you only want DBWR_IO_SLAVES or multiple DBWRn if you have datafiles spread over
multiple I/O points correct? We are using 'Network Appliance' hard disk array that Im
not all that familiar with. It looks like we have 3 I/O points and 5 mount points.
my boss told me that striping data files and r
-Original Message-
Sent: Monday, June 02, 2003 6:50 PM
To: Multiple recipients of list ORACLE-L
Your case against LOB is
That was probably the "case against LONG"?
--
Mladen Gogala
Oracle DBA
Phone:(203) 459-6855
Email:[EMAIL PROTECTED]
--
Please see the offic
-Original Message-
Sent: Monday, June 02, 2003 11:15 AM
To: Multiple recipients of list ORACLE-L
Yes, it is fixed in 9.2.0
Hemant
At 04:25 AM 02-06-03 -0800, you wrote:
>Hope it's fixed in 9iR2 !
>
>Jp.
>
>2003/06/02 20:54:41, <[EMAIL PROTECTED]> wrote:
> >Cannot create lob columns if
Prem,
Use DBMS_JOBS to run the analyze:
===
undef username
undef password
grant analyze any to &&username
/
connect &&username/&&password
CREATE or REPLACE PROCEDURE analyze_tables (
v_stat_type IN VARCHAR2 := 'COMPUTE') AS
CURSOR c IS
SELECT D
Naula - Don't worry about it. You've given me something to look for the next
time. I'd rather learn from someone else's problems. :-)
Dennis Williams
DBA, 80%OCP, 100% DBA
Lifetouch, Inc.
[EMAIL PROTECTED]
-Original Message-
Sent: Tuesday, June 03, 2003 5:25 AM
To: Multiple recipients o
>
>Hi Listers
>
>I have a unique performance problem. As a general
>rule by oracle while
>writing SQL scripts EXISTS should be used in place
>of IN.
>
As a general rule there is no general rule. Why do you want to use EXISTS if it goes
faster with IN ?
EXISTS is mostly used with a correlated
Title: RE: Benchmarking of ASP.NET with Oracle v/s MSSQL
Please send to me as well!
-Original Message-
From: Satya Prakash Viswanath [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, June 03, 2003 12:40 AM
To: Multiple recipients of list ORACLE-L
Subject: Benchmarking of ASP.NET with Ora
--_=_NextPart_001_01C329BC.79FDEA20
Content-Type: text/plain; charset="iso-8859-1"
Munish
I've got a funny feeling that this thing about using EXISTS rather than IN
is a bit of a myth. I do a lot of this sort of thing and I find that almost
invariably, an IN with a simple subquery is faster
Guys,
I would like to scedule the process of analyzing tables/indexes
using DBMS_STATS ?
Hope someone of u would have a script for the same.
can u share with me please ?!
BTW,Which is advisable : ANALYZE or DBMS_STATS ?
there was a discussion about the same on the list also.
but not found any c
Munish,
You are right, as a *general rule of thumb* EXISTS is *usually* faster than
IN..
There can however be problems when an EXISTS is used to manipulate or select
data from a very large table, where the row exists in a far smaller table,
as it will read every row in the large table, and then s
there is an in & exists thread on asktom. Generally speaking exists is better if the
sub-query will have a larger and most costly result set than the outer query. Its the
other way around for 'in'.
I may have them backward, though I think that is correct.
>
> From: Munish Bajaj <[EMAIL PROTECT
Well, I was talking about cursors, sort areas and hash areas. I probably
did confuse "GA stuff".
On 2003.06.02 22:27 Tim Gorman wrote:
> Almost. It is the UGA areas (not the PGA) for Shared Server (a.k.a.
> multi-threaded server) that are re-located to the Large Pool, if it exists.
> Otherwise, th
Hi Listers
I have a unique performance problem. As a
general rule by oracle while writing SQL scripts EXISTS should be used in
place of IN.
I'm having 2 sql for comparison using IN and
EXISTS operators.
With IN operator
SELECT /*+ PUSH_SUBQ USE_NL (a) INDEX (a) */
COUNT(1)
FROM mam_assets a
When I queried the table to which SQL Loader loaded data from the .dat
file, the first record in the table was the 5th record and
4,49998...
Hi All,
Sorry I feel *really* stupid about this but here goes.
The database was picking up the wrong init.ora (the one it was picking up
had the archiving parameter set to false)
I've deleted all the old versions of the init.ora so this should n't happen
again.
