I've found both as a manager in Oracle and in Miracle that if you schedule
regular meetings with the guys (and girls), then they start not showing up
after a while, or the meetings become boring. If you don't hold regular meetings
they'll complain and wish for regular meetings.
So I've come
Hi,
We're planning to consolidate a large amount of our production
databases into one database. Our customers will be charged based on
their use of the central machine's resources (other suggestions are
very welcome). I saw a method of keeping track of resource usage, by
means of the auditing
Hi All,
Suppose there two layer a(global ) and b(local).
I have a process running on layer b. This process access both Layer b and a.
and execute some sql-statement on both layer.
Is there any way to identify the corresponding trace on layer a if I know
the sql-tracefile on layer b?
Manoj
--
Hi,
First of all, you should use nologging. This helps a lot.
Second, when using parallel clause, the sort_area_size of the
ora_pxxx size is not the sort_area_size of your session, it is the
instance's default size, when it is started. I have verified it.
So if you want to
Hi!
You could easily use rownum for doing this kind of update... Just do a
commit after each update.
SQL create table u (a number);
Table created.
SQL alter table u modify a constraint un not null;
Table altered.
SQL insert into u values (0);
1 row created.
SQL insert into u values (0);
1
Hi!
Put it that way that since ROWNUM is a pseudocolumn and a reserved word, you
shouldn't include ROWNUM in your outer query if you want to limit results
from your subquery. The correct example would be:
select r, object_id
from (select rownum r, object_id
from dba_objects
where
i didnt that try on Oracle, but our mainframe Rdms. We have that situation
there with the ascending en descending attributes in the primary key.
The workaround is to create a unique index (with and without desc) and
create afterwards a primary key constraint on it in oracle.
we have oracle
Hi Robin,
LogMiner is a good tool for forensics, after the fact type work i.e to
find out when your junior dba dropped that table. It shouldn't be used
to do run of the mill daily work. It has a few deficiencies that are
hard to work around:
o - log miner user PGA memory so cannot be used in MTS
Ryan is right do this:
DECLARE
TYPE myRef IS REF CURSOR;
myCursormyRef;
TYPE myType IS TABLE OF Table_Name.PKid%TYPE INDEX BY BINARY
INTEGER;
p_limit PLS_INTEGER := 1000;
P_PKid myType;
Hello everyone,
We are looking for high availabilty database solution for a product
which is under initial development.
We have suggested Oracle RAC as the database option. Can we go for IBM
DB2 ESE as another database option ?
Can anybody give good references of organisations having production
Yes, for fun.
Three years ago I found a freeware utility called crontab for Windows.
Works well, I don't use AT anymore.
Especially when one autumn morning after the time jumped back an hour, I
noticed that two AT backup jobs had failed to start.
That was on NT Server 4.
Patrice.
Pradeep,
I don't know of any specific cliff notes on RAC in production, but I have
been involved in extensive testing of RAC/TAF for our environment for about
18 months. If you have specific questions, feel free to send them to me.
[EMAIL PROTECTED]
-Ron-
-Original Message-
Pradeep
Thanks but I am not interested in winat, it relies on AT.
I have the resource kit.
In my opinion the problem is AT, not the interface.
Patrice.
-Original Message-
Sent: Wednesday, July 09, 2003 9:06 AM
To: Multiple recipients of list ORACLE-L
Attached is the GUI program to
I have a problem with utl_file in Oracle 9 on Linux, standard engine
It does not seem to want to read lines longer than 997 characters. It works
fine if the line is 997 characters or less.
I get a utl_file.write_error exception if the line is longer than 997
charcaters!!! Why a write
Title: RE: [PMX:#] update about 100K records
How about creating a function based index on the column ...
function my_func(table_column in number) return number
is
begin
if table_column = your_current_value then
return 1
else
return null;
end if;
end my_func;
/
Then use this function
Hi!
