Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> RE: Purely for your amusement

RE: Purely for your amusement

From: Cary Millsap <cary.millsap_at_hotsos.com>
Date: Fri, 28 Feb 2003 08:19:28 -0800
Message-ID: <F001.0055D013.20030228081928@fatcity.com>


I encourage you to keep pushing Oracle Support on the bug. The kernel should emit the right numbers in the first place.

There is a workaround. Push Oracle Support for that, too. Tell them that someone in Oracle knows the answer, because tkprof knows which bits to ignore and swap in order to print the correct answers.

To prove the point, do a SQL trace on something common until you'll see outrageous numbers in the "r=" field. Run tkprof on the trace data, and you'll see a completely different answer in the "Rows" column of the output.

If enough people push on the bug, they'll eventually fix it. (I haven't checked MetaLink; there may already be a patch for it.)

Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com

Upcoming events:
- RMOUG Training Days 2003, Mar 5-6 Denver - Hotsos Clinic 101, Mar 25-27 London

-----Original Message-----
Travis
Sent: Friday, February 28, 2003 9:29 AM
To: Multiple recipients of list ORACLE-L

We are running 64-bit Oracle on HP-UX 11.11. My logical and physical IOs get out of whack after about a day. I can never get a good BHR. They appear to have passed a defined integer value and start looking like this;

SQL> column "logical_reads" FORMAT 999,999,999,999,999,999,999
SQL>     column "phys_reads"    FORMAT 999,999,999,999,999,999,999
SQL>     column "phy_writes"    FORMAT 999,999,999,999,999,999,999
SQL>     select A.value + B.value  "logical_reads",
  2      C.value                   "phys_reads",
  3      D.value                   "phy_writes",
  4      round(100 * ((A.value+B.value)-C.value) / (A.value+B.value))
  5         "Buffer Hit Ratio"
  6      from V$SYSSTAT A, V$SYSSTAT B, V$SYSSTAT C, V$SYSSTAT D
  7      where A.statistic# = 38
  8      AND B.statistic# = 39
  9      AND C.statistic# = 40
 10      AND D.statistic# = 44;

               logical_reads                   phys_reads    phy_writes
Buffer Hit Ratio
---------------------------- ---------------------------- -------------
----------------

  18,446,744,019,800,509,435 18,446,744,057,038,275,898 34,511,604 0

I've never been able to get an answer on this problem. The numbers actually keep increasing (after the 18,446,744 part). Not sure when it actually goes bad. Any ideas?

Good thing we've moved to wait statistic tuning, eh?