Thanks for all the help and sugge
Hi,
When you create a RBS the creation entry will be there in the alert.log
Senthil.
-Original Message-
[EMAIL PROTECTED]
Sent: Tuesday, June 03, 2003 12:40 PM
To: Multiple recipients of list ORACLE-L
Is there any way to find out created time stamp for a rollback segment
apart from lo
>
>Is there any way to find out created time stamp for
>a rollback segment
>apart from looking into alert.log.
>
>TIA
>Kranti
>
If you think of some column in a dictionary table, I don't think so. Auditing is
another solution, besides scanning the alert.log file.
Regards,
Stephane Faroult
Oriol
Check your alert log file to see what file
is missing.
Cheers
-Original Message-From: Munish Bajaj
[mailto:[EMAIL PROTECTED]Sent: 03 June 2003 07:00To:
Multiple recipients of list ORACLE-LSubject: ORA-27101: shared
memory realm does not exist
Hi Listers,
What could be
Is there any way to find out created time stamp for a rollback segment
apart from looking into alert.log.
TIA
Kranti
This message is for the designated recipient only and may contain
privileged, proprietary, or otherwise private information. If you have
received it in error, please notify the
o Database is down.
o Wrong ORACLE_SID.
o If your database processes is indeed
up and you have the right ORACLE_SID, then
you database might be in trouble. You
can try to shut it down and restart it. Or you
may choose to reboot or use ipcs, ipcrm
to clean it up. Because as not to mess o
And you can do export/import of stats and change individual stats directly
using dbms_stats. Way to go.
Jankovic, Djordje wrote:
I have done some test before and found out that
analyze is faster than dbms_stats for smaller tables only (less than 10M)
- and the difference was
Hi Sai,
I'd like to help, but I'm not sure what your question is. Are you asking
whether a SLA (I don't really see them that often here in Denmark,
fortunately) should be static or dynamic? Or are you asking how to have
measurements that are independent of data growth and other varying
factors
Hi Listers,
What could be the problem?? I'm getting the
following error while connecting to the database on Solaris 8.
ERROR:ORA-01034: ORACLE not
availableORA-27101: shared memory realm does not existSVR4 Error: 2: No
such file or directory
Thanks to all
Best Regards
Munish Bajaj
<>
More against LONG:
(1) Only one LONG column per table, not LOBs.
(2) It's easier to manipulate LOBs than LONG, such as you can modify
a piece of LOB etc. So this means easier to code.
Richard Ji
-Original Message-
Sent: Monday, June 02, 2003 10:55 PM
To: Multiple recipients of list ORACL
hi gurus
this is a kind of query i have faced a few times in the recent past and which has really forced me to start this thread.
as everyone knows, there is always what we call a SLA or in other words a service level agreement (may be called differently in different places) which infact means
Hi List,
Anybody done performance benchmarking of a typical ASP.NET application
with IIS(or any web server) with Oracle database(9.2) v/s MS SQL
Server.(Latest version)
Any pointers for the same is most appreciated. If there is any document
on which POC has been done, plz mail to my id.
FYI: 80
Yes, I am talkin about the time it takes. The analyze command takes around 2 mins to
esimate statistics of a table with a Million records whereas the dbms_stats package
function is merely hanging !!
And besides, Djordje is telling that they nearly take up the same time !! But i am
not experi
I am looking at a project to upgrade a
number of Oracle development, test and production DB's to newer releases.
I have reviewed Oracle's list of tasks from OTN. I would like your
feedback on other tasks, etc. that should be included. Is there a list
somewhere on the web that gives a comp
Against LONG:
5) Those are stored in-line and can easily become a performance headache (with
chaining etc..)
- Kirti
--- Arup Nanda <[EMAIL PROTECTED]> wrote:
> Your case against LOB is
>
> (1) desupported in 10i
> (2) Can't replicate through Oracle replication
> (3) Can't do a Create Table
Hi
Thanks a lot.
I myself have been for too long
an 'upgrade /backup & recovery' type of dba
and have been 'struggling' to figure the java thing.
will read this one :)
Thanks
Cyril
On Tue, 03 Jun 2003 DENNIS WILLIAMS wrote :
>I just ran into this article, and it seems pretty good, aside
> fr
Almost. It is the UGA areas (not the PGA) for Shared Server (a.k.a.
multi-threaded server) that are re-located to the Large Pool, if it exists.
Otherwise, they reside in the Shared Pool and all hell breaks loose,
performance wise...