Desc utl_file shows:
FUNCTION FOPEN RETURNS RECORD
Argument Name TypeIn/Out Default?
-- --- --
ID BINARY_INTEGER OUT
DATATYPE
John,
UTL_FILE is one of the worst designed functions I've ever tried to use.
In my opinion, it's a major design flaw to use the newline character ('\n')
as a packet delimiter. If UTL_FILE gets input lines that are too long (too
many bytes between '\n' characters), you'll get an error. If you
Thought I already was??? I set max_linesize to 998. I have also tried
with 32767.
utl_file.FOPEN(var_transfer_dir,var_file_name||'.KIC','r',32767);
Still no joy
-Original Message-
Sent: 09 July 2003 14:40
To: Multiple recipients of list ORACLE-L
Hi!
Desc utl_file shows:
that is a really good idea. discard my dopey pl/sql post. this one is alot better.
i must have been sniffing glue when i wrote it.
From: Jamadagni, Rajendra [EMAIL PROTECTED]
Date: 2003/07/09 Wed AM 09:39:24 EDT
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Subject: RE:
seems like alot of long time DBAs prefer using perl over pro*c to do data loads and
unloads. is it just because its simpler? or is it more robust? or other reasons?
From: Cary Millsap [EMAIL PROTECTED]
Date: 2003/07/09 Wed AM 09:44:25 EDT
To: Multiple recipients of list ORACLE-L [EMAIL
Hi!
Actually I read your post more thorougly, it seems that you have specified
max linesize already.
But this var_current_line, how is this one defined? If it's reduce it's size
to varchar2 (to 998) as well.
Tanel.
- Original Message -
To: Multiple recipients of list ORACLE-L [EMAIL
Title: RE: RE: [PMX:#] update about 100K records
This kind of FBI is another way of selectively indexing or indexing a subset of rows by using FBI. Ahh ... the power of NULL value.
Raj
Rajendra dot Jamadagni at
Hallo,
I would like to pick out distinct both of the field Code and Id from this sql
statement? Anyone who has an example of this?
Maybe this sounds simple but really appreciate somehelp.
I have this query,
Select code, id, name, address, city, year,
and would as told like to have distinct of
Well, not quite sure what the requirement is here - in the first instance
try
select code, id from table_name group by code, id;
or you could get a bit more info with
select code, id, count(*) from table_name group by code, id;
... and those results may give you some ideas.
peter
Hi
If you do : select distinct code || ' ' || id, city, year ...
Regards
-Message d'origine-
De : [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Envoyé : mercredi 9 juillet 2003 14:35
À : Multiple recipients of list ORACLE-L
Objet : Question about sql
Hallo,
I would like to pick
I tried it on my sun box, still same
problem.
Also I tried with no
compressing/uncompressing of dmp file, no luck.
-Original
Message-From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]Sent: Tuesday, July 08,
2003 5:25 PMTo: Multiple recipients of list
ORACLE-LSubject: RE:
I suggest calling Oracle Support. Get a test case ready.
Please post how the problem gets resolved.
- Kirti
--- [EMAIL PROTECTED] wrote:
I tried it on my sun box, still same problem.
Also I tried with no compressing/uncompressing of dmp file, no luck.
-Original Message-
Surendra,
Go back to
the 8i database and run a query to see if parent records exist for all child
records. It sounds like your source database is
bad.
Tom Mercadante Oracle Certified
Professional
-Original Message-From:
[EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]Sent:
When all else fails...
Log a TAR with Oracle
Sit back and wait, while enjoying a few belts of quality Kentucky sipping Bourbon
Read through the phone book and locate all the Ursulas in the greater Frankfort area...begin:vcard
n:Fink;Daniel
tel;cell:303.808.3282
tel;work:303.272.3225
When using parallelization with higher values for S_A_S, keep in mind that *each* PQ
slave will
potentially use that much memory for sorting. Should this sorting use disk, it would
cause even
more direct i/o to temp tablespace. Parallelization can very easily paralyze you
system ;)
Also,
I have opened TAR this morning.
