RE: ROW CACHE HIGH - Priority 1

2003-09-26 Thread Muqthar Ahmed
Hi,

There are two BUGS 2605240 and 2683906 which got fixed in 9.2.0.4.0.  I will be 
applying the patch to fix the problem.

Thanks for your input.

Muqthar

-Original Message-
Sent: Friday, September 26, 2003 9:55 AM
To: Multiple recipients of list ORACLE-L


SQL parsing would also be reflected in rowcache because, to parse
successfully, you
need dictionary and parse locks. 

--
Mladen Gogala
Oracle DBA 



> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On 
> Behalf Of Sam Roberts
> Sent: Friday, September 26, 2003 2:20 AM
> To: Multiple recipients of list ORACLE-L
> Subject: Re: ROW CACHE HIGH - Priority 1
> 
> 
> sql parsing would be reflected in Library cache not rowcache
> 
> Basic stuff
> 
> Sam
> - Original Message -
> To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
> Sent: Wednesday, September 24, 2003 1:14 PM
> 
> 
> > General answer: upgrade to 9.2.0.4 and hope that the bug has been 
> > fixed. Row cache locks are data dictionary locks. You can see the 
> > contents of row cache by inspecting v$rowcache.  You may need to 
> > increase shared pool. Last but not least, how fast is your private 
> > network connection between
> the
> > two nodes? 100mbit/sec is not nearly fast enough. You need 
> at least a 
> > gigabit switch. Also check your SQL for hard parsing (see the 
> > executions and invalidations in
> > v$sqlarea), ad-hoc DDL and that kind of stuff.
> > And no, it's not my priority 1.
> >
> >
> > --
> > Mladen Gogala
> > Oracle DBA
> >
> >
> >
> > > -Original Message-
> > > From: [EMAIL PROTECTED] 
> [mailto:[EMAIL PROTECTED] On Behalf 
> > > Of Muqthar Ahmed
> > > Sent: Wednesday, September 24, 2003 3:50 PM
> > > To: Multiple recipients of list ORACLE-L
> > > Subject: ROW CACHE HIGH - Priority 1
> > >
> > >
> > > Hi,
> > >
> > > I have Oracle 9.2.0.2.0 RAC (two nodes) on IBM AIX. 
> Currently I am 
> > > seeing very high number for ROW CACHE LOCK in statspack.
> > >
> > > Top 5 Timed Events
> > > ~~
> > >  % Total
> > > Event  Waits
> > > Time (s)   Ela Time
> > >   
> > > --- 
> > > row cache lock11,310  5,441
> > >  86.97
> > > CPU time
> > > 522   8.34
> > > global cache cr request32,513  71
> > >1.14
> > > global cache null to x   21,507  57
> > >  91
> > > log file sync22,689  49
> > >  .78
> > >
> > > -
> > >
> > >  Get
> > > Spin &
> > > Latch Name   Requests   Misses
> > >   SleepsSleeps 1->4
> > > --   --
> > > ---  ---   
> > > library cache 7,094,208   31,499
> > >  1,480  30031/1456/12/0/0
> > > shared pool  2,385,408   6,739
> > >   5276212/527/0/0/0
> > > ges enqueue table freelist  1,492,275  1,903
> > > 1241780/122/1/0 /0
> > > library cache pin   3,201,008   1,437
> > > 1301307/130/0/0/0
> > > row cache objects1,400,498   1,020
> > >   56  964/56/0/0/0
> > > row cache enqueue latch  1,292,843 715
> > > 19  696/19/0/0/0
> > >
> > > It is holding row cache lock (v$session_wait), other 
> sessions are in 
> > > the queuedue to this,  number of concurrent sessions will 
> > > increase from 100 to 350 sessions on each node.  In less than one 
> > > minute everything will be cleared ( OLTP database)
> > >
> > > I have already opened a TAR with Priority 1.  Do you have any 
> > > suggestions.
> > >
> > > Thanks
> > > Muqthar Ahmed
> > > DBA
> > >
> > > --
> > > Please see the official ORACLE-L FAQ: http://www.orafaq.net
> > > --
> > > Author: Muqthar Ahmed
> > >   INET: [EMAIL PROTECTED]
> 

RE: ROW CACHE HIGH - Priority 1

2003-09-26 Thread Mladen Gogala
SQL parsing would also be reflected in rowcache because, to parse
successfully, you
need dictionary and parse locks. 

