OpenVMS: Second RAC Instance Startup Hangs and Never Opens, Ora-01033 When Connecting (Doc ID 1256414.1)

Last updated on SEPTEMBER 21, 2016

Applies to:

Oracle Database - Enterprise Edition - Version 10.2.0.4 and later
HP OpenVMS Itanium
HP OpenVMS Alpha
***Checked for relevance on 08-April-2013***

Symptoms

Newly created RAC database does not open on the second node.

The database is mounted and opened by the first instance.
The database is mounted but not opened by the second instance. The second instance hangs after ALTER DATABASE OPEN.


RAC related communication like lock requests, reconfiguration, is working correctly.
No errors are seen in the alert.log or tracefiles.

The system state dump shows one session with  "waiting for 'gc cr request' "
Example from the system state dump:

O/S info: user: ORACLE10, term: , ospid: 2060045F, machine:
program: DSA1:[ORACLE10.DATABASE.bin]sqlplus.exe;1
application name: DSA3:[ORACLE10.DATABASE.bin]sqlplus.exe;1, hash value=912116949
waiting for 'gc cr request' blocking sess=0x0 seq=2112 wait_time=0 seconds since wait started=0
file#=1, block#=9d, class#=4
Dumping Session Wait History
for 'cr request retry' count=1 wait_time=0
file#=1, block#=9d, =0
for 'gc cr block lost' count=1 wait_time=875000
file#=1, =9d, =4

SQLPlus connection to the hanging instance returns error ORA-01033.
ORA-01033 Oracle startup or shutdown in progress.

When the first instance shuts down, the hanging instance opens.

Changes

New RAC database with a db_block_size larger than 8K, or with tablespaces
built with large block sizes using (sub)buffer caches like db_16k_cache_size, db_32_cache_size.

Cause

Sign In with your My Oracle Support account

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

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms