Not sure of the terms. I have been approached to
troubleshoot the following scenario.
Using OID. Intergration server is used besides 3rd
party tools from Sun , IBM. The change logs need to
be picked up by the 3rd party tools. How do i do it.
TIA
--
Please see the official ORACLE-L FAQ: http
Wednesday, July 30, 2003, 1:19:31 PM, you wrote:
Q>> If you have 2 redo log groups with 4 members each, how many disks does
SK> Oracle recommend
SK>to keep the redo log files?
My first thought was 8, and I think you might be able to
make a case for 8. However, you should have at least 4,
becau
This is why you setup an orainventory file specific to each oracle home.
YMMV.
Ron Thomas
Hypercom, Inc
[EMAIL PROTECTED]
Each new user of a new system uncovers a new class of bugs. -- Kernighan
Jay:
When you do the install, insure that you use the 9i installer to install
the 8i software. If you don't , the 8i installer will write to the
inventory file and that would be bad. The 8i installer would not be able
to read the file and would not know that 9i is already installed. Believe
me,
Rick
Since nobody has responded to your question, I'll answer: No, haven't
tried that. Most of us work hard to keep a user from impersonating another
user. ;-)
Seriously, this whole system might get you fired someday. I can just see you
trying to explain how this is a legitimate request several
I haven't actually done it in this order, but I don't see any reason why
the order would matter as long as you install into separate homes. In case
you do run into problems, I would take out the refreneces to the Oracle 9
paths, including java from the path and classpath environment variables.
Both MS and Sun bought into this.
Sorry, don't have a url handy to support that.
Jared
"Jamadagni, Rajendra" <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED]
08/01/2003 01:49 PM
Please respond to ORACLE-L
To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
cc
No problem, did 4 of them just that way a couple days ago.
Jared
PS. As an added special bonus, here's a cmd file to apply patch 8.1.7.4.6
@REM from the instructions in the 8.1.7.4.6 README.txt file
@REM be sure to change these two variables.
set ORACLE_HOME=D:\oracle\PRD\ora81
set PATCH_HOM
I've done 8.1.7 client installs on my personal machine (win2k) that was also
running an 9.2 database with no troubles.
-Original Message-
Sent: Friday, August 01, 2003 4:50 PM
To: Multiple recipients of list ORACLE-L
Well, I'm on Linux, not Windows, but I installed 8.1.7 on a box running
Title: RE: Analysis of SCO vs Linux from Open Source Development Labs
Nah Micro$oft ... who else would like Linux to die?
Raj
Rajendra dot Jamadagni at nospamespn dot com
All Views expressed in this email a
Thanks, I didn't have much to do over the weekend anyway.
v/r
Stephen S. Wolfe, GS-11, DAFC
Data Services Manager
[EMAIL PROTECTED]
(813) 827-9974 DSN 651-9974
> -Original Message-
> From: Boivin, Patrice J [mailto:[EMAIL PROTECTED]
> Sent: Friday, August 01, 2003 11:04 AM
> To: Mu
Well, I'm on Linux, not Windows, but I installed 8.1.7 on a box running
9.2 a couple of weeks ago with no problems. Didn't do anything out of
the ordinary, just ran a plain ol' regular install.
--Walt Weaver
Bozeman, Montana
> -Original Message-
> From: Jay [mailto:[EMAIL PROTECTED]
> S
Hello:
I got an odd request today to install 8.1.7 on an existing server which is
running 9.2 (Windows 2000).
Has anyone successfully been able to install Oracle 8.1.7 on the same
Windows Server that already has 9.2 running?
I know that if you install 8.1.7 then 9.2 both in different homes it will
We were going to use logical standby but quickly abandoned the idea
after testing.
1. The APPLY process on the standby crashed regularly (every 10
minutes or so). Oracle claimed it was a fixed bug, but there was no
port to Windows, so we'd have had either to wait for the next RDBMS
release or r
Looks like it was an injection of cash.
Not sure but I think the other "undisclosed" company was SUN Microsystems
(?).
Patrice.
-Original Message-
Sent: Friday, August 01, 2003 4:09 PM
To: Multiple recipients of list ORACLE-L
Didn't Microsoft buy shares in SCO recently?
I vaguely reca
Cripes...I wasn't adding the format and just doing a straight TO_CHAR.
ghh.
Thanks people
-Original Message-
Sent: Friday, August 01, 2003 1:49 PM
To: Multiple recipients of list ORACLE-L
Try 9.000 instead.
- Kirti
--- "Odland, Brad" <[EMAIL PROTECTED]>
Didn't Microsoft buy shares in SCO recently?
I vaguely recall...
http://www.wired.com/news/business/0,1367,58904,00.html
Problaby I didn't understand...
Patrice.
-Original Message-
Sent: Friday, August 01, 2003 3:04 PM
To: Multiple recipients of list ORACLE-L
http://www.osdl.org/os
Title: MTS v$session
First
question I would have is Why does a user have access to V$SESSIONS in the first
place?
Dick GouletSenior Oracle DBAOracle Certified 8i
DBA
-Original Message-From: Wiegand, Kurt
[mailto:[EMAIL PROTECTED]Sent: Friday, August 01, 2003 11:19
AMTo: Mu
use a 0 instead of a 9 for leading/trailing zeroes.
SQL>select to_char(123.45, '999.990') from dual;
TO_CHAR(
123.450
SQL>select to_char(123.45678, '999.990') from dual;
TO_CHAR(
123.457
SQL>select to_char(123.45, '0999.990') from dual;
TO_CHAR(1
-
0123.450
> --
SQL> select to_char(1234.4, '0.000') from dual;
TO_CHAR(12
--
1234.400
-Original Message-
Sent: Friday, August 01, 2003 2:39 PM
To: Multiple recipients of list ORACLE-L
Tried that.
1234.1 won't show the trailing zeros...
-Original Message-
Sent: Friday, August
Could it be something with the program you are executing it from?
When I ran the 1234.1 in SQL*Plus I got the below. I'll see if I can dig up
another formatting optoin.
[EMAIL PROTECTED]:SQL> select TO_CHAR(1234.1,'9.999') FROM DUAL;
TO_CHAR(12
--
1234.100
[EMAIL PROTECTED]:SQL>
Try 9.000 instead.
- Kirti
--- "Odland, Brad" <[EMAIL PROTECTED]> wrote:
> Tried that.
>
> 1234.1 won't show the trailing zeros...
>
>
>
> -Original Message-
> Sent: Friday, August 01, 2003 1:19 PM
> To: Multiple recipients of list ORACLE-L
>
>
> Not sure if this is the best wa
Tried that.
1234.1 won't show the trailing zeros...
-Original Message-
Sent: Friday, August 01, 2003 1:19 PM
To: Multiple recipients of list ORACLE-L
Not sure if this is the best way but you could do something like SELECT
TO_CHAR(1,'999.999') FROM DUAL;
Only drawback that I h
It does not work as advertised, in AIX either... I played with this in AIX 5L.
- Kirti
--- "Hately, Mike (LogicaCMG)" <[EMAIL PROTECTED]> wrote:
> Stephen,
>
> The documentation is pretty wooly regarding this issue but the way it seems
> to be intended to work is this:
> At startup Oracle wi
Not sure if this is the best way but you could do something like SELECT
TO_CHAR(1,'999.999') FROM DUAL;
Only drawback that I have been running into is if you don't allot of enough
spaces before the decimal it display a value of .
Hope this helps.
- Original Message -
To:
okay that sounds a lot like the connection got broken (PC turned off,
sqlnet expire time, which I see you've set, etc) badly
being a "newbie" is the perfect time to ask. You can sound totally
innocent ("I was just wondering how you have things set up here")
--- "Wiegand, Kurt" <[EMAIL PROTECTED]
Title: RE: Metalink FTP Woes
we prefer to use something like getright or wget. Or I go home and
download it using cable modem, burn it on CD and bring it back to
work.
