You run oradebug from sqlplus.
SQL oradebug help
HELP [command] Describe one or all commands
SETMYPID Debug current process
SETOSPID ospid Set OS pid of process to debug
SETORAPID orapid ['force']Set
Samar,
This might be due to failure in resolving hostname to ip address. You can try adding
the servers ip address and hostname to the hosts file on client machine. This file
will be in c:\windows\system32\drivers\etc\hosts. Add the entry of servers ip followed
by server's hostname.
Cheers,
It probably means that you will be restoring your
database from the most recent cold backup, as you
cannot restore a datafile when your database is not in
archivelog mode.
Jared
On Wednesday 25 December 2002 21:59, shuan.tay\(PCI???\) wrote:
Dear all DBAs,
What is this error suppose to mean?
Hi, dba friends:
I trussed some process execution in my system, and get result like:
oracle@main-db1$truss -c -p 3995
^Csyscall seconds calls errors
read .47 6963
write .48 6555
lseek .01 402
kill .00 2
times .61 25964
semsys .01 44
sigprocmask .00 92
context .00 23
exp userid=system/manager file=schema.dmp rows=n owner=scott
vi schema.dmp
really.
Jared
On Wednesday 25 December 2002 09:53, Andrey Bronfin wrote:
Dear gurus !
I'm sure many of you have scripts to recreate an Oracle schema including
objects (i am interested in tables, indexes , comments,
Shuan,
Can't
tell you what exactly to do, we will need more detailed information than just
two ORA codes.
Your
database can't read one of your data file and that could be because of many
things.
Your
database is in NOARCHIVELOG mode, and this means that you have no protection
against
To disconnect the session use this :
Alter system disconnect session 'sid,serial#'
post_transaction;
--- Richard Ji [EMAIL PROTECTED] wrote:
You run oradebug from sqlplus.
SQL oradebug help
HELP [command] Describe
one or all commands
SETMYPID
This is probably due to wrong tcp/ip address in tnsnames.ora
or the server is on a separate network from the clients.
Try to ping with the address in the tnsnames.ora.
If it does not work check the tcp/ip address or call the network
administrator.
If it works try tnsping with the connection
This looks like a v9 trace file, which means
there may be new issues involved that I
haven't come across yet. For example, 9.2
introduces a mode 2 TM lock on pk/fk activity
for some reason that I haven't worked out,
so this may be a side-effect.
However, (assuming no big changes from v8)
this is
Could you clarify what you mean by 'two sets of slaves' ?
Does this mean you got 24 slaves ? Do you get any clue
about how these may be related by looking at v$px_sesstat ?
This may be related in some way to the fact that
when you do a direct insert on a table, Oracle
still has to do ordinary
Hi people,
I have this table X, whose
structure is described below:
Name
Null? Type--
NAME
VARCHAR2(20)EMP
NUMBER(5)BOSS
NUMBER(5)
The table is having the following
data:
NAME
SQL SELECT * FROM
emp;
NAME
EMP BOSS-- --
--SAMAR
10
20ASHOK
20
30ASHWINI
30
40MONIKA
11
21RASHI
21
31SMRITI
12
22SUMEET
22 32
7 rows selected.
SQL SELECT *
2 FROM emp 3
WHERE emp NOT IN ( SELECT emp
4
FROM emp
5
START WITH emp = 10 CONNECT BY
Hi, list friends:
I am reading oracle 9.2 performance planning manual, and see it says:
Today's date. SELECT SYSDATE FROM DUAL can account for over 60% of the
workload on a database.
{page 33 of that manual}
How to understand that words? And does it mean to get the
Does your app connect to the database using the listener or
directly (BEQ) without specifying a connect string?
It should connect using the listener and a connect string.
Regards,
Waleed
Waleed
Yes, it's a web application and uses a listener. The listener entry has
not been
Title: Message
From: Bob Metelsky [mailto:[EMAIL PROTECTED]]
I have a developers stored procedure making repeaded calls (logons/request for data and logoff)
Each call takes only seconds but it seems like PMON dosnt close the
process in a timely manner, leaving what look to be ghost
What you can do is like this...
1. At the startup of Application Server, do a SELECT sysdate FROM dual, store
the sysdate, and also get the system date of the application server, that way
you will know the difference between the App Server date and the DB server
date.
2. Everytime you want to use
Do you know if the connections (sessions) created for the web app are shared
or dedicated?
If it's dedicated, did you try to connect using sqlplus from that win2k
server and other hosts?
Trying to isolate the problem?
Regards,
Waleed
-Original Message-
Sent: Thursday, December 26,
Hi,
It would account for 100% of a database's workload or it could account
for 0.01% of it. Or maybe 42%, which is another good number.