--
Mladen Gogala
Oracle DBA 



> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On 
> Behalf Of Sam Roberts
> Sent: Friday, September 26, 2003 2:20 AM
> To: Multiple recipients of list ORACLE-L
> Subject: Re: ROW CACHE HIGH - Priority 1
> 
> 
> sql parsing would be reflected in Library cache not rowcache
> 
> Basic stuff
> 
> Sam
> - Original Message -
> To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
> Sent: Wednesday, September 24, 2003 1:14 PM
> 
> 
> > General answer: upgrade to 9.2.0.4 and hope that the bug has been 
> > fixed. Row cache locks are data dictionary locks. You can see the 
> > contents of row cache by inspecting v$rowcache.  You may need to 
> > increase shared pool. Last but not least, how fast is your private 
> > network connection between
> the
> > two nodes? 100mbit/sec is not nearly fast enough. You need 
> at least a 
> > gigabit switch. Also check your SQL for hard parsing (see the 
> > executions and invalidations in
> > v$sqlarea), ad-hoc DDL and that kind of stuff.
> > And no, it's not my priority 1.
> >
> >
> > --
> > Mladen Gogala
> > Oracle DBA
> >
> >
> >
> > > -Original Message-
> > > From: [EMAIL PROTECTED] 
> [mailto:[EMAIL PROTECTED] On Behalf 
> > > Of Muqthar Ahmed
> > > Sent: Wednesday, September 24, 2003 3:50 PM
> > > To: Multiple recipients of list ORACLE-L
> > > Subject: ROW CACHE HIGH - Priority 1
> > >
> > >
> > > Hi,
> > >
> > > I have Oracle 9.2.0.2.0 RAC (two nodes) on IBM AIX. 
> Currently I am 
> > > seeing very high number for ROW CACHE LOCK in statspack.
> > >
> > > Top 5 Timed Events
> > > ~~
> > >  % Total
> > > Event  Waits
> > > Time (s)   Ela Time
> > >   
> > > --- 
> > > row cache lock11,310  5,441
> > >  86.97
> > > CPU time
> > > 522   8.34
> > > global cache cr request32,513  71
> > >1.14
> > > global cache null to x   21,507  57
> > >  91
> > > log file sync22,689  49
> > >  .78
> > >
> > > -
> > >
> > >  Get
> > > Spin &
> > > Latch Name   Requests   Misses
> > >   SleepsSleeps 1->4
> > > --   --
> > > ---  ---   
> > > library cache 7,094,208   31,499
> > >  1,480  30031/1456/12/0/0
> > > shared pool  2,385,408   6,739
> > >   5276212/527/0/0/0
> > > ges enqueue table freelist  1,492,275  1,903
> > > 1241780/122/1/0 /0
> > > library cache pin   3,201,008   1,437
> > > 1301307/130/0/0/0
> > > row cache objects1,400,498   1,020
> > >   56  964/56/0/0/0
> > > row cache enqueue latch  1,292,843 715
> > > 19  696/19/0/0/0
> > >
> > > It is holding row cache lock (v$session_wait), other 
> sessions are in 
> > > the queuedue to this,  number of concurrent sessions will 
> > > increase from 100 to 350 sessions on each node.  In less than one 
> > > minute everything will be cleared ( OLTP database)
> > >
> > > I have already opened a TAR with Priority 1.  Do you have any 
> > > suggestions.
> > >
> > > Thanks
> > > Muqthar Ahmed
> > > DBA
> > >
> > > --
> > > Please see the official ORACLE-L FAQ: http://www.orafaq.net
> > > --
> > > Author: Muqthar Ahmed
> > >   INET: [EMAIL PROTECTED]
> > >
> > > Fat City Network Services-- 858-538-5051 
> http://www.fatcity.com
> > > San Diego, California-- Mailing list and web 
> hosting services
> > > 
> --

RE: ROW CACHE HIGH - Priority 1

2003-09-26 Thread Muqthar Ahmed
Hi,

Thanks for your suggestions.  Usually Friday we have low traffic on our website.  I 
will run the database with one node to check the row cache problem.