Raj
Rajendra dot Jamadagni at nospamespn
http://www.osdl.org/osdlpress/2003_07_31_beaverton.html
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author:
INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web hosting services
http://linuxtoday.com/news_story.php3?ltsn=2003-08-01-013-26-OS-BZ-DV
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author:
INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web ho
Thanks Tanel.
I didn't want to know badly enough to test it myself. :)
Looks like about 3.5 hours on your system.
Jared
"Tanel Poder" <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED]
08/01/2003 06:49 AM
Please respond to ORACLE-L
To: Multiple recipients of list ORACLE-L <[EM
Title: RE: MTS v$session
No, they are not "SNIPED"; I've worked with those before.
They are in fact all "INACTIVE". I suspect they are waiting on the
client for their next messages. As for your 2nd question:
I've got 7 users with the DBA role on this database! Why?
I'm too new here to a
How do I force display of a FLOAT
in a format like
Original number Result
1234.34 1234.340
12345.45612345.456
123.1123.100
123123.000
The dev on a project wants to preformat the data for a report on the
database side.
He wants 3 place
Title: RE: Metalink FTP Woes
I have been having similar problems downloading patches... all of a sudden it just... dies and provides the ever so useful error screens... I think they are having problems.
April Wells
Oracle DBA/Oracle Apps DBA
Corporate Systems
Amarillo Texas
Few people reall
I've been doing a lot (unfortunately) on Metastink lately and have been
having trouble uploading files with their web interface. I've
successfully uploaded several files and then when I try and upload
another file, I get an error that says the link I'm going to doesn't
exist or something about "me
Title: MTS v$session
Yes;
at least on the server side.
#
SQLNET.ORA Network Configuration File:
/u01/app/oracle/product/817/network/admin/sqlnet.ora# Generated by Oracle
configuration tools. #NAMES.DIRECTORY_PATH=
(TNSNAMES)SQLNET.EXPIRE_TIME= 10
-Original Message-From: Nual
Tanel Poder wrote:
>
> Hi!
>
> Statistic#'s vary between versions (and possibly platforms), so one should
> use statistic names instead of numbers in scripts.
> Statistic# 99 is "physical reads direct (lob)" in my test environment for
> example.
>
> Tanel.
An alternative is to use v$sess_io and
Title: MTS v$session
Hi
Kurt,
Have
you set the sqlnet.expire_time in the sqlnet.ora file?
N.
-Original Message-From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]On Behalf Of Wiegand,
KurtSent: 01 August 2003 16:19To: Multiple recipients of
list ORACLE-LSubject: MTS v$ses
Not only that, but the query shown will show redo generation for the
entire life of a session. So if a daemon has been logged in for a
long time but generating redo slowly, it can still look like the top
redo generator. To do this in a purely analytical way, a delta of two
selects from v$sesstat
Yes, this is a trace file that
contains
ORA-07445: exception encountered: core
dump...
- Original Message -
From:
Jamadagni, Rajendra
To: Multiple recipients of list ORACLE-L
Sent: Thursday, July 31, 2003 3:39
PM
Subject: RE: "union all" problems
What do
Hi!
Statistic#'s vary between versions (and possibly platforms), so one should
use statistic names instead of numbers in scripts.
Statistic# 99 is "physical reads direct (lob)" in my test environment for
example.
Tanel.
- Original Message -
To: "Multiple recipients of list ORACLE-L" <[EM
Hi DBAs,
Oracle 8.1.7.4
I want to be able to have users log into the database with an OS integrated
account and set up only certain users that log in to act on the behalf of
another integrated user account. (i.e. A manager is given access to a
database where he may delegate his authority to
This is a good time to mention that it's a lot better idea to query
V$SQL than to query V$SQLAREA when you really only need what's in V$SQL.
It's less workload to use V$SQL, and therefore it's more scalable.
Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com
Upcoming events:
- Hotsos Cl
He should monitor v$session and v$sqlarea for all processes with status =
'ACTIVE' and track those DML which are resulting so much redolog specially
when they are generating lot of redologs.
HTH,
Regards
Rafiq
Reply-To: [EMAIL PROTECTED]
To: Multiple recipients of list ORACLE-L <[EMAIL PROTECT
Thanks Mladen.