This is of course nonsense taken as a statement by itself. Check out
Connor McDonald's article on Dual (and probably many others have written
about it,
You could use NTP to synchronouse the time on your app server and db server.
-Original Message-
Sent: Thursday, December 26, 2002 10:59 AM
To: Multiple recipients of list ORACLE-L
What you can do is like this...
1. At the startup of Application Server, do a SELECT sysdate FROM dual,
Title: RE: Cache on sysdate? --From 9i performance planning manual
Can someone please explain me why you have to use
select sysdate
from dual
/
when
my_date_Var := sysdate;
just works fine? Maybe I am clueless ... but I can take an explanation
Raj
I see you've started two threads on SYSDATE at once.
A couple of thoughts (rhetorical questions, not criticisms)
a) Does your benchmark actually mean anything to YOU ?
Are you really doing lots of
select sysdate from dual
in your application ?
If so why do you want
Raj:
Both are same. It is internally translated as a select call to dual.
KG
--- Jamadagni, Rajendra [EMAIL PROTECTED] wrote:
Can someone please explain me why you have to use
select sysdate
from dual
/
when
my_date_Var := sysdate;
just works fine? Maybe I am clueless ...
In PL/SQL, every statement which references SYSDATE or USER
or USERENV or similar functions (such as your second
suggestion) is automatically translated into an individual
SELECT xxx FROM DUAL statement, behind the scenes.
You don't have a choice. Call it a weakness of PL/SQL...
Can someone
Title: RE: Cache on sysdate? --From 9i performance planning manual
I thought the very same Tim ...
But ...
oraclei@elara-ABC1 sys
SQL*Plus: Release 9.2.0.2.0 - Production on Thu Dec 26 13:14:58 2002
Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved.
Connected.
SQL
Hi
Wondering one of database is showing single-task message high wait.
Let me know what to do to fix this pl?
thx
-Seema
_
Add photos to your e-mail with MSN 8. Get 3 months FREE*.
Actually this may not be true any more starting from Oracle 8.1.7.
Many of the functions started to have a C language interface in 8.1.7.
Check stdbody.sql to find which function has a C prgama.
Regards,
Waleed
-Original Message-
Sent: Thursday, December 26, 2002 1:24 PM
To: Multiple
Raj:
SQL trace or the event 10046 is NOT the right to trace the PL/SQL
executions. You need to either use the dbms package dbms_trace or
event 10938 to populate the SYS table PLSQL_TRACE_EVENTs and select
from that table.
SQL_TRACE is just SQL Trace. NOT PL/SQL Trace..
Best Regards,
K
jared,
Thanks for explanation. Still not convinced because of following two reasons
1) Same scenario can happen with explicit privileges as well. User A grants
ALL privileges on MY_TABLE to B without GRANT OPTION. Now B can create a
stored procedure to do DML on MY_TABLE and grant execute
That's what I thought too. But the results of testing are somewhat different.
Maybe it's evaluated within the PL/SQL engine and does not require a context
switch to the SQL engine.
Platform : Sun Solaris 2.6
Oracle : 8.1.7.4 (32 bit)
DEV:43#14739-23049@dual1
DEV:43#14739-23049declare
2
Following is from stdbody.sql. Now select sysdate from dual is used only
when pessdt returns NULL.
-- Bug 1287775: back to calling ICD.
-- Special: if the ICD returns NULL, that means we should do the old
-- 'SELECT SYSDATE FROM DUAL;' thing. This allows us to do the SELECT
from
--
Title: Rebuilding Indexes...
Anyone have any useful scripts for doing this?
TIA,
Rich
Johnathan,
I saw 24 slaves, two sets of 12. I got this info from v$px_session:
select
decode(px.qcinst_id,NULL,s.username,
' - '||lower(substr(s.program,length(s.program)-4,4) ) ) Username,
decode(px.qcinst_id,NULL, 'QC', '(Slave)') QC/Slave ,
to_char( px.server_set) Slave Set,
Raj,
Good point, they seem to have included some optimizations
for SYSDATE, but not for other functions. Your test
reproduced on 8174 and 9201 for me as well as your 9202, but
this test shows that the optimization only applys to SYSDATE
on 8174 and 9201...
- begin SQL*Plus
Hi,
Has any enabled cursor_Space_for_time=true in your databases,IF yes have you
faced any issues,Cause we are facing lots of contention on latch free for
library cache issue and one of the main culprit is that SYSDATE FROM
DUAL,Just wondering if we enable this parameter will it by some chance be
Convinced or not, that's the reason, fallible as it may be.
On Thursday 26 December 2002 11:18, Shaleen wrote:
jared,
Thanks for explanation. Still not convinced because of following two
reasons
1) Same scenario can happen with explicit privileges as well. User A grants
ALL privileges on
Used to, no longer ;-) Change happened in 8.1.7.2 or so.