Sure, will let you know.
Thank you all for your time.
Surendra
-Original Message-
Sent: Wednesday, July 09, 2003 11:30 AM
To: Multiple recipients of list ORACLE-L
I suggest calling Oracle Support. Get a test case ready.
Please post how the problem
Hi there
Hope this helps.
First the ksh file.
#!/bin/ksh
#
# File : check_alertlog.ksh
# Created : 01/02/2002
# Modified : 04/02/2002
# : 20/02/2002 Removed the CC for the email
#
# By: George Leonard ([EMAIL PROTECTED])
#
Pradeep,
I think if you contact Oracle directly they could certainly provide you with
some reference RAC customers that you could talk to.
RF
-Original Message-
To: Multiple recipients of list ORACLE-L
Sent: 7/9/2003 5:34 AM
Hello everyone,
We are looking for high availabilty
Simpler, portability
Richard
-Original Message-
Sent: Wednesday, July 09, 2003 10:04 AM
To: Multiple recipients of list ORACLE-L
seems like alot of long time DBAs prefer using perl over pro*c to do data loads and
unloads. is it just because its simpler? or is it more robust? or other
I did that already. Please see
below..
ERROR 1:
=
IMP-00017: following statement failed with
ORACLE error 2298:"ALTER TABLE "TWESAU" ENABLE CONSTRAINT
"FK_LC_AU_SYSID""IMP-3: ORACLE error 2298 encounteredORA-02298:
cannot validate (CMS.FK_LC_AU_SYSID) - parent keys not found
Joe and Pete,
Thanks for the response, that confirms my thoughts.
Robin
Pete Finnigan wrote:
Hi Robin,
LogMiner is a good tool for forensics, after the fact type work i.e to
find out when your junior dba dropped that table. It shouldn't be used
to do run of the mill daily work. It has a
Hi Pete,
I am a little prerplexed by selects are not recorded in redo prior to 9i .
AFAIK selects are nevere recorded in the redo, and therefore archived logs -
so they are never unearthed by LogMiner, even in 9i Release2. Isn't that
true?
Thanks.
Arup
- Original Message -
To: Multiple
I definitely dont fully understand ROWNUM yet,
and you guys so far provided more info than a couple of books
by Oracle, that I have here. For a second I thought I'm
beginning to get it, but the queries suggested produced no results...
SQL SELECT ROWNUM, GENDER
2 FROM (SELECT ROWNUM,
Arup, I kinda shook my head on that one also. So Pete, you got some
inside-scoop that i've missed, i admit i've not much touched logminer in
9ir2(did alot of it and presentations back in 8i days), but have been
spending my days now doing DR/RMAN implementations.
joe
Arup Nanda wrote:
Hi
MaryAnn,
the best way to understand rownum is to do the following:
SQL SELECT ROWNUM, GENDER
2 FROM (SELECT ROWNUM, GENDER
3 FROM EMP2
4 WHERE ROWNUM = 20)
You will quickly see that, no matter how you order the result set, the first
record returned is rownum #1,
Only less and = queries with regards to rownum are supported.
RF
-Original Message-
To: Multiple recipients of list ORACLE-L
Sent: 7/9/2003 11:19 AM
I definitely dont fully understand ROWNUM yet,
and you guys so far provided more info than a couple of books
by Oracle, that I have
To add on to my previous comments, change your code a bit:
SQL SELECT RN, GENDER
2 FROM (SELECT ROWNUM RN, GENDER
3 FROM EMP2
4 WHERE ROWNUM = 20)
5 WHERE RN 10;
-Original Message-
To: Multiple recipients of list ORACLE-L
Sent: 7/9/2003 11:19 AM
I
Each row that is retrieved from the database that matches the WHERE condition is
placed in a result set. The sequence within the result set is the value in the
pseudocolumn ROWNUM. If Oracle is using a full-table scan, the first row of the first
block is the first in the result set and is
Its obvious I hadnt fully understood ROWNUM yet, as you see
we are learning bits and pieces as we go along.
Help me out here, will you? Talk to me like I'm a 10-year old, its ok.
MaryAnn, the best way to understand rownum is to do the following:
SQL SELECT ROWNUM, GENDER
2 FROM
Think of it this way: when Oracle builds a result set, it goes
into the table and tries to qualify the records in the table
by evaluating the where clause; if the record does not qualify,
the record will not appear in the result set, if the record does
qualify, then the record will be returned in
Hey Jeff, just out of curiousity, do you guys all get company cars?
--Walt Weaver
Bozeman, Montana
-Original Message-
From: Eberhard, Jeff [mailto:[EMAIL PROTECTED]
Sent: Wednesday, July 09, 2003 10:35 AM
To: Multiple recipients of list ORACLE-L
Subject: RE: Tech meetings
When
THATS THE ONE!!! THAT DID IT!!!
--- Freeman Robert - IL [EMAIL PROTECTED] wrote:
To add on to my previous comments, change your code a bit:
SQL SELECT RN, GENDER
2 FROM (SELECT ROWNUM RN, GENDER
3 FROM EMP2
4 WHERE ROWNUM = 20)
5 WHERE RN 10;
The last
I definitely dont fully understand ROWNUM yet,
and you guys so far provided more info than a
couple of books
by Oracle, that I have here. For a second I thought
I'm
beginning to get it, but the queries suggested
produced no results...
SQL SELECT ROWNUM, GENDER
2 FROM (SELECT ROWNUM,
Those things, and its got a much better return on investment in terms of
extensibility. As a systems chappie, I started out writing things in
shell scripts, small C programs, etc. But I noticed that every time I
did anything that provided information (ran a report, data aggregation,
log mining,
I never liked those 'Power Lunch' meetings, even if the lunch was provided by the
employer ( back
when they had budgets for thost things :)
I always believed, and still do, that those '30-minutes' are for me. It's my lunch
time. It's time
for a break from the non-stop work. And while I am
Mary Ann,
Assume that the rows are selected in random order. Primary Keys and Order By exist
for ordering and qualifying your data. If you just select * from EMP2, most likely
the rows willl come back in the order that they were inserted, but this is not
necessarily true, especially if
Thanks for the explanations you've all provided.
I still believe that it's easier and faster to expand the RBS
tablespace ...
Not an option. It never was. The requirements were not set by me
but the agency(dont ask which one).We cant just go there and tell them,
hey you, increase RBS, NOW!
Oracle does not guarantee the order in which records are retrieved (read in
from the disk or data buffers) nor does Oracle guarantee the order in which
records are presented to you unless you specify an ORDER BY. The ROWNUM is
assigned after Oracle has selected the record to appear in the result
Can the inner-outer idea be extended to an update?
UPDATE EMP2
SETGENDER = 'F'
WHERE GENDER = ' '
ANDROWNUM BETWEEN 10 AND 20;
That does NOT work, because as we explained the returned rows
are numbered starting from 1, and so rownum never really matches
a 10 or 20 or anything
You should not have to relink Oracle, however, I would make sure that Oracle
has no issues with the release of the firmware being installed. I have seen
several cases where disk firmware caused performance problems.
Thank You
Stephen P. Karniotis
Strategic Alliance Manager
Compuware Corporation
IF you are looking to break up the update into more 'manageable' pieces, here is a
thought...
UPDATE EMP2
SET GENDER = 'F'
WHERE GENDER = ' '
AND ROWNUM = 100;
UPDATE EMP2
SET GENDER = 'F'
WHERE GENDER = ' '
AND ROWNUM = 100;
*Since the first 100 records have already been updated, the second
just what is it you want to return in your query?