So, I was looking at the wrong place. Since I was referring to the column
elapsed_time in v$sqlarea, I expected to find it documented under
v$sqlarea. I did not have the foresight or the hindsight to look at v$sql :
-)
Thanks
Raj
--
Please see the official ORACLE-L FAQ: http:/
I would have to disagree there. We run multiple SQL Server instances (which consists
of the memory structure, physical files and logical structure) on one server. The
reason for this is multiple sort code, which is like the same as the character set in
Oracle. Personally I think the developers
One more thing to look at , if there were any object or tablespace in
nologging mode earlier ?
-ak
- Original Message -
From:
Munish Bajaj
To: Multiple recipients of list ORACLE-L
Sent: Friday, August 01, 2003 4:24
AM
Subject: RE: Redo Logs Problem
I'
Title: RE: Redo Logs Problem
Update activity is
obviously occuring at an alarming rate. Look at the times of the log
switches
and you should be able to pinpoint the 'when' Then, catch them in the
act! Look at v$sess_io and
find your 'heavy hitters', then grab their SQL or ,bet
Hi,
you may be able to identify the sessions that are generating a lot of redo
by running this query. It will list the amount of redo generated by eash
session.
select s.sid,s.value
from v$sesstat s
where s.statistic#=99
and s.value!=0
Cheers,
Mike Hately
-Original Message-
Sent: 01 A
folks
-
listening to a preso
from a vendor where they will use logical standby to accomplish some of the
99.999 availability requirements... so I'm wondering, does anyone
have some extensive experience with the logical standby to share? the good
stuff, the bad stuff, pitfalls, g
Probably an application has started generating lots of changes,
causing the jump in redo generation rate. The client needs to get a
delta of each session's value for 'redo size' in v$sesstat when this
is happening, then they can identify the culprit session.
There is one bug that could cause this
Title: Message
Sure. Check v$backup. STATUS will give you
part of your answer, but TIME will be the real key.
Match
it to the alert log and/or to the time you normally run your backup and you
should discover
what
happened. Typically this is caused by a failed backup. Some backup
scripts
Title: Message
This
may have already been asked, but is it possible there is a datafile in backup
mode even though a backup is not running?
-Original Message-From: Whittle Jerome
Contr NCI [mailto:[EMAIL PROTECTED] Sent: Friday, August
01, 2003 10:24 AMTo: Multiple recipien
did you look at the status column in v$session for the ones that don't
match?
IIRC, there can be a status of "SNIPED" where the session is not
actually connected but has not been cleaned up.
curiosity -- why is a user looking at v$session?
--- "Wiegand, Kurt" <[EMAIL PROTECTED]> wrote:
> I've got
Title: RE: Redo Logs Problem
>> This has suddenly started from the last 1 week without any changes to Database Configuration or any other system settings (as per client).
Clients lie.
Jerry Whittle
ASIFICS DBA
NCI Information Systems Inc.
[EMAIL PROTECTED]
618-622-4145
-Original
Title: MTS v$session
I've got an 8.1.7.4.0 instance running under Solaris 2.8.
I've never worked with MTS before, but got a call from a user telling me that there
were about 700 rows in v$session but they only had about 130 sessions open.
The user believes that when the session is ended,
Charu,
Allthough the trigger text is stored as a long, you can take the long
field, chop it at the linefeed characters and print it out (with line
numbers). The lines you get will match the line numbers in the error
message. Somewhere I have a pl/sql anonymous block that does that using
instr.
c
Stephen,
I suggest reading the concepts manual...
(for 9i Release 2, see
http://download-west.oracle.com/docs/cd/B10501_01/server.920/a96524/toc.htm
in .pdf format
http://download-east.oracle.com/docs/cd/B10501_01/server.920/a96524.pdf )
If you're going to work with Oracle you will have to read
Hi!
Which version of DB?
Do you allow end users into database?
Maybe someone is doing massive DML?