On Thursday 26 December 2002 10:23, you wrote:
Raj:
Both are same. It is internally translated as a select call to dual.
KG
--- Jamadagni, Rajendra [EMAIL PROTECTED] wrote:
Can someone please explain me why you have to use
Waleed, thanks for the reply here is what I have
I have 26 sessions and 34 process, only one of which (my sqlplus
session) that seems to indicate shared
The rest seem to be remanents of the stored procedure.
At the point of running this test, it was only myself monitoring and the
developer
That's because doing dt := sysdate is more optimized, it's still select
sysdate into dt from dual but more optimized. select from dual doing 4 db
block gets in 8i and 2 db block gets in 9i, but you can tune it to cut it
down.
Richard Ji
-Original Message-
Sent: Thursday, December 26,
Friends,
I need help with OID as I am new to it. How do I
delete a subtree (recursive delete) from DIT through
Oracle Directory Manager? I am able to delete it at
the leaf level only.
TIA,
Ravi.
__
Do you Yahoo!?
Yahoo! Mail Plus - Powerful.
Rich,
select 'alter index '||owner||'.'||index_name||' rebuild;' from
sys.dba_indexes where owner = 'PUTOWNERNAMEHERE';
is a good start for normal non-bitmap indexes, enhance it to do bitmap,
etc. :)
joe
Richard Huntley wrote:
Anyone have any useful scripts for doing this?
TIA,
Rich
--
List,
Can anybody tell me what is happening here. I am constantly getting the
PLS-00201 error when I try to compile. It is looking inside the package at
a procedure and saying the identifier must be declared. I have gone over
metalink docs and notes. I first compile the package spec and then
This may be more convincing:
http://metalink.oracle.com/metalink/plsql/ml2_documents.showDocument?p_datab
ase_id=NOTp_id=1031418.6
regards,
Waleed
-Original Message-
Sent: Thursday, December 26, 2002 2:19 PM
To: Multiple recipients of list ORACLE-L
--Resolved
jared,
Thanks for
Title: Rebuilding Indexes...
Here is the index
rebuilding part of a ksh script I use. Prior to this, I some variable
setting, check the existence and permissions on some files, check the existence
of tablespaces, check that the script is not already running, etc. In this
script, all the
Hi All,
Need help with OID. How do I delete a subtree
(recursive delete) from DIT through Oracle Directory
Manager? I am able to delete it at the leaf level
only.
TIA,
Ravi.
__
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
Thanks Waleed. For the first time I have understood the reason behind this.
Which is, Larry shirked!!
The reason why roles are disabled in stored procedures is not due to the
roles themselves, but has to do with the dependency model. If roles were
enabled in stored procedures, Oracle would
Title: Rebuilding Indexes...
Script "gen_rebuild_idx.sql" on www.evdbt.com/tools.htmgenerates
SQL to rebuildUNUSABLE indexes, index partitions/subpartitions. You
can modify the WHERE clauses as you prefer...
- Original Message -
From:
Richard Huntley
To: Multiple
Title: RE: Cache on sysdate? --From 9i performance planning manual
Richard,
If you look in the tracefile ... there is no select from dual. It used to be like that but things changed (as Anjo mentions maybe be around 8iR3). The sysdate call is now a C function call.
Raj
Does the compiling ID have a direct grant on the package P_MESSAGES
If you have access to it via a grant on a role and not a direct grant, then
you will get error messages like this.Check your authorities and see if
you have a direct grant.
-Original Message-
Sent: Thursday,
Title: RE: compile errors
You probably do not have enough privileges to access HSD_TYPES (a package maybe?) If it is a package, you need a synonym and exec privilege. If it is a table, you still need a synonym and a direct grant to select from it.
Raj
I don't agree that anyone shirked. Roles are, by design, changeable
within a session. The SET ROLE command is not DDL, altering the metadata of
the database. Instead, it is only altering already-granted permissions to
used subsequently by the session. So, why should permanent objects (such
as
Yes, I read that off of metalink. I granted execute any procedure to the
user. And still got the same error. Do I still need a direct grant off of
that package?
thanks.
-Original Message-
Sent: Thursday, December 26, 2002 4:14 PM
To: Multiple recipients of list ORACLE-L
Does the
I think you do.
I would suggest that you go ahead and try it with the grant. Should be
very easy to test out that way.
-Original Message-
Sent: Thursday, December 26, 2002 4:40 PM
To: Multiple recipients of list ORACLE-L
Yes, I read that off of metalink. I granted execute any
I thought I had these files stopped but apparently not.
Is there somone out there who can tell me how to stop the Audit files from
appearing in the audit_file_dest ??? I thought if I set the audit_trail to
false then these would stop as well ... Apparently not.
Anyone have an idea how
Hmm. Makes sense. Thanks Tim.