Muqthar

-Original Message-
Sent: Friday, September 26, 2003 7:15 AM
To: Multiple recipients of list ORACLE-L


> sql parsing would be reflected in Library cache not rowcache
>
> Basic stuff

Not that basic. Rowcache will get hit during any hard parse, since you need
to get information about objects from DD. And if objects information isn't
cached yet, it has to be loaded into row cache.

Btw, row cache locks might come from LMT extent (de)allocation. Now, if both
RAC instances start allocating or deallocating extents in a datafile, they
need to access the same bitmap blocks, which results in GC traffic or pings,
maybe that's the problem?

Tanel.

>
> Sam
> - Original Message -
> To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
> Sent: Wednesday, September 24, 2003 1:14 PM
>
>
> > General answer: upgrade to 9.2.0.4 and hope that the bug has been fixed.
> > Row cache locks are data dictionary locks. You can see the contents of
row
> > cache
> > by inspecting v$rowcache.  You may need to increase shared pool.
> > Last but not least, how fast is your private network connection between
> the
> > two nodes? 100mbit/sec is not nearly fast enough. You need at least a
> > gigabit
> > switch. Also check your SQL for hard parsing (see the executions and
> > invalidations in
> > v$sqlarea), ad-hoc DDL and that kind of stuff.
> > And no, it's not my priority 1.
> >
> >
> > --
> > Mladen Gogala
> > Oracle DBA
> >
> >
> >
> > > -Original Message-
> > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
> > > Behalf Of Muqthar Ahmed
> > > Sent: Wednesday, September 24, 2003 3:50 PM
> > > To: Multiple recipients of list ORACLE-L
> > > Subject: ROW CACHE HIGH - Priority 1
> > >
> > >
> > > Hi,
> > >
> > > I have Oracle 9.2.0.2.0 RAC (two nodes) on IBM AIX.
> > > Currently I am seeing very high number for ROW CACHE LOCK in
> > > statspack.
> > >
> > > Top 5 Timed Events
> > > ~~
> > >  % Total
> > > Event  Waits
> > > Time (s)   Ela Time
> > >   
> > > --- 
> > > row cache lock11,310  5,441
> > >  86.97
> > > CPU time
> > > 522   8.34
> > > global cache cr request32,513  71
> > >1.14
> > > global cache null to x   21,507  57
> > >  91
> > > log file sync22,689  49
> > >  .78
> > >
> > > -
> > >
> > >  Get
> > > Spin &
> > > Latch Name   Requests   Misses
> > >   SleepsSleeps 1->4
> > > --   --
> > > ---  ---   
> > > library cache 7,094,208   31,499
> > >  1,480  30031/1456/12/0/0
> > > shared pool  2,385,408   6,739
> > >   5276212/527/0/0/0
> > > ges enqueue table freelist  1,492,275  1,903
> > > 1241780/122/1/0 /0
> > > library cache pin   3,201,008   1,437
> > > 1301307/130/0/0/0
> > > row cache objects1,400,498   1,020
> > >   56  964/56/0/0/0
> > > row cache enqueue latch  1,292,843 715
> > > 19  696/19/0/0/0
> > >
> > > It is holding row cache lock (v$session_wait), other sessions
> > > are in the queuedue to this,  number of concurrent
> > > sessions will increase from 100 to 350 sessions on each node.
> > >  In less than one minute everything will be cleared ( OLTP database)
> > >
> > > I have already opened a TAR with Priority 1.  Do you have any
> > > suggestions.
> > >
> > > Thanks
> > > Muqthar Ahmed
> > > DBA
> > >
> > > --
> > > Please see the official ORACLE-L FAQ: http://www.orafaq.net
> > > --
> > > Author: Muqthar 

Re: ROW CACHE HIGH - Priority 1

2003-09-26 Thread Tanel Poder
> sql parsing would be reflected in Library cache not rowcache
>
> Basic stuff

Not that basic. Rowcache will get hit during any hard parse, since you need
to get information about objects from DD. And if objects information isn't
cached yet, it has to be loaded into row cache.