If the query results are to be displayed a chunk at a time on a web page
then you should give strong consideration to using OCI and implementing
scrollable cursors with the OCIStmtFetch2() function and its
OCI_FETCH_NEXT, OCI_FETCH_PRIOR,
So, there I am, reading about 9iR2's compressed tables feature at:
http://otn.oracle.com/oramag/webcolumns/2003/techarticles/poess_tablecomp.ht
ml
(wrap URL if necessary, or go to http://tinyurl.com/gg3p )
Other than it being the best thing since sliced bread, what's the downside?
Are there
Hi, I'm using the OEM 92010 with oracle 92 when I look
for the option SHOW TABLESPACE MAP I can't find it,
I remember using the OEM with 806 oracle DB and that
option was available choosing
STORAGE-TABLESPACES-tb_name and then click right
button, in the database's tree.
What happened with the TS
next tested it but id guess. dont know about 8i. why would you want to do that? if you
post what your goal is, ther emay be a better method.
UPDATE EMP2
SETGENDER = 'F'
WHERE GENDER = (Select gender.(nested stuff) here.
same as from that link.
From: MaryAnn Atkinson [EMAIL
Well, you must load the dimension tables before the fact tables.
Just beware of the precedence.
Usually you cleansed the data and put it in a staging table that is
identical to the target table, then you do load the target table.
In my experiences we were not loading the time dimension, we
I wish. I always get that question. I work at a gear box manufacturing
plant which primarily manufactures gear boxes for jet engines. Rolls-Royce
plc doesn't even make the cars anymore. In 1998 Rolls-Royce Motor Cars was
sold by Vickers to Volkswagen, although BMW hold the rights to the name
Just wanted to say thanks for the information.
Bryan
-Original Message-
Sent: Tuesday, July 08, 2003 6:54 PM
To: Multiple recipients of list ORACLE-L
|Oracle says it is not their problem. They are waiting for the confirmation
|that the file is written. It is written just like any
It's still there. It's labeled as Show Tablespace Map in the second
section. Make sure that you have the Tablespace selected before you
right-click.
Tom Mercadante
Oracle Certified Professional
-Original Message-
Sent: Wednesday, July 09, 2003 2:15 PM
To: Multiple recipients of list
Title: RE: Compressed tables
Waleed,
the attached paper was compressed by Listguru ... for you, with good compression ratio too.
Raj
Rajendra dot Jamadagni at nospamespn dot com
All Views expressed in this
Title: RE: Tech meetings
Okay ... so you don't get a car ... do you at-least get an engine once in say 5 years ??
Same thing happens to me, I always get the question .. Do you get tickets to all the games? the answer is a big NO. We may get to see them on our internal cable network, but
Tools - Tuning Pack - Tablespace Map
- Kirti
--- Gabriel Aragon [EMAIL PROTECTED] wrote:
Hi, I'm using the OEM 92010 with oracle 92 when I look
for the option SHOW TABLESPACE MAP I can't find it,
I remember using the OEM with 806 oracle DB and that
option was available choosing
Thanks, Waleed! I also found that paper on OTN. It seems to answer my
questions.
Thanks again!
Rich
Rich Jesse System/Database Administrator
[EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA
-Original Message-
From: Khedr, Waleed
Click on:
$ - $$ - $$$
Just keep issuing the P.O.s to Oracle and sooner or later you'll get to
it...
Rich
Rich Jesse System/Database Administrator
[EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA
-Original Message-
From: Kirtikumar
Hi all. Thanks to all who replied. I'm still trying
to speed the things up and wondering whether someone
can explain what is paging to file system is
referring to? Does this indicate that I don't have
enough memory? Is it related to reads? Anything else?
thanks
--- zhu chao [EMAIL PROTECTED]
mmm I dont have the Tuning Pack option in my Tools
menu, I dont understand, it's supossed to be the last
version:
Oracle(R) Entrepise Manager
Version 9.2.0.1.0 Production
Copyright 1998,2002, Oracle Corporation
maybe a bad installation?
--- Kirtikumar Deshpande
[EMAIL PROTECTED] wrote:
Title: RE: Compressed tables
Sorry,
I think it's here too: http://otn.oracle.com/products/bi/pdf/o9ir2_compression_twp.pdf
Also
one of the things I like about it is moving some of the disk IO system
contention to the CPUs when there is extra CPU resources that is not being
utilized.
IF you are looking to break up the update into more 'manageable'
pieces, here is a thought...
UPDATE EMP2
SET GENDER = 'F'
WHERE GENDER = ' '
AND ROWNUM = 100;
Yes, that was exactly my objective, to update a large number
of records and commit them 1000 at a time. Here is what I did:
Is there a way to prevent end users from connecting directly
to the database via sqlplus without restricting access of those
same users via application code. The application is a third party
package which prompts for an id and password and then uses that
id/password to connect to the database.
I
Title: RE: Compressed tables
Thanks ... I did achieve about 70+% savings in space ... also one of the
thing that Oracle recommends is for greater compression
ratio
1. (somehow) load the data sorted ...
2. use larger block size if your version (and DBA and technical director
and Management
Title: RE: Restricting access via sqlplus
Something what I have used in past ...
DROP TABLE SYSTEM.SQLPLUS_COMMANDS CASCADE CONSTRAINTS
/
CREATE TABLE SYSTEM.SQLPLUS_COMMANDS
(
COMMAND VARCHAR2(20 BYTE) NULL
)
TABLESPACE TOOLS
PCTUSED 0
PCTFREE 10
INITRANS 1
MAXTRANS 255
STORAGE (
You mean you created an index A on the table with the descending column order, and
then added a primary key constraint, which created another index on the same columns
(except that all the columns are in ascending order).
Or did you do something different? If you did something different I'd be
Hi!
I think sqlplus product profile isn't a good idea, because some smarter ones
might be using TOAD, SQL Navigator or SQL Worksheet... What you might want
to do is to write an after logon trigger which checks the app name from
v$session and allows logon if and only if app name
INSERT INTO system.sqlplus_product_profile (userid, attribute, char_value)
VALUES (oracle_user_name, 'CONNECT', 'DISABLED')
/
Works for us. Most people can't connect via SQL*Plus. I can connect to the
production database under my own username, but not perform other DML such as
updates, because
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of Mark Richard
Sent: 08 July 2003 06:09
To: Multiple recipients of list ORACLE-L
Subject: Re: concatenated index
I didn't mean that including RULE will prevent the index from
being used - I
Few sql queries run fast (less than a second) when I use RULE hint and
without RULE hint it takes 30 seconds. The queries are not that
complicated, but joins 10 - 12 tables. I compared the plans between RULE
hint and witout RULE hint, they are almost the same except join methods.
Also, I checked
Title: RE: Restricting access via sqlplus
Tanel,
If I change TOAD.EXE to TODD.EXE, this scheme fails instantly ...
Raj
Rajendra dot Jamadagni at nospamespn dot com
All Views expressed in this email are
Thanks Stephane. Is there some place (some article) other
than the Oracle
Manual which deals with this(latest) features on the composite index
Asktom has or had an article on this try (will likely wrap)
http://asktom.oracle.com/pls/ask/f?p=4950:8:409221425463834426::NO::F495
Title: RE: concatenated index
no_index is a valid hint ... it tells Oracle that exclude specified index from your consideration.
Raj
Rajendra dot Jamadagni at nospamespn dot com
All Views expressed in this
Title: RE: Tech meetings
A
RR jet engine in a car Cute idea! There is an urban legend about a
military guy who tried
to
mount a JATO rocket on his Chevy Impala (nothing so fancy as R R). It
didn't end up
very
good for him. May be he should have tried with R R jet
engine.
