Re: 11gR2 Rac install issue

  • From: "Martin Bach" <development@xxxxxxxxxxxxxxxxx>
  • To: "Martin Brown" <martinfbrown@xxxxxxxxxxx>,mzito@xxxxxxxxxxx,oracle-l@xxxxxxxxxxxxx
  • Date: Thu, 09 Sep 2010 09:40:43 +0100

Very interesting case. I absolutely love digging into RAC details, I am now 
very curious. Has the database been restored from a cluster with more than 7 
nodes by any chance? Are you mapping threads to instances as I described in the 
spfile? I will check the bug number as well.

Martin

Martin Bach

Oracle Certified Master 10g
http://martincarstenbach.wordpress.com
http://www.linkedin.com/in/martincarstenbach

----- Reply message -----
From: "Martin Brown" <martinfbrown@xxxxxxxxxxx>
Date: Wed, Sep 8, 2010 21:54
Subject: 11gR2 Rac install issue
To: <mzito@xxxxxxxxxxx>, <development@xxxxxxxxxxxxxxxxx>, 
<oracle-l@xxxxxxxxxxxxx>


Thanks a bunch, fellas. This is so bizarre that I should have looked here but 
didn't. 
Notice the private/public settings as well. I'm sure we'll be addressing this. 
 
SQL> column instance format a20
SQL> /
   THREAD# STATUS ENABLED  INSTANCE
---------- ------ -------- --------------------
         1 OPEN   PUBLIC   ormsperf1
         2 OPEN   PRIVATE  ormsperf2
         3 OPEN   PRIVATE  ormsperf3
         4 OPEN   PRIVATE  ormsperf4
         5 OPEN   PRIVATE  ormsperf5
         6 OPEN   PRIVATE  ormsperf6
         7 OPEN   PRIVATE  ormsperf7
         8 CLOSED PUBLIC   UNNAMED_INSTANCE_8
         9 CLOSED PUBLIC   UNNAMED_INSTANCE_9
        10 CLOSED PUBLIC   UNNAMED_INSTANCE_10
        11 CLOSED PUBLIC   UNNAMED_INSTANCE_11
   THREAD# STATUS ENABLED  INSTANCE
---------- ------ -------- --------------------
        12 CLOSED PUBLIC   UNNAMED_INSTANCE_12
        13 CLOSED PUBLIC   UNNAMED_INSTANCE_13
13 rows selected.
SQL> 
 
> Subject: RE: 11gR2 Rac install issue
> Date: Wed, 8 Sep 2010 15:47:36 -0400
> From: mzito@xxxxxxxxxxx
> To: development@xxxxxxxxxxxxxxxxx; oracle-l@xxxxxxxxxxxxx
> 
> One of our customers noted this the other day - we worked with Oracle,
> turns out it's bug 9383000 - assuming you're hitting the same issue our
> customer is.
> 
> There's a patch for it, and it's fixed in 11.2.0.2.
> 
> Matt
> 
> -----Original Message-----
> From: oracle-l-bounce@xxxxxxxxxxxxx
> [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Martin Bach
> Sent: Wednesday, September 08, 2010 3:44 PM
> To: oracle-l@xxxxxxxxxxxxx
> Subject: Re: 11gR2 Rac install issue
> 
> Hi Martin,
> 
> that's interesting. Normally you have a 1:1 mapping between threads and 
> instances. The mapping of thread to instance usually happens in the 
> initialisation file (spfile), where you find something like
> 
> instance_n.thread=n i.e.
> prod1.thread=1
> prod2.thread=2
> ...
> prod7.thread=7
> 
> In a seven node cluster I wouldn't expect more than 7 public threads, 
> unless there were even more nodes in the cluster which haven't been 
> deconfigured properly. What's the output of this query?
> 
> select thread#,status,enabled,instance from v$thread;
> 
> Best regards,
> 
> Martin
> 
> On 08/09/2010 18:48, Martin Brown wrote:
> > Hello,
> > I am looking around in a new install and I've noticed that the install
> > seems to have created too many threads and corresponding redo logs. We
> > use a 7-node cluster and I see 13 public threads. Is this normal? Any
> > ideas where to look to see what caused it if it isn't?
> >
> > TIA
> 
> 
> -- 
> Martin Bach
> OCM 10g
> http://martincarstenbach.wordpress.com
> http://www.linkedin.com/in/martincarstenbach
> --
> //www.freelists.org/webpage/oracle-l
> 
> 
> --
> //www.freelists.org/webpage/oracle-l
> 
> 
                                          

Other related posts: