Re: Strange 4031 issue.
- From: "Jonathan Lewis" <jonathan@xxxxxxxxxxxxxxxxxx>
- To: <oracle-l@xxxxxxxxxxxxx>
- Date: Thu, 2 Nov 2006 21:52:15 -0000
The amount of memory free is nearly irrelevant if you want
to allocate a single large chunk. It is common for the heap
to end up as a very large collection of very small pieces,
and if a lot of those pieces are pinned in some way, it
may not be possible to free enough adjacent pieces to
produce a single chunk large enough.
27K is about the size of the 'session parameters' on some
platforms, so if you keep seeing the same size chunk
causing a failure then perhaps that's what it is. You could
try increasing your reserved_shared_pool so that it is
large enough to hold sessions * 27KB plus a few dozen
MB.
Cross-check you timestamps on the traces with the
large amounts of free memory. A hunt for free memory
can clear a lot of stuff from the shared pool, and it
can take some time to reload. Do you any of the
reports of large percentages of free memory show up
a few minutes after a trace file is dumped ?
Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
Author: Cost Based Oracle: Fundamentals
http://www.jlcomp.demon.co.uk/cbo_book/ind_book.html
The Co-operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html
----- Original Message -----
From: "Wolfson Larry - lwolfs" <lawrence.wolfson@xxxxxxxxxx>
To: <oracle-l@xxxxxxxxxxxxx>
Sent: Thursday, November 02, 2006 9:36 PM
Subject: Strange 4031 issue.
9.2.0.6
Getting multiple 4031s on SQLLDR processes.
Shows up as
-rw-r--r-- 1 dba 14755 Oct 31 14:08 p_j013_834.trc
-rw-r--r-- 1 dba 14880 Oct 31 14:06 p_j004_29359.trc
-rw-r--r-- 1 dba 14887 Oct 31 13:58 p_j008_23565.trc
-rw-r--r-- 1 dba 14763 Oct 31 13:56 p_j000_22346.trc
-rw-r--r-- 1 dba 14764 Oct 31 13:53 p_j011_19931.trc
-rw-r--r-- 1 dba 14766 Oct 31 13:51 p_j005_18595.trc
like it came from dba_jobs.
We've been monitoring SP every 15 minutes and it looks like there's
generally lots of memory available.
DB TIME Shared Pool Size Free Bytes Percent Free
---- -------------------- ---------------- ---------------- ------------
XXXX 25-OCT-2006 07:15:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 07:30:00 167,772,160 8,299,488 4.9468803
XXXX 25-OCT-2006 07:30:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 07:45:00 167,772,160 14,386,336 8.5749244
XXXX 25-OCT-2006 07:45:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 08:00:00 167,772,160 12,982,504 7.7381753
XXXX 25-OCT-2006 08:00:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 08:15:00 167,772,160 25,204,720 15.0231838
XXXX 25-OCT-2006 08:15:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 08:30:00 167,772,160 8,200,736 4.8880195
XXXX 25-OCT-2006 08:30:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 08:45:00 167,772,160 13,062,288 7.7857303
XXXX 25-OCT-2006 08:45:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 09:00:00 167,772,160 10,246,144 6.1071777
XXXX 25-OCT-2006 09:00:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 09:15:00 167,772,160 9,018,072 5.3751897
XXXX 25-OCT-2006 09:15:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 09:30:00 167,772,160 28,939,400 17.2492266
XXXX 25-OCT-2006 09:30:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 09:45:00 167,772,160 21,258,496 12.671051
XXXX 25-OCT-2006 09:45:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 10:00:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 10:00:00 167,772,160 52,412,056 31.2400198
XXXX 25-OCT-2006 10:15:00 167,772,160 28,882,048 17.2150421
XXXX 25-OCT-2006 10:15:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 10:30:00 167,772,160 15,986,504 9.5286989
XXXX 25-OCT-2006 10:30:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 10:45:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 10:45:00 167,772,160 51,772,144 30.8586025
XXXX 25-OCT-2006 11:00:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 11:00:00 167,772,160 34,621,608 20.6360865
XXXX 25-OCT-2006 11:15:00 167,772,160 8,650,376 5.1560258
??
XXXX 25-OCT-2006 11:15:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 11:30:00 167,772,160 9,714,240 5.7901382
??
XXXX 25-OCT-2006 11:30:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 11:45:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 11:45:00 167,772,160 57,708,392 34.3968821
XXXX 25-OCT-2006 12:00:00 167,772,160 28,971,008 17.2680664
XXXX 25-OCT-2006 12:00:00 167,772,160 37,005,832 22.0571947
Vendor keeps saying nothing special happens at times we experience
4031s or dips in available memory
I summarize what's in SP and I don't see anything really big
TYPE NAME
MEMORY KEPT
---------------------------- ----------------------------------------
------------ ---
JAVA CLASS sun/io/CharacterEncoding
109,204 YES
PACKAGE BODY S2P_INV_CTL
109,978 NO
PACKAGE S2P_C
116,318 NO
PACKAGE BODY S2P_ORDER_CTL
116,835 NO
PACKAGE S2P_RF_UTIL
117,385 NO
JAVA CLASS java/util/TimeZoneData
121,338 YES
PACKAGE BODY S2P_SKUXLOC
122,842 NO
PACKAGE BODY S2P_TASKHDR
123,554 NO
PACKAGE BODY S2P_RF_PICK
126,569 NO
CURSOR SELECT * FROM ORDERS WHERE ORDERKEY =
127,326 NO
:b1
PACKAGE BODY S2P_RF_UTIL
127,705 NO
PACKAGE BODY S2P_TASKDTL
128,138 NO
JAVA CLASS /1adcad9a_OracleStatement
132,077 YES
PACKAGE BODY S2P_LOC
137,626 NO
PACKAGE BODY S2P_ORDERS
150,408 NO
PACKAGE BODY S2P_TASK
157,804 NO
CURSOR SELECT * FROM TASKHDR WHERE TASKHDR.D
158,892 NO
EVICE = :b1 AND TASKHDR.STATE = :b2 AN
D TASKHDR.STS = :b3 ORDER BY TASKHDR.PIC
KEDUPSEQUENCE DESC
PACKAGE BODY S2P_STORAGE
161,386 NO
PACKAGE S2P_RF_TASK
163,922 NO
PACKAGE BODY S2P_RF_AUTO
165,138 NO
PACKAGE BODY S2P_TASKTRN
166,442 NO
PACKAGE BODY S2P_CONT_INV_CTL
184,500 NO
PACKAGE BODY S2P_INVENTORY
188,598 NO
PACKAGE BODY S2P_RF_TASK
251,338 NO
CURSOR table_1_0_141_0_0_0
353,117 NO
PACKAGE BODY S2P_TASKTRN_CTL
359,722 NO
PACKAGE STANDARD
438,556 NO
Then we get 4031s on sqlplus jobs, but connected users don't see this.
OOPS! Actually we did see ONE 4031 from connected user (out of 330) at
this time.
And same size every time:
ORA-04031: unable to allocate 27512 bytes of shared memory
We're tracking all errors:
create or replace trigger log_errors_trig
after servererror on database
declare
var_user varchar2(30);
var_osuser varchar2(30);
var_machine varchar2(64);
var_process varchar2(09);
var_program varchar2(48);
var_sqltext varchar2(4000);
sql_text ora_name_list_t;
n number;
begin
Saw Karen Morton suggested log_buffer might be too small. It's 512K
here, but haven't increased it.
Did see someone complain about SP stress when running sqlldr in 8.1.5
Is there a parm that can cause problem?
Saw someting about perfstat reporting miscellanrous growth over time,
but don't have perfstat installed here (yet).
I'll see if I can get script from vendor.
Any ideas?
Thanks
Larry
***************************************************************************
The information contained in this communication is confidential, is
intended only for the use of the recipient named above, and may be legally
privileged.
If the reader of this message is not the intended recipient, you are
hereby notified that any dissemination, distribution or copying of this
communication is strictly prohibited.
If you have received this communication in error, please resend this
communication to the sender and delete the original message or any copy
of it from your computer system.
Thank You.
****************************************************************************
--
http://www.freelists.org/webpage/oracle-l
--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.409 / Virus Database: 268.13.22/512 - Release Date: 01/11/2006
--
http://www.freelists.org/webpage/oracle-l
- References:
- Strange 4031 issue.
- From: Wolfson Larry - lwolfs
Other related posts:
- » Strange 4031 issue.
- » RE: Strange 4031 issue.
- » Re: Strange 4031 issue.
- » RE: Strange 4031 issue.
- Strange 4031 issue.
- From: Wolfson Larry - lwolfs