 0x7f marked for de-assignment
NOTE: Disk DG_NADCLPDBSCCMC2A_VOTE__0002 in mode 0x7f marked for de-assignment
NOTE: Disk DG_NADCLPDBSCCMC2A_VOTE__0003 in mode 0x7f marked for de-assignment
NOTE: Disk DG_NADCLPDBSCCMC2A_VOTE__0004 in mode 0x7f marked for de-assignment
SUCCESS: diskgroup DG_NADCLPDBSCCMC2A_VOTE_01 was dismounted
NOTE: cache deleting context for group DG_NADCLPDBSCCMC2A_VOTE_01 37/0xfab98400
SUCCESS: ALTER DISKGROUP ALL DISMOUNT /* asm agent *//* {0:0:15891} */
Sun Dec 11 12:02:17 2016

====>> 当該ディスクグループ (diskgroup number 37) は、11.2 ホームからのスタック停止で正常にディスマウントされました。

NOTE: cache registered group DG_NADCLPDBSCCMC2A_VOTE_01 38/0x6090142B
NOTE: cache began mount (first) of group DG_NADCLPDBSCCMC2A_VOTE_01 38/0x6090142B
Sun Dec 11 12:04:53 2016
.... 
NOTE: Assigning number (38,0) to disk (/dev/emcpowere1)
NOTE: Assigning number (38,1) to disk (/dev/emcpowera1)
....====>> 当該ディスクグループ(diskgroup number 37 の代わりに 38)は、12.1 ホームからのスタック起動でマウントされます。
Sun Dec 11 12:05:11 2016
NOTE: cache mounting group 36/0xFAA0142A (DG_CCONASP2_UNDO_TEMP_01) succeeded
NOTE: cache ending mount (success) of group DG_CCONASP2_UNDO_TEMP_01 number=36 incarn=0xfaa0142a
Sun Dec 11 12:05:11 2016
ERROR: GMON found another heartbeat for grp 38 (DG_NADCLPDBSCCMC2A_VOTE_01)
Sun Dec 11 12:05:11 2016
ERROR: GMON could not check any PST heartbeat (grp 38)
Sun Dec 11 12:05:11 2016
NOTE: cache dismounting (clean) group 38/0x6090142B (DG_NADCLPDBSCCMC2A_VOTE_01)
NOTE: messaging CKPT to quiesce pins Unix process pid: 38023, image: oracle@nadclpdbsccmc2a.unix.medcity.net (TNS V1)
....
NOTE: Disk DG_NADCLPDBSCCMC2A_VOTE__0004 in mode 0x7f marked for de-assignment
ERROR: diskgroup DG_NADCLPDBSCCMC2A_VOTE_01 was not mounted


手動マウントは失敗します。

SQL> ALTER DISK GROUP DG_NADCLPDBSCCMC2A_VOTE_01 MOUNT;

ORA-15032: not all alterations performed
ORA-15017: diskgroup "DG_NADCLPDBSCCMC2A_VOTE_01" cannot be mounted
ORA-15003: diskgroup "DG_NADCLPDBSCCMC2A_VOTE_01" already mounted in another lock name space

 

原因

To view full details, sign in with your My Oracle Support account.

Don't have a My Oracle Support account? Click to get started!


本書の内容
現象
原因
解決策
参照情報

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.