Also, if DBWR_IO_SLAVES > 0, then those IPC queues will reside i
I would use web services as the baseline technology.
Then the UI would pretty independent (Java, .Net, just about anything).
This would also allow you an easily expandable platform.
Awhile ago I saw a demo of JDeveloper and it seemed to make the process very
"functional".
I'm currently on a projec
Hi list,
I have to develop an interface to allow another company, with 60
offices, to make some kind of transactions in my system with my clients.
I am thinking of using Forms9i.
What should I consider ?
Any comments would be highly appreciated.
TIA,
Ramon E. Estevez
[EMAIL PROTECTED]
809-535-
Your case against LOB is
(1) desupported in 10i
(2) Can't replicate through Oracle replication
(3) Can't do a Create Table As Select if a column is LONG
(4) LONG columns can't queried like SUBSTR(), INSTR(), DECODE, etc.
You case for them to move from LONG to LOB
(1) All of the above
(2) The exi
I just ran into this article, and it seems pretty good, aside from being
oriented toward Oracle9iAS and Oracle9iAS TopLink. If you support Java, take
a look at it. If anyone has found any issues with the statements it makes, I
would be interested in hearing them.
http://otn.oracle.com/oramag/orac
John Shaw wrote:
>
> We are looking at purchasing a third party software package that is still using
> longs. I am trying to make the case for forcing them to upgrade to lob's before we
> will purchase their software. Aside from the pain of having to deal with them - are
> they going to stop be
I don't think that your vendor is supported now, at least not by oracle.
Mladen Gogala
Oracle DBA
Phone:(203) 459-6855
Email:[EMAIL PROTECTED]
-Original Message-
Sent: Monday, June 02, 2003 5:45 PM
To: Multiple recipients of list ORACLE-L
We are looking at purchasing a third party soft
Jay - Personally I did the Tuning exam next after B&R and saved the Network
for last because most of it is new to me. The B&R and Tuning exams share
quite a bit of material in terms of they both require a clear understanding
of the Oracle architecture, how the rollback, redo, SGA, etc. all interact
And in today's news...
> http://biz.yahoo.com/rb/030602/tech_jdedwards_peoplesoft_6.html
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: Vergara, Michael (TEM)
INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, Ca
We are looking at purchasing a third party software package that is still using longs.
I am trying to make the case for forcing them to upgrade to lob's before we will
purchase their software. Aside from the pain of having to deal with them - are they
going to stop being supported soon? If so do
Hello:
I resently finished my 8i Backup and Recovery test and was thinking about
taking the 8i Networking testing before performance tuning. I was wondering
if those of you that have taken both test feel that would be an ok choice of
action or would taking performance tuning first be better?
Nope, it's not accurate. PGA areas for shared server sessions are also
allocated from the large pool.
Mladen Gogala
Oracle DBA
Phone:(203) 459-6855
Email:[EMAIL PROTECTED]
-Original Message-
Sent: Monday, June 02, 2003 4:35 PM
To: Multiple recipients of list ORACLE-L
for some reason w
>From the doc:
LARGE_POOL_SIZE lets you specify the size (in bytes) of the large pool
allocation heap. The large pool allocation heap is used in shared server
systems for session memory, by parallel execution for message buffers, and
by backup processes for disk I/O buffers. (Parallel execution al
That is pretty much my understanding. Here is a Web page that explains it
pretty well:
http://www.interealm.com/roby/technotes/tnlargepool.html
Dennis Williams
DBA, 80%OCP, 100% DBA
Lifetouch, Inc.
[EMAIL PROTECTED]
-Original Message-
Sent: Monday, June 02, 2003 3:35 PM
To: Multiple r
for some reason we have 100MB large pool. I dont think we need it at all. I read that
its only used by RMAN or Parallel server. Is that accurate?
>
> From: DENNIS WILLIAMS <[EMAIL PROTECTED]>
> Date: 2003/06/02 Mon PM 03:39:42 EDT
> To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
>
Use the large pool to store what? I can think of 3 aspects of a transaction:
- Rollback (you've probably read about SET TRANSACTION)
- SQL statements, execution plans (more an issue with bind variables)
- Data blocks
It sounds like you might be thinking of data blocks. You didn't mention your
to get a speed advantage from dbms_stats in parallel do you have to the tablespace
and/or segments broken up over several mount points?
>
> From: "Jankovic, Djordje" <[EMAIL PROTECTED]>
> Date: 2003/06/02 Mon PM 02:34:41 EDT
> To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
> Subjec
Hi,
I have installed Oracle Administration Assistant 8.1.7.4 on Win 2000.
