Thanks for the replies (K,Jonathan,Anjo). I'm getting
deadlocks to occur but they're not producing the
deadlock graphs I was expecting to see.
I'm looking to reproduce the scenario that will
generate the following kind of deadlock graph--where
it looks like a self-deadlock:
---------Blocker(s)-------- ---
------Waiter(s)---------
Resource Name process session holds waits
process session holds waits
TX-00050032-00002143 22 23
X 22 23 S
session 23: DID 0001-0016-000017E7 session 23:
DID 0001-0016-000017E7
Rows waited on:
Session 23: no row
But, the deadlock graph that my testing is generating
looks like:
---------Blocker(s)-------- ----
-----Waiter(s)---------
Resource Name process session holds waits
process session holds waits
TX-00030052-00001fb9 11 11
X 12 15 X
TX-00040058-000023ef 12 15
X 11 11 X
session 11: DID 0001-000B-00000002 session 15: DID
0001-000C-00000002
session 15: DID 0001-000C-00000002 session 11: DID
0001-000B-00000002
Rows waited on:
Session 15: obj - rowid = 0000153E - AAABU+AAFAAAALIAAA
Session 11: obj - rowid = 0000153E - AAABU+AAFAAAAACAAA
It is my understanding that the deadlock graph I am
trying to reproduce is caused by ITL shortage
deadlocks. I have verified, via
dbms_rowid.rowid_block_number(), that I am updating
rows in different blocks.
Thanks again.
-w
>It gets more complicated under Oracle 9 because
>initrans has a minimum value of 2 on tables; but
>under oracle 8, the following should work:
>
>create table t1
>initrans 1
>maxtrans 1
>
>insert a few dozen rows into t1
>so that there are rows in at least
>two blocks, and commit.
>
>session 1
> update a row in block 1
>session 2
> update a row in block 2
>session 1
> update a row in block 2 -- will wait on a TX/4
>session 2
> update a row in block 1 -- will wait on a TX/4
>
>After ca, three seconds session 1 will report ORA-
00060.
>
>
>In Oracle 9, you may get lucky with just two blocks
>of data and three concurrent sessions, but for
>consistency you need to have three blocks of data and
>rotate through the sessions updating one row in each
>block from each session in turn.
>
>
>
>Jonathan Lewis
>http://www.jlcomp.demon.co.uk
>
|Walter K wrote:
|
|> Can someone post an example of how to trigger a
|> deadlock (ORA-0060) due to ITL shortage? This is for
|> informational/fact-finding purposes.
|>
|> I've created a test table with MAXTRANS=1 and can
|> cause the enqueue waits between two sessions
|> contending for the same block but I can't seem to
|> cause a deadlock to occur.
|>
|> Thanks.
|> -w
|>
--
Please see the official ORACLE-L FAQ:
http://www.orafaq.com
--
Author: Jonathan Lewis
INET: [EMAIL PROTECTED]
Fat City Network Services -- (858) 538-5051 FAX:
(858) 538-5051
San Diego, California -- Public Internet
access / Mailing Lists
-------------------------------------------------------
-------------
To REMOVE yourself from this mailing list, send an E-
Mail message
to: [EMAIL PROTECTED] (note EXACT spelling
of 'ListGuru') and in
the message BODY, include a line containing: UNSUB
ORACLE-L
(or the name of mailing list you want to be removed
from). You may
also send the HELP command for other information (like
subscribing).
--
Please see the official ORACLE-L FAQ: http://www.orafaq.com
--
Author: Walter K
INET: [EMAIL PROTECTED]
Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
San Diego, California -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from). You may
also send the HELP command for other information (like subscribing).
