Stephane,
I *did* warn you about the OT list before I approved your registration on
it..
btw, this sounds like it should have gone private, not to the list
Rachel
|+--->
|| |
|| |
||
as one of the women in this profession, do I want to know who you consider
"honorary men"? :)
What I do want to know is how you determine what the criteria are for being an
"honorary man".
Inquiring minds want to know :)
Rachel
|+--->
||
way back (7.3 version) I wrote a set of scripts to export to a pipe, split and
compress the split files.. It was based on the Metalink doc and I *think* copies
of them are still out on the archives on www.lazydba.com
but now, when you can split files via export, so the split is done by Oracle a
there is a BIG difference between the "COMPRESS=Y" parameter on an export and
compressing a file!
the parameter changes the "create table" statement placed in the export file so
that the initial extent is large enough to hold the entire table. It does NOT
affect the size of the export dump file
If you are on Unix, you can pipe the export into a "split" command and break the
file into multiples and compress on the fly. There's a note on metalink about it
(note 30528.1)
Also, I *think* in 8.1.7 you can specify the size and names of the export files,
so that Oracle will automatically bre
true -- but 99% of the time, it's in the application code (and you listed 4, not
3, additional reasons :) )
|+--->
|| |
|| |
|| kaygopal@yaho|
|| o.com|
|
no you can't recompile them... I tried that once, the recompiled procedure
referenced the renamed (old) table
it's the object_id thing. Oracle doesn't care what you name or rename an object,
it tracks it by the object_id.
Trust me, it screwed up triggers etc. real pita
|+--
the problem is in the application code... find the sql (it's in the trace files)
and start from there
|+--->
|| |
|| |
|| oracledbam@ho|
|| tmail.com|
||
there is a hardware problem that necessitated Oracle coming down. the only
question is does the database come down cleanly when YOU want it to or does it
come down with a crash, time undetermined, when the file is accessed?
|+--->
||
ashoke,
that's going to corrupt his database.. at some point. If you do that, you'd
better be willing to immediately shutdown, recreate the tablespace and then do a
backup
|+->
|| |
||
It's not just grants. Any procedure that references that table will have to be
recreated, oracle uses object_id not object_name so the procedure will point to
the old table etc etc
you CAN try to resize the datafile down to something really small, smaller than
the smallest extent if possible,
I suppose you can if you don't care about wasted disk space, I know everyone
says "disk is cheap" but that's only until you explain to your manager that you
need a 100GB disk for a 10GB (real used space) database because you are forcing
order by storing each row in one block and never reusing sp
that works unless there is also a reason to see insert not just in order but by
"inserted on a particular date or time". I suppose in that case, add two fields,
a date field for the time range and a numeric field for the sequence of insert
|+--->
||
Do you want to physically order them or do you just want to know by time the
order in which they were entered.
if the first, no, not that I know of. If the later, yes, add another column
(ins_date date) and a trigger to populate that column with sysdate when you
insert a row. You can then order
You can use the oerr facility on Unix to look at the events... I don't know if
that file exists under Windows, I vaguely recall a conversation on the list a
while ago where people were saying it didn't. I know that I downloaded the file
from my Unix box to my PC so I'd have a copy.
Anyone out t
yeah, but it's confusing when you look at that file...
|+--->
|| |
|| |
|| Rajendra.Jamadagn|
|| [EMAIL PROTECTED] |
||
actually that's the list of error messages, not events
and it's not under the admin directory in any case the path to the error
messages is:
$ORACLE_HOME/rdbms/mesg/roaus.msg
|+--->
|| |
||
log on using an account with CREATE ANY SYNONYM privs and create the synonym as:
create .synonym for
|+-->
|| |
|| |
|| Beth.Seefelt@tet|
|| leyusa.co
a default value.
|+->
|| |
|| |
|| andreyb@elronte|
|| lesoft.com |
|| |
|| 05/13/2002 |
||
give me a break! It took me 25+ years to get to the point where I have a group
who will listen to me. And that's only one small group in the larger group I
support, I doubt I'll be that lucky across the board
|+--->
|| |
||
That's CORRECT... no one other than SYS should have the SYSTEM tablespace as the
default tablespace. No one should have SYSTEM as the temporary tablespace.
for years oracle would say "change the default tablespace". Now they do it for
you (and about bloodly time!)
|+
Dennis,
It's possible the version you have was corrected. I know that it was the first
printing that had some serious, unintentinal omissions I think only in the
first few chapters.
What really hurts the author is when they catch the error, send the correction
in in time to be fixed for th
supposedly no additional cost in 9i, but when I checked Oracle's store, there
was a $20 charge for each. Considering that for 7.3 and 8.0 they were $100 for
EVERY NAMED USER, regardless of how many people were actually going to use them,
that's "no cost" :)
|+--->
Folks, just a word of warning. Through NO FAULT of the authors, there are a
number of errata in the first edition of the book.