However the utility ocfgutil.exe was not installed.
Apparently this is a common problem according to MetaLink. Does anyone
have this utility that can be e-mailed
offline? If so please rename the extension.
Thanks
Rick
Check the disk capacity for the archive log destination.
Bonnie Finke
-Original Message-
Cullen
Sent: Monday, June 02, 2003 9:20 AM
To: Multiple recipients of list ORACLE-L
Hi All,
I keep getting a cannot allocate new log in the alert log. This freezes the
entire application and I
Wolfgang,
Thanks for your respond.
I tried gathering stats on the function based index using
- analyze index ACFD_INDX1 compute statistics;
- exec
dbms_stats.gather_table_stats(ownname=>'ACPO',tabname=>'AC_FORWARD_DEST',cascade=>TRUE);
- exec dbms_stats.gather_index_stats('ACPO','ACFD_INDX1');
bu
I think I read this somewhere, but I cant find it. Is it possible to use the large
pool for a specific transaction? We run alot of large batch DML statements over night.
We have one that involves an 8GB table. The blocks from this table are being knocked
out of the buffer cache by shorter and qu
I have
done some test before and found out that analyze is faster than dbms_stats for
smaller tables only (less than 10M) - and the difference was not that big (for
example very small tables dbms_stats would do a compute instead of
estimate). For large tables they are pretty much the same.
I am siding with Scott and would double-check that automatic archiving is
on. Also, what are your log_archive_dest settings? If you are using
log_archive_dest_n make sure you add the "reopen = nn" attribute.
At 08:49 AM 6/2/2003 -0800, you wrote:
Nuala,
It looks like the DB is waiting for you t
nice of you to be so generous with my already empty wallet :)
I will buy the list the virtual coffee drink of their choice (even tea
if they prefer)
--- "Daniel W. Fink" <[EMAIL PROTECTED]> wrote:
> As penance, I think you should pay for dinner for the entire list!
>
> Rachel Carmichael wrote:
Binley - Thanks for filling in the details. I suspected the answer might be
more complicated than just "balanced" or "unbalanced". Have you tried
reverse key indexes yourself? Results?
Dennis Williams
DBA, 80%OCP, 100% DBA
Lifetouch, Inc.
[EMAIL PROTECTED]
-Original Message-
Sent: Monda
I think the query is displaying data from the table in the order in which rows were inserted by SQL*Loader.
Whereas, the query to read the same data from the external table is reading the file from the first physical record. So it appears reversed..
- Kirti Stephen Andert <[EMAIL PROTECTED]> wr
Yoiur logfile size is too small to handle peaktime DML activities.
Possible solutions:
First track that job resulting in so much archiving. Check through v$sqlarea
with active sessions resuling in DMA.
Increase size of your logfiles and meantime also try to add more
group/members. For specifi
Try to add some more RedoLog groups.
Igor Neyman, OCP DBA
[EMAIL PROTECTED]
-Original Message-
Cullen
Sent: Monday, June 02, 2003 11:25 AM
To: Multiple recipients of list ORACLE-L
Hi Denis,
Thanks for the quick reply,
We're not getting an oracle error no: This is the message in the
Hi Tom
I always copy the diagrams to VISIO, print it from there.
You can adjust stuff (pagesize) better then within Word or the like.
But in cases, the diagrams get so big, that just a plotter
and 2 square meter paper does the job ;).
PS: If you have the Acrobat Distiller installed, it also
wo
Nuala,
It looks like the DB is waiting for you to manually archive the
logs. Like Lisa suggested, could you do an archive log list from
the sqlplus prompt and post the results? There are two steps in
setting up archiving. You have to set the proper parameters in the
init.ora file, and then bri
Nuala
Running off to a meeting, and Metalink seems to be down. I did a Google
search and found this link which seems to somewhat touch on your issue:
http://www.geocities.com/oraperf/redo/arch_tuning.html
>From your message, I infer that the database just "hangs" for awhile, then
corrects itself
Craig,
Could you please send the explain plans for each statement, the
value of db_file_multiblock_read_count and the following columns related
to tables, columns and indexes (user_tables.blocks,
user_tables.num_rows, user_indexes.blevel, user_indexes.leaf_blocks,
user_indexes.distinct_keys,
As penance, I think you should pay for dinner for the entire list!
Rachel Carmichael wrote:
sigh, it's Monday morning and I'm not properly caffeinated.
