cancel
Showing results for 
Search instead for 
Did you mean: 

Oracle Fail Safe cannot start DB == ORA-12631: Username retrieval failed

Former Member
0 Kudos

I cannot longer startup the database (Oracle 10.2.0.4) resource via MSCS (OFS 3.4.2).

It used to work fine, until last week.

We had some issues with the cluster, now it is OK.

There were no changes in the database,

Here are the error messages:

//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////

<DB_RES> Oracle Fail Safe resource ORCL.WORLD is starting

<DB_RES> ORCL.WORLD starting

<DB_RES> ORCL.WORLD connect as user "SYS", password is supplied, SYSDBA, PRELIM_AUTH

<DB_RES> using bequeath

<DB_RES> ORCL.WORLD OCIServerAttach attach string "(DESCRIPTION=(ADDRESS=(PROTOCOL=BEQ)(PROGRAM=oracle)(ARGV0=oracleORCL)(ARGS='(DESCRIPTION=(LOCAL=YES)

(ADDRESS=(PROTOCOL=beq)))'))(CONNECT_DATA=(SID=ORCL)

(CID=(PROGRAM=C:\WINDOWS\cluster\resrcmon.exe)(HOST=host1)(USER=DOMAIN\winservice))))"

<DB_RES> ORCL.WORLD ***** OCI routine OCIServerAttach Bequeath returned error -1 - OCI_ERROR

<COMMON> ORA-12631: Username retrieval failed

<DB_RES> ORCL.WORLD disconnect

<DB_RES> Event start

Oracle Fail Safe resource ORCL.WORLD failed to start.

ORA-12631: Username retrieval failed

//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////

Best regards,

Tom

Accepted Solutions (0)

Answers (2)

Answers (2)

Former Member
0 Kudos

Are you maintaining any password file on nodes?

If yes, just copy from source to target and try (from node 1 to node 2).

Regards,

Nick Loy

Former Member
0 Kudos

Yes, the password file is kept locally, on each node.

It is the same on both nodes, and it hasn't changed since the creation of the database/instances.

The issue now is that the cluster resource (database resource) is NOT coming up on node #1.

Node #2 is down, I am not trying to failover and/or move the cluster group.

Also, before rebooting node #1, and AFTER shutting down node #2, I was ABLE to start cluster resource (database resource) on node #1. Then, after rebooting node #1, the cluster resource (database resource) is NOT coming up!!

Former Member
0 Kudos

Here is the error, in more details:

///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////

Mon Sep 19 22:49:59 [srvr1] <DB_RES> ORCL.WORLD starting

Mon Sep 19 22:49:59 [srvr1] <DB_RES> ORCL.WORLD connect as user "SYS", password is supplied, SYSDBA, PRELIM_AUTH

Mon Sep 19 22:49:59 [srvr1] <DB_RES> using bequeath

Mon Sep 19 22:49:59 [srvr1] <DB_RES> ORCL.WORLD OCIServerAttach attach string "(DESCRIPTION=(ADDRESS=(PROTOCOL=BEQ)(PROGRAM=oracle)(ARGV0=oracleORCL)(ARGS='(DESCRIPTION=(LOCAL=YES)(ADDRESS=(PROTOCOL=beq)))'))(CONNECT_DATA=(SID=ORCL)(CID=(PROGRAM=C:\WINDOWS\cluster\resrcmon.exe)(HOST=srvr1)(USER=DOMAIN\userservice))))"

Mon Sep 19 22:49:59 [srvr1] <DB_RES> ORCL.WORLD ***** OCI routine OCIServerAttach Bequeath returned error -1 - OCI_ERROR

Mon Sep 19 22:49:59 [srvr1] <COMMON> ORA-12631: Username retrieval failed

Mon Sep 19 22:49:59 [srvr1] <DB_RES> ORCL.WORLD disconnect

Mon Sep 19 22:49:59 [srvr1] <DB_RES> Event start

Oracle Fail Safe resource ORCL.WORLD failed to start.

ORA-12631: Username retrieval failed

///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////

Former Member
0 Kudos

It seems authentication error. As a try reboot the server and let us know.

Regards,

Nick Loy

Former Member
0 Kudos

Hi Nick, thank you for your reply!

Since this is a PROD env. I have to wait until the weekend.

Best regards,

Tom

Former Member
0 Kudos

Just an FYI.

The 2nd node was rebooted, don't ask me why the 1st wasn't !!!

Now, I am able to bring up the database 'resource', via Windows Cluster Administrator (ver. 5.2), I wasn't able before to do that.

BUT, I cannot move the cluster group (the one that has the database 'resource') from node #1 to node #2.

I keep getting the same error (as before):

<<< Oracle Fail Safe resource ORCL.WORLD failed to start. >>>

<<< ORA-12631: Username retrieval failed >>>

.

Also, I keep seeing the following errors in Window's Event Viewer Logs:

SAP-Basis System: Operating system call recv failed (error no. 10054)

SAP-Basis System: Operating system call connect failed (error no. 10061)

Best regards,

Tom

Former Member
0 Kudos

I just reboot BOTH nodes.

Actually, I left the 2nd node down (it is still down).

Node #1 is up, but the database cluster resource is NOT coming up!!!!!

I keep getting the same error (as before):

<<< Oracle Fail Safe resource ORCL.WORLD failed to start. >>>

<<< ORA-12631: Username retrieval failed >>>

.

Also, I keep seeing the following errors in Window's Event Viewer Logs:

SAP-Basis System: Operating system call recv failed (error no. 10054)

SAP-Basis System: Operating system call connect failed (error no. 10061)

I am not sure what else has left to do ?

Best regards,

Tom