Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Startup RAC database.

Re: Startup RAC database.

From: Steve <sroesch3_at_attbi.com>
Date: 3 Jun 2002 22:14:56 -0700
Message-ID: <efc3f20b.0206032114.4d434d23@posting.google.com>


fluo_at_stirlingdouglas.com (frank12345) wrote in message news:<5025b594.0206030628.1c89939b_at_posting.google.com>...
> sroesch3_at_attbi.com (Steve) wrote in message news:<efc3f20b.0206011709.31a162e3_at_posting.google.com>...
> > "news1.on.sympatico.ca" <luosh_at_hotmail.com> wrote in message news:<8f5K8.18861$Sn2.2524971_at_news20.bellglobal.com>...
> > > Hi Gurus,
> > >
> > > Here is my problem, can anyone give some hints?
> > > I installed 9i in three AIX 64 bits unix box. I also create my database use
> > > dbca tools with 3 instances each per node. The oracle file system are
> > > created on raw device. It went through until postinstallation step, the last
> > > step of installcation.
> > >
> > > Postinstallation already created other 2 groups redo log files for other two
> > > instances. But it is hanging due to only one instance startup. the other
> > > two are hang. The hanging place is SMON checking dead thread from alert log
> > > file.
> > >
> > > I tried to shut down all instances and restart manually by using sqlplus, I
> > > got same result. The funny thing is I can ONLY start one instance from any
> > > of three node, but I can not start them all.
> > >
> > > Is I missed something?
> > >
> > > I really need get this done for my benchmarking my applications. I
> > > appreciated any suggestion .
> > >
> > > Thank you in advance.
> > >
> > > Frank Luo
> >
> > Frank two things:
> >
> > - Check if cluster_database is set to true.
> > - Check if the other redo logs thread are already enabled.
> >
> > If this doesn help post your alert.log file
> >
> > Regards
> > Steve
>
> Hi Steve,
>
> Thank you for you suggestion. I did these two thing before. Anyway I
> tried again, It still does not work.
>
> Here is alert file from hanging instance (second one).
>
> Starting up ORACLE RDBMS Version: 9.0.1.3.0.
> System parameters with non-default values:
> processes = 150
> timed_statistics = TRUE
> shared_pool_size = 117440512
> large_pool_size = 1048576
> java_pool_size = 117440512
> resource_manager_plan = SYSTEM_PLAN
> control_files = /oracle/admin/oradata/marvin_ctrl1,
> /oracle/admin/oradata/pepe_ctrl1
> db_block_size = 32768
> db_cache_size = 67108864
> compatible = 9.0.0
> cluster_database = TRUE
> cluster_database_instances= 3
> thread = 2
> fast_start_mttr_target = 300
> instance_number = 2
> undo_management = AUTO
> undo_tablespace = UNDOTBS2
> remote_login_passwordfile= EXCLUSIVE
> db_domain =
> instance_name = sdgdb2
> dispatchers = (PROTOCOL=TCP)(SER=MODOSE),
> (PROTOCOL=TCP)(PRE=oracle.aurora.server.GiopServer),
> (PROTOCOL=TCP)(PRE=oracle.aurora.server.SGiopServer)
> remote_listener = LISTENERS_SDGNCR
> background_dump_dest = /oracle/admin/sdgdb/bdump
> user_dump_dest = /oracle/admin/sdgdb/udump
> core_dump_dest = /oracle/admin/sdgdb/cdump
> sort_area_size = 524288
> db_name = sdgdb
> open_cursors = 300
> Mon Jun 3 08:37:55 2002
> cluster interconnect IPC version:Oracle UDP/IP
> IPC Vendor 1 proto 2 Version 1.0
> PMON started with pid=2
> Syntax error in listener string LISTENERS_SDGNCR
> DIAG started with pid=3
> LMON started with pid=4
> LMD0 started with pid=5
> LMS0 started with pid=6
> LMS1 started with pid=7
> DBW0 started with pid=8
> LGWR started with pid=9
> CKPT started with pid=10
> SMON started with pid=11
> RECO started with pid=12
> Mon Jun 3 08:37:59 2002
> starting up 1 shared server(s) ...
> starting up 1 dispatcher(s) for network address
> '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...
> starting up 1 dispatcher(s) for network address
> '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...
> starting up 1 dispatcher(s) for network address
> '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...
> Mon Jun 3 08:38:02 2002
> ALTER DATABASE MOUNT
> Mon Jun 3 08:38:02 2002
> lmon registered with NM - instance id 2 (internal mem no 1)
> Mon Jun 3 08:38:03 2002
> Reconfiguration started
> List of nodes: 0,1,
> Global Resource Directory frozen
> Communication channels reestablished
> Server queues filtered
> Master broadcasted resource hash value bitmaps
> Non-local Process blocks cleaned out
> Resources and enqueues cleaned out
> Resources remastered 0
> 0 GCS shadows traversed, 0 cancelled, 0 closed
> 0 GCS resources traversed, 0 cancelled
> set master node info
> 0 GCS shadows traversed, 0 replayed, 0 unopened
> Submitted all remote-enqueue requests
> Update rdomain variables
> 0 write requests issued in 415 GCS resources
> 0 PIs marked suspect, 0 flush PI msgs
> Dwn-cvts replayed, VALBLKs dubious
> All grantable enqueues granted
> Mon Jun 3 08:38:04 2002
> Reconfiguration complete
> LCK0 started with pid=18
> Mon Jun 3 08:38:10 2002
> Successful mount of redo thread 2, with mount id 3913904395.
> Mon Jun 3 08:38:10 2002
> Database mounted in Shared Mode (CLUSTER_DATABASE=TRUE).
> Completed: ALTER DATABASE MOUNT
> Mon Jun 3 08:38:10 2002
> ALTER DATABASE OPEN
> Picked Lamport scheme to generate SCNs
> Mon Jun 3 08:38:11 2002
> Thread 2 advanced to log sequence 6
> Thread 2 opened at log sequence 6
> Current log# 4 seq# 6 mem# 0: /oracle/admin/oradata/pepe_redo2
> Successful open of redo thread 2.
> Mon Jun 3 08:38:11 2002
> SMON: enabling cache recovery
>
>
> Then It is waiting here forever.
>
> Following is part of init.ora for cluster setting
>
> ###########################################
> # Cluster Database
> ###########################################
> cluster_database=true
> cluster_database_instances=3
> ....
> ###########################################
> # System Managed Undo and Rollback Segments
> ###########################################
> undo_management=AUTO
> undo_tablespace=UNDOTBS2
> sdgdb1.instance_name=sdgdb1
> sdgdb2.instance_name=sdgdb2
> sdgdb3.instance_name=sdgdb3
> sdgdb2.instance_number=2
> sdgdb1.instance_number=1
> sdgdb3.instance_number=3
> sdgdb3.thread=3
> sdgdb2.thread=2
> sdgdb1.thread=1
> sdgdb2.undo_tablespace=UNDOTBS2
> sdgdb1.undo_tablespace=UNDOTBS
> sdgdb3.undo_tablespace=UNDOTBS3

Hi Fred,

several things:
- your listener entry is wrong, comment it out (remote_listeners) - disable the settings for mts. The current settings doesn't make a lot of sense (dispatchers)
- if this doesn't help set event 10500 and wait for 5 minutes. This will create a trace file with a system state dump. Feel free to send it to me.

Regards
Steve Received on Tue Jun 04 2002 - 00:14:56 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US