Btw, row cache locks might come from LMT extent (de)allocation. Now, if both
RAC instances start allocating or deallocating extents in a datafile, they
need to access the same bitmap blocks, which results in GC traffic or pings,
maybe that's the problem?

Tanel.

>
> Sam
> - Original Message -
> To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
> Sent: Wednesday, September 24, 2003 1:14 PM
>
>
> > General answer: upgrade to 9.2.0.4 and hope that the bug has been fixed.
> > Row cache locks are data dictionary locks. You can see the contents of
row
> > cache
> > by inspecting v$rowcache.  You may need to increase shared pool.
> > Last but not least, how fast is your private network connection between
> the
> > two nodes? 100mbit/sec is not nearly fast enough. You need at least a
> > gigabit
> > switch. Also check your SQL for hard parsing (see the executions and
> > invalidations in
> > v$sqlarea), ad-hoc DDL and that kind of stuff.
> > And no, it's not my priority 1.
> >
> >
> > --
> > Mladen Gogala
> > Oracle DBA
> >
> >
> >
> > > -Original Message-
> > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
> > > Behalf Of Muqthar Ahmed
> > > Sent: Wednesday, September 24, 2003 3:50 PM
> > > To: Multiple recipients of list ORACLE-L
> > > Subject: ROW CACHE HIGH - Priority 1
> > >
> > >
> > > Hi,
> > >
> > > I have Oracle 9.2.0.2.0 RAC (two nodes) on IBM AIX.
> > > Currently I am seeing very high number for ROW CACHE LOCK in
> > > statspack.
> > >
> > > Top 5 Timed Events
> > > ~~
> > >  % Total
> > > Event  Waits
> > > Time (s)   Ela Time
> > >   
> > > --- 
> > > row cache lock11,310  5,441
> > >  86.97
> > > CPU time
> > > 522   8.34
> > > global cache cr request32,513  71
> > >1.14
> > > global cache null to x   21,507  57
> > >  91
> > > log file sync22,689  49
> > >  .78
> > >
> > > -
> > >
> > >  Get
> > > Spin &
> > > Latch Name   Requests   Misses
> > >   SleepsSleeps 1->4
> > > --   --
> > > ---  ---   
> > > library cache 7,094,208   31,499
> > >  1,480  30031/1456/12/0/0
> > > shared pool  2,385,408   6,739
> > >   5276212/527/0/0/0
> > > ges enqueue table freelist  1,492,275  1,903
> > > 1241780/122/1/0 /0
> > > library cache pin   3,201,008   1,437
> > > 1301307/130/0/0/0
> > > row cache objects1,400,498   1,020
> > >   56  964/56/0/0/0
> > > row cache enqueue latch  1,292,843 715
> > > 19  696/19/0/0/0
> > >
> > > It is holding row cache lock (v$session_wait), other sessions
> > > are in the queuedue to this,  number of concurrent
> > > sessions will increase from 100 to 350 sessions on each node.
> > >  In less than one minute everything will be cleared ( OLTP database)
> > >
> > > I have already opened a TAR with Priority 1.  Do you have any
> > > suggestions.
> > >
> > > Thanks
> > > Muqthar Ahmed
> > > DBA
> > >
> > > --
> > > Please see the official ORACLE-L FAQ: http://www.orafaq.net
> > > --
> > > Author: Muqthar Ahmed
> > >   INET: [EMAIL PROTECTED]
> > >
> > > Fat City Network Services-- 858-538-5051 http://www.fatcity.com
> > > San Diego, California-- Mailing list and web hosting services
> > > --

Re: ROW CACHE HIGH - Priority 1

2003-09-25 Thread Sam Roberts
sql parsing would be reflected in Library cache not rowcache

Basic stuff

Sam
- Original Message -
To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
Sent: Wednesday, September 24, 2003 1:14 PM