There is a complete errata list on the Osborne site:
http://shop.osborne.com/cgi-bin/oraclepress/errata.html
Page proofing is one of the worst tasks on earth (I know
Don,
Next time I need to resign, will you help me write the letter?
Having worked at several large and not-so large companies, all I can say is, the
problems you documented seem to be ubiquitous throughout most management. I have
found it nearly impossible in some places to do my job with any
How is this essentially different than what you can get from utldtree.sql?
when Kevin Loney and I were writing the Annotated Archives a few years ago, I
tried to write a script that would show the dependencies, because part of the
reason for the book was to provide "home-grown" scripts that wre
And I respectfully agree that I was wrong :)It happens, I think I already
posted a note to the list apologizing and correcting my error.
This is what happens when I post from memory without double checking
obviously I had done a "flush" of my shared_pool!
Rachel
|+---
and stats have to have been collected on the index and base table as well
just checked the 9i docs... I was wrong you do need the upper on the column, my
bad.. but you need the stats, the compatibility set (as Beth says)
|+-->
||
you don't need the upper(ename) in your where clause... that forces oracle to
NOT use an index
|+->
|| |
|| |
|| Harvinder.Singh@met|
|| ratech.com
I knew someone on the list would have it :) thanks Suzy.
actually, I'd do dbms_stats.delete_schema_stats('SYS') this time through to fix
the problem and then change the dbms_stats command to
dbms_stats.gather_schema_stats instead of database
|+--->
||
analyzing SYSTEM is not the same as analyzing SYS, which is what you have done
by using gather_database_stats
run dbms_stats.delete_schema_stats('SYS') to remove the statistics
then either always run the delete after the gather_database_stats or switch to
gather_schema_stats
|+
In some versions of Oracle, you need a SECOND rollback segment created in the
SYSTEM tablespace in order to create another tablespace. It can't hurt to add
it (you can drop it immediately after you get the rollback tablespace created)
but it could be the solution.
Since you didn't include the
that's interesting... since the data dictionary is NOT analyzed, setting
optimizer_mode=choose would force the query against dba_extents to RULE which it
what it is supposed to be doing anyway, Hm. The question now is, what is the
optimizer_mode set to when the problem happens? Did any of the da
it stops me. I refuse to deliberately look like an idiot... I do enough damage
inadvertently
|+--->
|| |
|| |
|| [EMAIL PROTECTED]|
|| ms.osd.mil
HEY! I resent that. ask the last person who worked for me, I think I was a
pretty good manager!
|+--->
|| |
|| |
|| bthater2@nets|
|| cape.net |
||
you people are soo funny. Writing a book takes time, hard work and more
energy than I care to commit to the project.. especially on a subject with which
I have zero experience
|+--->
|| |
|| |
|
won't be me... why don't YOU write one and then talk to me about the joys of
authorship (says the woman going blind looking a page proofs that are totally
messed up)
|+--->
|| |
|| |
|| c
I know Joe Testa blew up his DB with it... it was in the initial release and was
supposedly fixed in the patchsets
|+--->
|| |
|| |
|| optimaldba@ya|
|| hoo.com |
|
nuh uh, for two reasons, the first and foremost being, there already IS one
the second is that I have no plans to write any new books
|+--->
|| |
|| |
|| cgrabowy@fcg.|
||
I've always been a cynic about storage -- too much is never enough.
|+-->
|| |
|| |
|| JayMiller@tdwate|
|| rhouse.com |
||
Right now I'm collecting information.. I don't KNOW what this will be.. other
than a "learning experience" of course.
That which does not kill us makes us strong, right?
rachel, anticipating great strength
|+--->
|| |
||
Yechiel,
have you used their tools? We are trying to decide whether or not to use them,
so if anyone has had recent experience with them, I'd appreciate your thoughts
on ease of use, understandability, quality of the product, etc
Thanks
Rachel
|+--->
||
Dennis,
Forgetting about normalization won't be a problem, I've always been more
practical than "by the book". As for amounts of data being collected, I can see
them wanting data aggregated hourly.
I greatly doubt the tech people will allow adhoc queries, they seem to "do
things right" here. W
Okay, my background is OLTP, but we are looking at a data warehousing project
here
any and all help appreciated! Specifically:
1) does anyone have any experience with a product called "SAS Datawarehousing
Administrator" (or SAS)?
2) how do I go about doing rough estimates of sizing needs,
you really might want to read the manuals -- specifically the reference manual
on dba_data_files
select sum(bytes) from dba_data_files will tell you the total size (in bytes) of
all the datafiles in your database. There will also be space used by the control
files, redo log files, archived log
You could talk to your sales rep about short-term licensing, or you could just
do it and hope they don't discover it. I've found that "temporary servers" have
a tendency to become permanent though.
|+--->
|| |
||
Option 4:
Install 8.1.7
shutdown the database
change ORACLE_HOME etc etc to point to the 8.1.7 ORACLE_HOME
startup the database
You shouldn't have to rebuild the database when moving to a minor release
Rachel
|+--->
|| |
||
47 matches
Mail list logo