Maybe your temporary tablespaces are dictionary managed and in logging
mode - some user has forgot few where clauses in SQL and is doing huge
sorting/joining?
You should check redo size sta
Oracle, in a Real Application Clusters configuration, can run multiple
instances against one database.
--- Wolfe Stephen S GS-11 6 MDSS/SGSI <[EMAIL PROTECTED]>
wrote:
> I think I understand what you have said, Oracle runs an instance per
> database where MSSQL is one instance running multiple da
Hi!
> Oracle says this is what you would expect. It will grab the entire value
of
> SGA_MAX_SIZE.
>
> This certainly seems to be correct I just think it is stupid. Why grab
the
> memory from the os but never use it.
Check from man pages of ps, top or other utility, how to check how much of
the
You don't mention the database version, but if it is 8i or later, you could use
logminer to view the contents of the redo logs.
-Original Message-
Sent: Friday, August 01, 2003 6:34 AM
To: Multiple recipients of list ORACLE-L
Hi Listers,
One of my remote Clients is facing a problem
Hi!
Just try it out in test env...
There's gonna be a lot of compiling, inserts on source$, idl_ub1$ and
idl_sb4$ tables (actually more tables such are procedure$, access$, obj$
etc..). Of course if your trigger code is small, then no huge amounts of
data will be generated... Also 5 implicit
You didnt answer. Do u have more
load?
You can make out that info from
v$archived_log
Regards
Naveen
-Original Message-From: Munish Bajaj
[mailto:[EMAIL PROTECTED]Sent: Friday, August 01, 2003 4:54
PMTo: Multiple recipients of list ORACLE-LSubject: RE:
Redo Logs Problem
The line number referenced by pl/sql for packages, stored procedures
and functions is the dba_source.line column.
This won't work for triggers however, as they are stored in
dba_triggers as long text: no line numbers.
Jared
On Fri, 2003-08-01 at 05:20, Charu Joshi wrote:
> Is there any we way c
Stephen,
The documentation is pretty wooly regarding this issue but the way it seems
to be intended to work is this:
At startup Oracle will allocate an SGA sized as specified in the
sga_max_size parameter. This is to ensure that the system has enough memory
accomodate what you see as a maximum req
Sorry, I must have been shooting from the hip.
Not taking enough time to read more carefully.
The LAST_HOME is likely being read by TOAD.
Could be that TOAD stores the Oracle home it
was last initialized with, compares it to the
current on on subsequent startups, and tries
to revert by using L
Sorry for omitting that info.
Version: 8.1.7.4.1 ( maybe patched to 8.1.7.4.6, would
have to get on the server to check )
OS: Windows 2000 Server
Thanks for the insights John, I hadn't considered that it
might actually be a bug, given that this app often has
similar issues.
There are snapsh
Hi Poster ;)
Just a guess: Is it possible, that a
tablespace is still in hot backup mode ?
Or: Do you just happen to have more load
?
Stefan
Stefan Jahnke Consultant BOV
Aktiengesellschaft Tel.: +49 201/45
13-289 mailto:[EMAIL PROTECTED]
http://www.bov.de Abonnieren Sie unseren Newsle
Venu,
this does help and this is what the guys on metalink said.
What I do not understand - what a waste.
If I have to grab 100M initially on the OS side I may aswell use it.
In my metalink tar I quoted many references in the documentation:
E.G.
Is there any we way can see the 'line number' along with the
trigger code text. For ex. with stored procs, packages etc. we can
get the exact line number for a line of code. This helps greatly
when any error occurs, it mentions the line number which you can
jump to. Can we not do the same about tri
Hi!
AFAIK, Rule Based Optimizer always converts ORs to Union alls (except when
doing an outer join or connect by query). That's called OR expansion. CBO
seems to prefer inlist iterators:
SQL> create table t as select * from sys.obj$;
Table created.
SQL> select * from t where obj# = 1 or obj# =
Tanel,
I have done this.
ipcs -m would not give me the size of the memory acquired on my system so I
used ipcs -a
I have also looked into the ps vax
Which gives more detail:
Currently this shows 199M reserved to the SGA
EM9I /oracle/app/oracle/product/9.2.0/dbs > ps avx|head -1
PID TTY
I've Checked this in both the v$backup table
and v$datafile_headers table (fuzzy column). But the v$backup table shows "Not
Achive" and Fuzzy is Null.