> General answer: upgrade to 9.2.0.4 and hope that the bug has been fixed.
> Row cache locks are data dictionary locks. You can see the contents of row
> cache
> by inspecting v$rowcache.  You may need to increase shared pool.
> Last but not least, how fast is your private network connection between
the
> two nodes? 100mbit/sec is not nearly fast enough. You need at least a
> gigabit
> switch. Also check your SQL for hard parsing (see the executions and
> invalidations in
> v$sqlarea), ad-hoc DDL and that kind of stuff.
> And no, it's not my priority 1.
>
>
> --
> Mladen Gogala
> Oracle DBA
>
>
>
> > -Original Message-
> > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
> > Behalf Of Muqthar Ahmed
> > Sent: Wednesday, September 24, 2003 3:50 PM
> > To: Multiple recipients of list ORACLE-L
> > Subject: ROW CACHE HIGH - Priority 1
> >
> >
> > Hi,
> >
> > I have Oracle 9.2.0.2.0 RAC (two nodes) on IBM AIX.
> > Currently I am seeing very high number for ROW CACHE LOCK in
> > statspack.
> >
> > Top 5 Timed Events
> > ~~
> >  % Total
> > Event  Waits
> > Time (s)   Ela Time
> >   
> > --- 
> > row cache lock11,310  5,441
> >  86.97
> > CPU time
> > 522   8.34
> > global cache cr request32,513  71
> >1.14
> > global cache null to x   21,507  57
> >  91
> > log file sync22,689  49
> >  .78
> >
> > -
> >
> >  Get
> > Spin &
> > Latch Name   Requests   Misses
> >   SleepsSleeps 1->4
> > --   --
> > ---  ---   
> > library cache 7,094,208   31,499
> >  1,480  30031/1456/12/0/0
> > shared pool  2,385,408   6,739
> >   5276212/527/0/0/0
> > ges enqueue table freelist  1,492,275  1,903
> > 1241780/122/1/0 /0
> > library cache pin   3,201,008   1,437
> > 1301307/130/0/0/0
> > row cache objects1,400,498   1,020
> >   56  964/56/0/0/0
> > row cache enqueue latch  1,292,843 715
> > 19  696/19/0/0/0
> >
> > It is holding row cache lock (v$session_wait), other sessions
> > are in the queuedue to this,  number of concurrent
> > sessions will increase from 100 to 350 sessions on each node.
> >  In less than one minute everything will be cleared ( OLTP database)
> >
> > I have already opened a TAR with Priority 1.  Do you have any
> > suggestions.
> >
> > Thanks
> > Muqthar Ahmed
> > DBA
> >
> > --
> > Please see the official ORACLE-L FAQ: http://www.orafaq.net
> > --
> > Author: Muqthar Ahmed
> >   INET: [EMAIL PROTECTED]
> >
> > Fat City Network Services-- 858-538-5051 http://www.fatcity.com
> > San Diego, California-- Mailing list and web hosting services
> > -
> > To REMOVE yourself from this mailing list, send an E-Mail message
> > to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru')
> > and in the message BODY, include a line containing: UNSUB
> > ORACLE-L (or the name of mailing list you want to be removed
> > from).  You may also send the HELP command for other
> > information (like subscribing).
> >
>
>
>
>
> Note:
> This message is for the named person's use only.  It may contain
confidential, proprietary or legally privileged information.  No
confidentiality or privilege is waived or lost by any mistransmission.  If
you receive this message in error, please immediately delete it and all
copies of it from your system, destroy any hard copies of it and notify the
sender.  You must not, directly or indirectly, use, disclose, distribute,
print, or copy any part of this message if you are not the intended
recipien

RE: ROW CACHE HIGH - Priority 1

2003-09-25 Thread Mladen Gogala
Oh, it's not me, I was just replying to somebody. My company is not running
RAC.
I know what clustering means, I have 5+ years of experience with OPS on both
versions 7 and 8 plus I used to teach people VMS and how to tune it. Yeah, 
you're absolutely right clustering is flustering, don't do it unless  that
is
what you really, really want & need. Especially need.

--
Mladen Gogala
Oracle DBA 



> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On 
> Behalf Of Mogens Nørgaard
> Sent: Wednesday, September 24, 2003 7:35 PM
> To: Multiple recipients of list ORACLE-L
> Subject: Re: ROW CACHE HIGH - Priority 1
> 
> 
> I HAVE to ask this question: Is it possible for you to turn 
> off RAC and 
> thereby completely avoiding this issue?
> 
> Yeah, that's kind of funny. Except it isn't, really. Invalidations of 
> objects, drops, in general breaking of breakable parse locks 
> - they will 
> all need to be communicated to all the other instances. But 
> this kind of 
> activity is often forgotten in lieu of the "normal" buffer 
> cache activity.
> 
> Undskyld. But the best optimisation is not to do it :-).
> 
> Mogens
> 
> Mladen Gogala wrote:
> 
> >General answer: upgrade to 9.2.0.4 and hope that the bug has been 
> >fixed. Row cache locks are data dictionary locks. You can see the 
> >contents of row cache by inspecting v$rowcache.  You may need to 
> >increase shared pool. Last but not least, how fast is your private 
> >network connection between the two nodes? 100mbit/sec is not nearly 
> >fast enough. You need at least a gigabit
> >switch. Also check your SQL for hard parsing (see the executions and
> >invalidations in 
> >v$sqlarea), ad-hoc DDL and that kind of stuff.
> >And no, it's not my priority 1. 
> >
> >
> >--
> >Mladen Gogala
> >Oracle DBA
> >
> >
> >
> >  
> >
> >>-Original Message-
> >>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
> >>Behalf Of Muqthar Ahmed
> >>Sent: Wednesday, September 24, 2003 3:50 PM
> >>To: Multiple recipients of list ORACLE-L
> >>Subject: ROW CACHE HIGH - Priority 1
> >>
> >>
> >>Hi,
> >>
> >>I have Oracle 9.2.0.2.0 RAC (two nodes) on IBM AIX.
> >>Currently I am seeing very high number for ROW CACHE LOCK in 
> >>statspack.
> >>
> >>Top 5 Timed Events
> >>~~
> >> % Total
> >>Event  Waits
> >>Time (s)   Ela Time
> >>  
> >>--- 
> >>row cache lock11,310  5,441   
> >> 86.97
> >>CPU time   
> >>522   8.34
> >>global cache cr request32,513  71 
> >>   1.14
> >>global cache null to x   21,507  57   
> >> 91
> >>log file sync22,689  49   
> >> .78
> >>  
> >>-
> >>
> >> Get  
> >>Spin &
> >>Latch Name   Requests   Misses
> >>  SleepsSleeps 1->4
> >>--   --  
> >>---  ---   
> >>library cache 7,094,208   31,499  
> >> 1,480  30031/1456/12/0/0
> >>shared pool  2,385,408   6,739
> >>  5276212/527/0/0/0
> >>ges enqueue table freelist  1,492,275  1,903  
> >>1241780/122/1/0 /0
> >>library cache pin   3,201,008   1,437 
> >>1301307/130/0/0/0
> >>row cache objects1,400,498   1,020
> >>  56  964/56/0/0/0
> >>row cache enqueue latch  1,292,843 715   
> >>19  696/19/0/0/0
> >>
> >>It is holding row cache lock (v$session_wait), other sessions
> >>are in the queuedue to this,  number of concurrent 
> >>sessions will increase from 100 to 350 sessions on each node. 
> >> In less than

Re: ROW CACHE HIGH - Priority 1

2003-09-24 Thread Mogens Nørgaard
I HAVE to ask this question: Is it possible for you to turn off RAC and 
thereby completely avoiding this issue?

Yeah, that's kind of funny. Except it isn't, really. Invalidations of 
objects, drops, in general breaking of breakable parse locks - they will 
all need to be communicated to all the other instances. But this kind of 
activity is often forgotten in lieu of the "normal" buffer cache activity.

Undskyld. But the best optimisation is not to do it :-).

Mogens

Mladen Gogala wrote:

General answer: upgrade to 9.2.0.4 and hope that the bug has been fixed.
Row cache locks are data dictionary locks. You can see the contents of row
cache
by inspecting v$rowcache.  You may need to increase shared pool.
Last but not least, how fast is your private network connection between the
two nodes? 100mbit/sec is not nearly fast enough. You need at least a
gigabit 
switch. Also check your SQL for hard parsing (see the executions and
invalidations in 
v$sqlarea), ad-hoc DDL and that kind of stuff.
And no, it's not my priority 1. 