Mladen
Title: RE: Tech meetings
-Original Message-From: Gogala, Mladen
[mailto:[EMAIL PROTECTED]Sent: Wednesday, July 09, 2003 4:59
PMTo: Multiple recipients of list ORACLE-LSubject: RE:
Tech meetings
A
RR jet engine in a car Cute idea! There is an urban legend about a
what are your
optimizer_index_caching
optimizer_index_cost_adj
optimizer_max_permutations
set to?
[EMAIL PROTECTED] 7/9/2003 3:39:31 PM
Few sql queries run fast (less than a second) when I use RULE hint and
without RULE hint it takes 30 seconds. The queries are not that
complicated, but
Title: RE: Restricting access via sqlplus
Hi!
No, if you code your trigger to check if the
program is your apps name, then renaming TOAD to TODD doesn't change
anything.
But of course if you change TOAD to your apps name,
then this scheme fails. But as I stated, these kinds of methods only
Hi!
First of course get the company, who sold you your development environment,
basic functionality like bind variables to work.
You could reduce your optimizer_max_permutations (default 8), but be
aware, that CBO might not be able to calculate that good execution plans for
Tanel,
Thanks for your suggestions, I'll do some more research on these parameters.
Thanks,
Bala.
- Original Message -
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Sent: Wednesday, July 09, 2003 5:39 PM
Hi!
First of course get the company, who sold you your
John,
I did not change these parameters from its defaults :
They are currently set to :
optimizer_index_caching = 0
optimizer_index_cost_adj = 100
optimizer_max_permutations = 8
Thanks,
Bala.
- Original Message -
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Sent:
Hi!
It's generally good idea to review optimizer_index_caching and
optimizer_index_cost_adj in CBO environments, but since you got parsing
performance problem currently, I'd not look at them first.
You can get elaborate information about those if you search for Tim Gorman's
The Search for
MaryAnn,
That was my original suggestion to you (although my typo in the rownum
query probably caused confusion). What you have done is a fairly safe way
to update in batches - the query is fairly simple and should never have
rollback issues although it will waste some time rescanning the table
You also might try out first_rows hint instead of rule. Oracle keeps
threatening to take the rule hint away.
Jay
-Original Message-
Sent: Wednesday, July 09, 2003 7:24 PM
To: Multiple recipients of list ORACLE-L
Hi!
It's generally good idea to review optimizer_index_caching and
Here is a good one:
create or replace directory bdump as '/oracle/admin/o9i/bdump'
/
grant read on directory bdump to public
/
create table alert_log_ext (
line varchar2(2000) )
organization external
(
type oracle_loader
default directory bdump
access parameters
(
records delimited by
John,
Thanks. I forgot all about LAST_DDL_TIME. Of course this hasn't happened
again, but at least I've got a strategy if/when it does.
Henry
-Original Message-
John Kanagaraj
Sent: Monday, July 07, 2003 3:10 PM
To: Multiple recipients of list ORACLE-L
Henry,
These locks should
To prevent these kinds of problems, the only solution I've ever thought was feasible
was how we implemented it at a previous company of mine.
I think this kind of solution was discussed on the list many months ago.
Problem description at my previous company:
We had a two-tier application
To prevent these kinds of problems, the only solution I've ever thought was feasible
was how we implemented it at a previous company of mine.
I think this kind of solution was discussed on the list many months ago.
Problem description at my previous company:
We had a two-tier application
Hi, yong:
This is the database of my friend, and it is in beijing.
I have found some suspected question and asked them to modify the
application to see if this works. They used row level security. I traced the
session with most library cache lock/pin wait event and found two
Could someone share with me if you come across similar
kind of configuration.
--- Mladen Gogala [EMAIL PROTECTED] wrote:
You obviously do need special configuration.
You are welcome.
On 2003.07.08 17:09, Bob Robert wrote:
All,
I am getting following error message when I try to
1 - 100 of 106 matches
Mail list logo