Any Other Advice.
Regards
Munish Bajaj
-Original Message-From: Naveen Nahata
[mailto:[EMAIL PROTECTED]Sent: Friday, August 01
Hi!
You should check mem usage from OS instead. Using ipcs -m for example.
Tanel.
- Original Message -
To: "Multiple recipients of list ORACLE-L" <[EMAIL PROTECTED]>
Sent: Friday, August 01, 2003 12:54 PM
>
>
> Hi,
>
> does anybody have any experience with setting the SGA_MAX_SIZE in 9
Redo is not effected by database
configuration but by database activity.
Has the no. of transactions increased
significantly?
If the activity is the same, a wild
guess - Has someone issued a command "ALTER TABLESPACE BEGIN
BACKUP" and forgot to issue the command "ALTER TABLESPACE END
Have you checked whether
any tablespace is left in hot backup mode?
Kind Regards,
Hatzistavrou Yannis
Database Administrator
SchlumbergerSema
Phone ext. 478
Email: [EMAIL PROTECTED]
-Original
Message-
From: Munish Bajaj
[mailto:[EMAIL PROTECTED]
Sent: Friday,
Stephen,
That's the way it's meant to work.
This is how Oracle guarantees that you can grow upto the value of
SGA_MAX_SIZE. Look at the case below:
DAY 1: you have your sga_max set to 100m but you are currently using
only 50m of it.
DAY 2: you want to use all 100m and you try to increase the sga
> Of course it would make complete sense to give DBA's an option like
> "immediate" on such commands.
you do recall the oxymoron:
common sense Oracle
:)
--- Mark Richard <[EMAIL PROTECTED]> wrote:
>
> This is very easily explained once you consider that a "transaction"
> can
> consist of many
Yep, and they use the same definition of an hour as Oracle did with the
Oracle Applications One-Hour Install :)
> Hmm, must be a one hour course then ;-)
>
> On Thursday 31 July 2003 12:34, you wrote:
> > I know Microsoft is offering a new course called 'Microsoft
> SQL Server
> > 2000 for Exp
That's weird. I'm not aware of any security
issues relating to sar and its use. Sounds like
paranoia mode just because the thing might
have the setuid bit on. Which means nothing
as far as security goes, unless sar allows interactive
command shells. Which it doesn't.
Cheers
Nuno Souto
[EMAIL
Hi Listers,
One of my remote
Clients is facing a problem with Redo Logs. The Redo Logs and the Archive logs
in turn have suddenly started to generate at an alarming Rate. This has suddenly
started from the last 1 week without any changes to Database Configuration or
any other system settin
Hi,
does anybody have any experience with setting the SGA_MAX_SIZE in 9i.
I assumed the purpose of this parameter was that SGA would grow as
requested to that limit.
Example:
You could configure your SGA to be 80M
Set the SGA_MAX_SIZE to be 250M.
I would have expected oracle to acquire 80M of
Interesting indeed... I didn't see a version number in your note, but I am
assuming that it is 8.1.7.x. We have also had some _very_ strange
occurrences of 'library cache lock' / 'library cache pin' problems -
sessions en-masse went into this state during high activity times on a
Solaris based 8.1.
Jared,
> However, I have set the
> default home to ORA_8i_HOME(which has Net8) using the 'Home
> Selector' utility. Also, TOAD is reading the TNSNames.Ora file
> from this home, so should be using Net8 rather than SQL*Net.
The question is, is it TOAD that uses the LAST_HOME entry, or
Hi,
Ilooked at metalink for bugs on ORA-27101 and did not find anything.
Possibly because I'm on 9i and looked for that as well.
These are my Oracle settings.
The only thing odd may be that the lkORA92 file is all capitals whereas
everything else is initCap for the sid.
Only 1 Oracle is insta
85 matches
Mail list logo