--
Mladen Gogala
Oracle DBA 



 

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On 
Behalf Of Muqthar Ahmed
Sent: Wednesday, September 24, 2003 3:50 PM
To: Multiple recipients of list ORACLE-L
Subject: ROW CACHE HIGH - Priority 1

Hi,

I have Oracle 9.2.0.2.0 RAC (two nodes) on IBM AIX.  
Currently I am seeing very high number for ROW CACHE LOCK in 
statspack.

Top 5 Timed Events
~~
% Total
Event  Waits
Time (s)   Ela Time
  
   --- 
row cache lock11,310  5,441   
86.97
CPU time   
522   8.34
global cache cr request32,513  71 
  1.14
global cache null to x   21,507  57   
91
log file sync22,689  49   
.78
 
-

Get  
   Spin &
Latch Name   Requests   Misses
 SleepsSleeps 1->4
--   --  
---  ---   
library cache 7,094,208   31,499  
1,480  30031/1456/12/0/0
shared pool  2,385,408   6,739
 5276212/527/0/0/0
ges enqueue table freelist  1,492,275  1,903  
1241780/122/1/0 /0
library cache pin   3,201,008   1,437 
   1301307/130/0/0/0
row cache objects1,400,498   1,020
 56  964/56/0/0/0
row cache enqueue latch  1,292,843 715   
19  696/19/0/0/0

It is holding row cache lock (v$session_wait), other sessions 
are in the queuedue to this,  number of concurrent 
sessions will increase from 100 to 350 sessions on each node. 
In less than one minute everything will be cleared ( OLTP database)

I have already opened a TAR with Priority 1.  Do you have any 
suggestions.

Thanks 
Muqthar Ahmed
DBA

--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: Muqthar Ahmed
 INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web hosting services
-
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') 
and in the message BODY, include a line containing: UNSUB 
ORACLE-L (or the name of mailing list you want to be removed 
from).  You may also send the HELP command for other 
information (like subscribing).

   





Note:
This message is for the named person's use only.  It may contain confidential, 
proprietary or legally privileged information.  No confidentiality or privilege is 
waived or lost by any mistransmission.  If you receive this message in error, please 
immediately delete it and all copies of it from your system, destroy any hard copies 
of it and notify the sender.  You must not, directly or indirectly, use, disclose, 
distribute, print, or copy any part of this message if you are not the intended 
recipient. Wang Trading LLC and any of its subsidiaries each reserve the right to 
monitor all e-mail communications through its networks.
Any views expressed in this message are those of the individual sender, except where 
the message states otherwise and the sender is authorized to state them to be the 
views of any such entity.
 

--
Please see the official ORACLE-L FAQ: 

RE: ROW CACHE HIGH - Priority 1

2003-09-24 Thread Mladen Gogala
General answer: upgrade to 9.2.0.4 and hope that the bug has been fixed.
Row cache locks are data dictionary locks. You can see the contents of row
cache
by inspecting v$rowcache.  You may need to increase shared pool.
Last but not least, how fast is your private network connection between the
two nodes? 100mbit/sec is not nearly fast enough. You need at least a
gigabit 
switch. Also check your SQL for hard parsing (see the executions and
invalidations in 
v$sqlarea), ad-hoc DDL and that kind of stuff.
And no, it's not my priority 1. 