- Original Message -
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Sent: Thursday, December 26, 2002 2:34 PM
I don't agree that anyone shirked. Roles are, by design, changeable
within a session. The SET ROLE command is not DDL, altering the
Yes it is 9013. This is not an array based update. As per the trace file
same statement is being executed by both sessions. I can directly send you
the trace file if there is a need.
There are triggers on the tables, I'll look into parent table activity. But
there are indexes on all foreign keys
Listers,
8.1.7.4 64 Bit Solaris
Does row migration utilize DB File Sequential Reads on the table? Off the
top of my head I would expect so, but I've never tested something like that
before.
Trying to figure out if row migration is the cause of the slowdown in a
package (well, it's probably
I remember seeing another reason (I think from Tom Kyte). Roles tend to be
used for a larger number of users than system privileges. If procedures
could be created using rights from roles, every time a role was changed
(grant a new priv, disable, ...), the procedure would need to be invalidated
Hi Listers,
I'm new on Oracle Database 9i after I migrated from Oracle 8i.
I try to use backup facility from Oracle 9i and I already follow the
instructions how to activate the automatic backup but when I see the status
on history I get an error comment 'Failed'. I've try again and again but the
They are generated when you connect internal, connect as sysdba, or connect
as sysoper. You can't turn them off.
In 9i, AUDIT_SYS_OPERATIONS=TRUE (default FALSE) will also create .aud
files in the AUDIT_FILE_DEST, but obviously that can be turned off. In all
versions of Oracle,
Just to clarify, server_set tells what set of slaves each process
belongs to.
Johnathan,
I saw 24 slaves, two sets of 12. I got this info from v$px_session:
select
decode(px.qcinst_id,NULL,s.username,
' - '||lower(substr(s.program,length(s.program)-4,4) ) ) Username,
IIRC, these files are generated whenever someone logs in as sysdba or
internal. I don't know of any way to stop them.
--- Kevin Lange [EMAIL PROTECTED] wrote:
I thought I had these files stopped but apparently not.
Is there somone out there who can tell me how to stop the Audit files
Does the owner of the package have *direct* privs (not through a role)
on the object in question?
List,
Can anybody tell me what is happening here. I am constantly getting
the
PLS-00201 error when I try to compile. It is looking inside the
package at
a procedure and saying the
Title: RE: Cache on sysdate? --From 9i performance planning manual
The Oracle kernel tallies time to this event
when an application that is linked single-task spends time between database
calls. Its usually time spent awaiting either user input or application
processing.
Cary Millsap
Sony,
What is an 'automatic' backup?
Is this something supplied by that 'Oracle Enterprise Manager' thingy?
Jared
On Thursday 26 December 2002 17:23, Sony kristanto wrote:
Hi Listers,
I'm new on Oracle Database 9i after I migrated from Oracle 8i.
I try to use backup facility from Oracle 9i
Though I have published a script for determining indexes that
need to be rebuilt, and then rebuilding them, I have to say that
this is almost never necessary.
Why are you rebuilding indexes? About the only reason for ever
doing so is that the BLEVEL = 5.
goto asktom.oracle.com, and do a
This is in conflict with what a few Oracle folks have said.
Interesting. I guess you'd have to ask the oracle security architect
to get the full story.
Jared
On Thursday 26 December 2002 12:39, Khedr, Waleed wrote:
This may be more convincing:
Not familiar with this. Is this something in Oracle Enterprise Manager,
perhaps the backup manager forms?
- Original Message -
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Sent: Thursday, December 26, 2002 6:23 PM
Hi Listers,
I'm new on Oracle Database 9i after I
What do you get when run this on the server hosting Oracle:
lsnrctl services
Waleed
-Original Message-
To: Multiple recipients of list ORACLE-L
Sent: 12/26/02 3:49 PM
Waleed, thanks for the reply here is what I have
I have 26 sessions and 34 process, only one of which (my sqlplus
That would be Mary-Ann Davidson. I'll ask her.
Mogens
Jared Still wrote:
This is in conflict with what a few Oracle folks have said.
Interesting. I guess you'd have to ask the oracle security architect
to get the full story.
Jared
On Thursday 26 December 2002 12:39, Khedr, Waleed wrote:
It is 8.1.7.4, it doesn't do the select sysdate from dual anymore for
dt:=sysdate;
Anjo.
On Thursday 26 December 2002 12:59, you wrote:
That's because doing dt := sysdate is more optimized, it's still select
sysdate into dt from dual but more optimized. select from dual doing 4 db
block
They were put there in 7.1 in order to comply with some security standard.
And their purpose is exactly to prevent a dba from logging in without being
monitored. It's in the 7.1 new features manual, as far as I remember. That's
also the version where it was suddenly not possible for the poor
71 matches
Mail list logo
