Re: 8.1.7 cache buffer chains contention

Sorry about the dump file size - I wasn't expecting it to do full leaf dumps. That must have changed since 8.1.5.

The interesting thing is that I think your dump is showing your root block at

    0x4803e803
which is block number 3e803 (256003) of
a file whose number I can't work out but which is greater than 64 (possibly 288).

This doesn't tally with your earlier hypothesis that the root block was at 15603 - so it would be interesting to find out what block 15603 really is,

Do a block dump - using the file number block number.

alter system dump datafile XXX block 15603;

and see what it looks like.

--
Jonathan Lewis
http://www.jlcomp.demon.co.uk

Host to The Co-Operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html

Author of:
Practical Oracle 8i: Building Efficient Databases

Screen saver or Life saver: http://www.ud.com
Use spare CPU to assist in cancer research.

Scott Gamble wrote in message ...


>

>

>"Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> wrote in

><999014469.27908.1.nnrp-07.9e984b29_at_news.demon.co.uk>:

>

>>That's a bit odd, I would have expected to

>>see at least a couple of CR copies of the block.

>>

>>Is the index in the DEFAULT pool, or one or

>>other of the KEEP / RECYCLE pools ?

>

>In the default pool.

>

>

>>

>>Has your value  for _db_block_max_cr_dba

>>been adjusted.

>

>No it hasnt, the default is 6

>

>>

>>How do you feel about doing:

>>

>>    alter session set events

>>    'immediate trace name treedump level NNN';

>>

>>where NNN is the object_id of the index in question.

>>This will dump a line of text for every block in the

>>index, and prove that the block IS the root block

>>(it will be the first one listed in the dumpt) as well

>>as telling us something about the degree of packing

>>of the block and the rest of the index.

>>

>

>Results of dump.  Not sure how much you wanted to see..

>

>Machine:        alpha

>Instance name: P013

>Redo thread mounted by this instance: 1

>Oracle process number: 528

>Unix process pid: 22118, image: oracle_at_prd0db05 (TNS V1-V3)

>

>*** 2001-08-28 13:36:31.358

>*** SESSION ID:(547.893) 2001-08-28 13:36:31.251

>----- begin tree dump

>branch: 0x4803e803 1208215555 (0: nrow: 61, level: 3)

>   branch: 0x4803fe24 1208221220 (-1: nrow: 44, level: 2)

>      branch: 0x4803e876 1208215670 (-1: nrow: 67, level: 1)

>         leaf: 0x4803e804 1208215556 (-1: nrow: 92 rrow: 92)

>Leaf block dump

>===============

>header address 18186911836=0x43c06405c

>kdxcolev 0

>kdxcolok 0

>kdxcoopc 0x80: opcode=0: iot flags=--- is converted=Y

>kdxconco 6

>kdxcosdc 0

>kdxconro 92

>kdxcofbo 220=0xdc

>kdxcofeo 292=0x124

>kdxcoavs 72

>kdxlespl 0

>kdxlende 0

>kdxlenxt 1208215557=0x4803e805

>kdxleprv 0=0x0

>kdxledsz 0

>kdxlebksz 3920

>row#0[3872] flag: -----, lock: 0

>col 0; len 15; (15):  2b 2b 55 4d 3a 43 41 4e 32 32 39 32 30 30 43

>col 1; len 3; (3):  31 30 30

>col 2; len 7; (7):  77 64 01 01 01 01 01

>col 3; len 7; (7):  78 c7 0c 1f 01 01 01

>col 4; len 2; (2):  c1 60

>col 5; len 6; (6):  34 01 83 bb 00 07

>row#1[3824] flag: -----, lock: 0

>col 0; len 15; (15):  2b 2b 55 4d 3a 4d 41 54 32 39 36 38 36 31 42

>col 1; len 3; (3):  31 30 30

>col 2; len 7; (7):  77 64 01 01 01 01 01

>

>

>