> -----Original Message-----
> From: Cary Millsap [mailto:cary.millsap_at_hotsos.com]
> Sent: Thursday, February 27, 2003 3:49 PM
> To: Multiple recipients of list ORACLE-L
> Subject: RE: Purely for your amusement
>
>
> Let me guess: this is a 64-bit implementation of Oracle on
> HP? There are apparently lots of Endian bugs in how the
> Oracle kernel emits statistics on this platform. It's common,
> for example, for 10046 trace files to contain row counts in
> the millions of trillions (or something like that). It's not
> *complete* nonsense, because with some research you can
> figure out which bits contain the relevant information, but
> it's inconvenient.
>
>
> Cary Millsap
> Hotsos Enterprises, Ltd.
> http://www.hotsos.com
>
> Upcoming events:
> - RMOUG Training Days 2003, Mar 5-6 Denver
> - Hotsos Clinic 101, Mar 25-27 London
>
>
> -----Original Message-----
> Patrice J
> Sent: Thursday, February 27, 2003 1:49 PM
> To: Multiple recipients of list ORACLE-L
>
> I have a hard time believing these figures... unbelievable!
>
> Patrice Boivin
> Systems Analyst (Oracle Certified DBA)
>
> Systems Admin & Operations | Admin. et Exploit. des systèmes
> Technology Services | Services technologiques
> Informatics Branch | Direction de l'informatique
> Maritimes Region, DFO | Région des Maritimes, MPO
>
> E-Mail: boivinp_at_mar.dfo-mpo.gc.ca
>
>
>
>
> -----Original Message-----
> Sent: Thursday, February 27, 2003 2:24 PM
> To: Multiple recipients of list ORACLE-L
>
>
> And you thought your job was bad . . .
> (Yes, this IS a production job)
> ----------------------------------------------------------
> 0 SELECT STATEMENT Optimizer=CHOOSE
> (Cost=623898383366807000000000 Card=81143391896031200000000000000
> 1 0 MERGE JOIN (Cost=623898383366807000000000
> Card=81143391896031200000000000000
> Bytes=535546386513806
> 2 1 SORT (JOIN)
> 3 2 MERGE JOIN (CARTESIAN)
> (Cost=1323033411994580000000
> Card=518075818229206000000000 Bytes=321725
> 4 3 MERGE JOIN (CARTESIAN)
> (Cost=190827283721623000 Card=28145586908741600000 Bytes=150578889961
> 5 4 MERGE JOIN (OUTER)
> (Cost=301983192525 Card=40967578733025
> Bytes=21385076098639100)
> 6 5 MERGE JOIN (Cost=301983182527
> Card=40967578733025 Bytes=19377664740720800)
> 7 6 SORT (JOIN)
> 8 7 MERGE JOIN (CARTESIAN)
> (Cost=2532679282 Card=1590809983140
> Bytes=322934426577420)
> 9 8 VIEW OF 'WO_CHARG_VIEW'
> (Cost=725242 Card=2615655 Bytes=347882115)
> 10 9 SORT (GROUP BY)
> 11 10 TABLE ACCESS (BY
> ROWID) OF 'WO_CHARG' (Cost=690469 Card=2615655
> Bytes=68007030
> 12 11 INDEX (FULL SCAN) OF
> 'PK_WO_CHARG' (UNIQUE)
> 13 8 SORT (JOIN)
> 14 13 TABLE ACCESS (FULL) OF
> 'PUB' (Cost=968 Card=608188 Bytes=42573160)
> 15 6 SORT (JOIN)
> 16 15 TABLE ACCESS (FULL) OF 'WO'
> (Cost=190903 Card=15662455 Bytes=4228862850)
> 17 5 SORT (JOIN)
> 18 17 TABLE ACCESS (FULL) OF 'NAD'
> (Cost=2792 Card=500401 Bytes=24519649)
> 19 4 SORT (JOIN)
> 20 19 TABLE ACCESS (FULL) OF 'CNR'
> 21 3 SORT (JOIN)
> 22 21 TABLE ACCESS (FULL) OF 'WOE'
> (Cost=47 Card=18407 Bytes=1583002)
> 23 1 SORT (JOIN)
> 24 23 TABLE ACCESS (FULL) OF 'WO'
> (Cost=190903 Card=15662455 Bytes=610835745)
>
>
>
> __________________________________________________
> Do you Yahoo!?
> Yahoo! Tax Center - forms, calculators, tips, more
> http://taxes.yahoo.com/
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.net
> --
> Author: Barbara Baker
> INET: barbarabbaker_at_yahoo.com
>
> Fat City Network Services -- 858-538-5051 http://www.fatcity.com
> San Diego, California -- Mailing list and web hosting services
> ---------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: ListGuru_at_fatcity.com (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.net
> --
> Author: Boivin, Patrice J
> INET: BoivinP_at_mar.dfo-mpo.gc.ca
>
> Fat City Network Services -- 858-538-5051 http://www.fatcity.com
> San Diego, California -- Mailing list and web hosting services
> ---------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: ListGuru_at_fatcity.com (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.net
> --
> Author: Cary Millsap
> INET: cary.millsap_at_hotsos.com
>
> Fat City Network Services -- 858-538-5051 http://www.fatcity.com
> San Diego, California -- Mailing list and web hosting services
> ---------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: ListGuru_at_fatcity.com (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.net
-- 
Author: Glenn Travis
  INET: Glenn.Travis_at_sas.com

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (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.net
-- 
Author: Cary Millsap
  INET: cary.millsap_at_hotsos.com

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (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).
Received on Fri Feb 28 2003 - 10:19:28 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US