--
Mladen Gogala
Oracle DBA 



> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On 
> Behalf Of Muqthar Ahmed
> Sent: Wednesday, September 24, 2003 3:50 PM
> To: Multiple recipients of list ORACLE-L
> Subject: ROW CACHE HIGH - Priority 1
> 
> 
> Hi,
> 
> I have Oracle 9.2.0.2.0 RAC (two nodes) on IBM AIX.  
> Currently I am seeing very high number for ROW CACHE LOCK in 
> statspack.
> 
> Top 5 Timed Events
> ~~
>  % Total
> Event  Waits
> Time (s)   Ela Time
>   
> --- 
> row cache lock11,310  5,441   
>  86.97
> CPU time   
> 522   8.34
> global cache cr request32,513  71 
>1.14
> global cache null to x   21,507  57   
>  91
> log file sync22,689  49   
>  .78
>   
> -
> 
>  Get  
> Spin &
> Latch Name   Requests   Misses
>   SleepsSleeps 1->4
> --   --  
> ---  ---   
> library cache 7,094,208   31,499  
>  1,480  30031/1456/12/0/0
> shared pool  2,385,408   6,739
>   5276212/527/0/0/0
> ges enqueue table freelist  1,492,275  1,903  
> 1241780/122/1/0 /0
> library cache pin   3,201,008   1,437 
> 1301307/130/0/0/0
> row cache objects1,400,498   1,020
>   56  964/56/0/0/0
> row cache enqueue latch  1,292,843 715   
> 19  696/19/0/0/0
> 
> It is holding row cache lock (v$session_wait), other sessions 
> are in the queuedue to this,  number of concurrent 
> sessions will increase from 100 to 350 sessions on each node. 
>  In less than one minute everything will be cleared ( OLTP database)
> 
> I have already opened a TAR with Priority 1.  Do you have any 
> suggestions.
> 
> Thanks 
> Muqthar Ahmed
> DBA
> 
> -- 
> Please see the official ORACLE-L FAQ: http://www.orafaq.net
> -- 
> Author: Muqthar Ahmed
>   INET: [EMAIL PROTECTED]
> 
> Fat City Network Services-- 858-538-5051 http://www.fatcity.com
> San Diego, California-- Mailing list and web hosting services
> -
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') 
> and in the message BODY, include a line containing: UNSUB 
> ORACLE-L (or the name of mailing list you want to be removed 
> from).  You may also send the HELP command for other 
> information (like subscribing).
> 




Note:
This message is for the named person's use only.  It may contain confidential, 
proprietary or legally privileged information.  No confidentiality or privilege is 
waived or lost by any mistransmission.  If you receive this message in error, please 
immediately delete it and all copies of it from your system, destroy any hard copies 
of it and notify the sender.  You must not, directly or indirectly, use, disclose, 
distribute, print, or copy any part of this message if you are not the intended 
recipient. Wang Trading LLC and any of its subsidiaries each reserve the right to 
monitor all e-mail communications through its networks.
Any views expressed in this message are those of the individual sender, except where 
the message states otherwise and the sender is authorized to state them to be the 
views of any such entity.

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.net
-- 
Author: Mladen Gogala
  INET: [EMAIL PROTECTED]

Fat City Network Service

ROW CACHE HIGH - Priority 1

2003-09-24 Thread Muqthar Ahmed
Hi,

I have Oracle 9.2.0.2.0 RAC (two nodes) on IBM AIX.  Currently I am seeing very high 
number for ROW CACHE LOCK in statspack.

Top 5 Timed Events
~~ % Total
Event  WaitsTime (s)   Ela Time
  ---  
   
row cache lock11,310  5,44186.97
CPU time   522   8.34
global cache cr request32,513  711.14
global cache null to x   21,507  5791
log file sync22,689  49.78
  -

 Get   
   Spin &
Latch Name   Requests   Misses  SleepsSleeps 1->4
--   --  ---  
---   
library cache 7,094,208   31,499   1,480  
30031/1456/12/0/0
shared pool  2,385,408   6,739  527
6212/527/0/0/0
ges enqueue table freelist  1,492,275  1,903  1241780/122/1/0 
/0
library cache pin   3,201,008   1,437 130
1307/130/0/0/0
row cache objects1,400,498   1,020  56  
964/56/0/0/0
row cache enqueue latch  1,292,843 715   19  696/19/0/0/0

It is holding row cache lock (v$session_wait), other sessions are in the queuedue 
to this,  number of concurrent sessions will increase from 100 to 350 sessions on each 
node.  In less than one minute everything will be cleared ( OLTP database)

I have already opened a TAR with Priority 1.  Do you have any suggestions.

Thanks 
Muqthar Ahmed
DBA

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.net
-- 
Author: Muqthar Ahmed
  INET: [EMAIL PROTECTED]

Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web hosting services
-
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).