From: Andreas Piesk on
Hello list,

is it possible to create a 11gr1 rac database on grid 11gr2? i thought
it is possible but i get errors while trying it. dbca fails while
trying to start the instance:

ORA-29702: error occurred in Cluster Group Service operation

the alert.log:

Mon Nov 30 18:13:03 2009
Starting ORACLE instance (normal)
LICENSE_MAX_SESSION = 0
LICENSE_SESSIONS_WARNING = 0
Errors in file /opt/oracle/base/diag/rdbms/rac11r1/RAC11R1/trace/
RAC11R1_ora_31146.trc:
ORA-27504: IPC error creating OSD context
ORA-27300: OS system dependent operation:skgxnqtsz failed with status:
0
ORA-27301: OS failure message: Error 0
ORA-27302: failure occurred at: SKGXN not av
clsssinit ret = 21
interconnect information is not available from OCR
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.
Shared memory segment for instance monitoring created
Picked latch-free SCN scheme 2
Using LOG_ARCHIVE_DEST_1 parameter default value as /opt/oracle/base/
product/11.1.0/db/dbs/arch
Using LOG_ARCHIVE_DEST_10 parameter default value as
USE_DB_RECOVERY_FILE_DEST
Autotune of undo retention is turned on.
IMODE=BR
ILAT =18
LICENSE_MAX_USERS = 0
SYS auditing is disabled
Starting up ORACLE RDBMS Version: 11.1.0.6.0.
Using parameter settings in client-side pfile /opt/oracle/base/admin/
RAC11R1/pfile/init.ora on machine rac11r2-1
System parameters with non-default values:
processes = 150
nls_territory = "GERMANY"
memory_target = 300M
db_block_size = 8192
compatible = "11.1.0.0.0"
db_create_file_dest = "+DATA"
db_recovery_file_dest = "+RECO"
db_recovery_file_dest_size= 2G
undo_tablespace = "UNDOTBS1"
remote_login_passwordfile= "EXCLUSIVE"
db_domain = "test"
dispatchers = "(PROTOCOL=TCP) (SERVICE=RAC11R1XDB)"
audit_file_dest = "/opt/oracle/base/admin/RAC11R1/adump"
audit_trail = "DB"
db_name = "RAC11R1"
open_cursors = 300
diagnostic_dest = "/opt/oracle/base"
Cluster communication is configured to use the following interface(s)
for this instance
10.10.0.101
cluster interconnect IPC version:Oracle UDP/IP (generic)
IPC Vendor 1 proto 2
Mon Nov 30 18:13:05 2009
PMON started with pid=2, OS id=31155
Mon Nov 30 18:13:05 2009
VKTM started with pid=3, OS id=31159 at elevated priority
VKTM running at (20)ms precision
Mon Nov 30 18:13:05 2009
DIAG started with pid=4, OS id=31165
Mon Nov 30 18:13:05 2009
DBRM started with pid=5, OS id=31169
Mon Nov 30 18:13:05 2009
PING started with pid=6, OS id=31173
Mon Nov 30 18:13:05 2009
PSP0 started with pid=7, OS id=31177
Mon Nov 30 18:13:05 2009
ACMS started with pid=8, OS id=31181
Mon Nov 30 18:13:05 2009
DSKM started with pid=9, OS id=31185
Mon Nov 30 18:13:05 2009
DIA0 started with pid=10, OS id=31189
Mon Nov 30 18:13:05 2009
LMON started with pid=11, OS id=31193
Mon Nov 30 18:13:05 2009
LMD0 started with pid=9, OS id=31197
Mon Nov 30 18:13:05 2009
RMS0 started with pid=12, OS id=31201
Mon Nov 30 18:13:05 2009
MMAN started with pid=13, OS id=31205
Mon Nov 30 18:13:05 2009
DBW0 started with pid=14, OS id=31209
Mon Nov 30 18:13:05 2009
LGWR started with pid=15, OS id=31213
Mon Nov 30 18:13:05 2009
CKPT started with pid=16, OS id=31217
Mon Nov 30 18:13:05 2009
SMON started with pid=17, OS id=31221
Mon Nov 30 18:13:05 2009
RECO started with pid=18, OS id=31225
Mon Nov 30 18:13:05 2009
MMON started with pid=19, OS id=31229
starting up 1 dispatcher(s) for network address '(ADDRESS=(PARTIAL=YES)
(PROTOCOL=TCP))'...
Mon Nov 30 18:13:05 2009
MMNL started with pid=20, OS id=31233
starting up 1 shared server(s) ...
USER (ospid: 31146): terminating the instance due to error 29702
Instance terminated by USER, pid = 31146

11gr1 cannot get the cluster ip and therefore cannot communicate with
the clusterware a.k.a. grid.
on the same system i can deploy and start 11gr1 RAC or 11gr2 RAC with
any problems, so the requirements for 11gr1 and 11gr2 are met.

am i completely wrong and it's not possible?

regards,
-ap
From: Andreas Piesk on
On 30 Nov., 19:14, Andreas Piesk <a.pi...(a)gmx.net> wrote:
>
> am i completely wrong and it's not possible?
>

it's possible, of course. i forgot to pin ALL nodes. after fixing my
error it works perfectly. sorry for the noise.

regards,
-ap