At least I didn't name names :)
For those who didn't know, Stephane Faroult is speaking at the NYOUG
meeting next week.
I'm going to go
Pradeep,
I don't know what you mean when you say SQL*Loader is "starting from
the 50,000th record and going up". If there are 50k records, are you
saying it is not loading them? Or is it reading them from 50,000 then
49,999 (i.e. in revers order)? How are you determining what order they
are be
It will be hard to find what changed since you don't have the information
on exactly what production looked like when it was using the index.
Whenever you do somthing which may affect the statistics, make a backup of
the current statistics with dbms_stats.export_table_stats(..., cascade =>
true
Hi Denis,
Thanks for the quick reply,
We're not getting an oracle error no: This is the message in the alert log
Mon Jun 02 08:19:19 2003
Thread 1 cannot allocate new log, sequence 711
All online logs needed archiving
Current log# 2 seq# 710 mem# 0: D:\ORACLE\ORADATA\GDA\GDA_21.LOG
Current l
MAIL> ROLLBACK
ORA-2: committed mail could not be retrieved
I'd suggest Tazo's "Awake" tea, preferrably the loose leaf stuff for maximum
caffeinage. :)
Rich
Rich JesseSystem/Database Administrator
[EMAIL PROTECTED] Quad/Tech International, Sussex, WI USA
Thanks...
Having been through two books and having aborted the third, I can only
repeat what a friend told me after completing his MBA at night school: "You
can work your job, live your life, and go to school. But, only two at a
time"...
Substitute "write a book" for "go to school" and that p
Nuala,
20M for log files might be too small. look in the alert log file and see
how often logs are switching. it could be that all four sets of logs are
full, and you are waiting for the archiver process to clear a log for you.
the database will hang until this process completes.
log switches s
The documentation is correct, but left out a lot of background information.
Oracle*trees are height balanced. That means that the level from root to
branch block is always the same. In the case of monotonically increasing
indexes, keys are always inserted at the high end. If keys are also deleted
Craig
I'm sorry, you said they were on the same instance, I just didn't read
carefully enough.
I believe that all the statistics that CBO uses to make a decision are in
USER_TABLES and USER_INDEXES. You might compare the values for both tables
to see if there is a difference that might cause
Jared and all,
thanks! GhostScript worked fine. Had to get a Postscript Print driver
first, but everything is just hunky-dorey!
thanks again
Tom Mercadante
Oracle Certified Professional
-Original Message-
Sent: Friday, May 30, 2003 10:10 PM
To: Multiple recipients of list ORACLE-L
Kesh - What do you define as performance?
1. How fast a table is analyzed?
2. The performance of the queries that use the statistics that were
collected?
What Oracle version are you on?
In theory, you can analyze tables when the system load isn't so great. In 9i
you can even have Oracle decide when
Nuala - You didn't include your error number, so the following are some
guesses: Is your database in archivelogmode? Check the path to the archive
logs and make sure that drive hasn't filled. For the logs themselves, check
the paths and permissions.
Dennis Williams
DBA, 80%OCP, 100% DBA
Lifetouch,
oops... let's hope she's not a member of this list...
-Original Message-
Sent: Monday, June 02, 2003 10:40 AM
To: Multiple recipients of list ORACLE-L
Stephane,
Are we still on for dinner on the 10th? Have you contacted Sean? I
think the group is trying to organize a dinner for the spe
Hi Folks,
Many thanks to Dennis, Raj, Mark and Roy. We are in the processing to
evaluate which method would be better to use in our case. Personally I
am very interested in Dennis's solution.
Thanks again,
Joan
DENNIS WILLIAMS wrote:
>
> Joan
> Okay, a coworker has experience in this area and
> -Original Message-
> From: Stephane Faroult [mailto:[EMAIL PROTECTED]
>
>Does 'etc' include doing something to tables as well? I
> would, from your posting, being tempted to say 'no'. Could it
> have anything to do with histograms ? The only thing I can
> currently think of is diff
Nuala,
Are you running your database in archive log mode?
Type ARCHIVE LOG LIST at the Sql*Plus command line to find out.
What error message are you getting? What is in your alert log?
Lisa Koivu
Oracle Database Monkey
Fairfield Resorts, Inc.
5259 Coconut Creek Parkway
Ft. Lauderdale, FL, U
Have you tried methods 3 or preferrably method 4:
method 3: analyze index compute statistics
method 4: exec dbms_stats.gather_index_stats(...)
What do you mean by "the output below is similar for both methods"? What
are the differences? Can you use Tom Kyte's print_table procedure to list
the
sigh, it's Monday morning and I'm not properly caffeinated.
At least I didn't name names :)
For those who didn't know, Stephane Faroult is speaking at the NYOUG
meeting next week.
I'm going to go get a caffeine drip now.
--- Rachel Carmichael <[EMAIL PROTECTED]> wrote:
> Stephane,
>
> Are we
Hi
all,
Can
anyone tell me which of the following performs better in collecting statistics
of a table:
Analyze
command or
Dbms_stats.gather_table_stats
I tried my queries with
both the options -- but always Analyze is out-performing the
dbms_stats.
But i read in a
doc
SQL> select * from v$version;
BANNER
Oracle9i Enterprise Edition Release 9.2.0.1.0 - Production
PL/SQL Release 9.2.0.1.0 - Production
CORE9.2.0.1.0 Production
TNS for 32-bit Windows: Version 9.2.0.1.0 - Production
NLSRTL Ve
- Original Message -
To: "Multiple recipients of list ORACLE-L" <[EMAIL PROTECTED]>
Sent: Monday, June 02, 2003 9:54 PM
>
> Cannot create lob columns if segment space management is set to auto.
>
>
Why not ?
SQL> create tablespace new datafile 'c:\bowie\new01.dbf' reuse
2 segment spa
Yes, it is fixed in 9.2.0
Hemant
At 04:25 AM 02-06-03 -0800, you wrote:
Hope it's fixed in 9iR2 !
Jp.
2003/06/02 20:54:41, <[EMAIL PROTECTED]> wrote:
>Cannot create lob columns if segment space management is set to auto.
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author:
This is [hugely] not my area, but is TSPITR a possible option?
Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com
Upcoming events:
- Hotsos Clinic 101 in Reykjavik, Ottawa, Dallas, Washington, Denver, Sydney
- Visit www.hotsos.com for schedule details...
-Original Message-
Far
Does this look familiar?
Bug No. 2782919
Filed 03-FEB-2003 Updated 20-MAY-2003
Product Oracle Server - Enterprise Edition V7 Product Version 9.2.0.2.0
Platform Sun SPARC Solaris (64-BIT) Platform Version 2.8
Database Version 9.2.0.2.0 Affects Platforms Generic
Priority Severe Loss of Servic
>OK, I'm confused. Maybe it's Monday morning and my
>brain's not working.
>We have a production schema and a test schema on
>the same Oracle 8.1.7
>instance, running on Windows. They both have a
>customer table, with 3
>million and 2 million records respectively. They
>both have the same
>indexes,
Craig - Compare the init.ora file for these instances.
Dennis Williams
DBA, 80%OCP, 100% DBA
Lifetouch, Inc.
[EMAIL PROTECTED]
-Original Message-
Sent: Monday, June 02, 2003 8:45 AM
To: Multiple recipients of list ORACLE-L
OK, I'm confused. Maybe it's Monday morning and my brain's not
Stephane,
Are we still on for dinner on the 10th? Have you contacted Sean? I
think the group is trying to organize a dinner for the speakers, but
knowing the person who is doing the organizing... it's not going to
happen. she's one of the more incompetent members of the board.
In any case...
Indy - First, I would urge you to contact Oracle Support. You pay them for
support and they are pretty good at walking you through a recovery. A
problem with an email list like this is that you may get a variety of
opinions. For many issues that is good, but in a critical situation,
multiple opinio
Hi All,
I keep getting a cannot allocate new log in the alert log. This freezes the
entire application and I am unable to connect to oracle in order to do a
statspack snap shot.
This is an upgrade to Version 8.1.7.4 set up on a WK2000 (NEW) Server with 2
gigs of memory.
We are writing to 3 sepa
Hi,
Can someone shed some light on the differences of gathering statistics on a
function-based index using the following two methods?
method 1: analyze table compute statistics
vs
method 2: create unique index on
(upper(columne_name)) compute statistics;
I could not get the CBO optimizer to u
- Original Message -
To: "Multiple recipients of list ORACLE-L" <[EMAIL PROTECTED]>
Sent: Monday, June 02, 2003 7:39 PM
> Shuan, this sounds like a problem for support. Are you using I/O slaves?
>
>
> > Mladen,
> >
> > There's only a line in the .trc file:
> > "kxfpqsrls: Timed out t
93 matches
Mail list logo