Re: consistent read and LIO question

From: lsllcm <lsllcm_at_gmail.com>
Date: Sat, 18 Apr 2009 21:47:30 -0700 (PDT)
Message-ID: <e60a0c65-6d55-4099-9ebe-869b234a9470_at_s38g2000prg.googlegroups.com>

On Apr 18, 11:58 am, sriv..._at_hotmail.com wrote:
> Number of consistent reads does not always equal what is reported in
> FETCH
>
> You can use the following events for getting the required info
>
> 10200 "consistent read buffer status"
> 10201 "consistent read undo application"
> 10202 "consistent read block header"
> 10226 "trace CR applications of undo for data operations"
>
> regards
> srivenu

Thanks for your comments

I use 10200 event and get below result

• 2009-04-19 11:47:38.945 ktrgtc2(): started for block <0x0005 : 0x0140b08c> objd: 0x00010fc5 ----------------This is index root block env: (scn: 0x0000.00148442 xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 statement num=0 parent xid: xid: 0x0000.000.00000000 scn: 0x0000.00000000 33sch: scn: 0x0000.00000000 mascn: (scn: 0x0000.00000000) ktrexc(): returning 2 on: 059864D0 scn: 0xffff.ffffffff xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 scn: 0xffff.ffffffff sfl: 0 ktrgtc2(): completed for block <0x0005 : 0x0140b08c> objd: 0x00010fc5 ktrgtc2(): started for block <0x0005 : 0x018080c4> objd: 0x00010fc5 ----------------This is index branch block, level 1 env: (scn: 0x0000.00148442 xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 statement num=0 parent xid: xid: 0x0000.000.00000000 scn: 0x0000.00000000 32sch: scn: 0x0000.00000000 mascn: (scn: 0x0000.00000000) ktrexc(): returning 2 on: 059864D0 scn: 0xffff.ffffffff xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 scn: 0xffff.ffffffff sfl: 0 ktrgtc2(): completed for block <0x0005 : 0x018080c4> objd: 0x00010fc5 ktrget2(): started for block <0x0005 : 0x0140b095> objd: 0x00010fc5 ----------------This is index leaf block, level 0 env: (scn: 0x0000.00148442 xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 statement num=0 parent xid: xid: 0x0000.000.00000000 scn: 0x0000.00000000 32sch: scn: 0x0000.00000000 mascn: (scn: 0x0000.00000000) ktrexf(): returning 9 on: 059864D0 scn: 0xffff.ffffffff xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 scn: 0xffff.ffffffff sfl: 0 ktrgcm(): completed for block <0x0005 : 0x0140b095> objd: 0x00010fc5 ktrget2(): completed for block <0x0005 : 0x0140b095> objd: 0x00010fc5 ktrget2(): started for block <0x0005 : 0x0140005f> objd: 0x00010fc4 ----------------This is trans data block env: (scn: 0x0000.00148442 xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 statement num=0 parent xid: xid: 0x0000.000.00000000 scn: 0x0000.00000000 32sch: scn: 0x0000.00000000 mascn: (scn: 0x0000.00000000) ktrexf(): returning 9 on: 059864D0 scn: 0xffff.ffffffff xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 scn: 0xffff.ffffffff sfl: 0 ktrgcm(): completed for block <0x0005 : 0x0140005f> objd: 0x00010fc4 ktrget2(): completed for block <0x0005 : 0x0140005f> objd: 0x00010fc4 ktrget2(): started for block <0x0005 : 0x0140b095> objd: 0x00010fc5 ----------------This is index leaf block, level 0 env: (scn: 0x0000.00148442 xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 statement num=0 parent xid: xid: 0x0000.000.00000000 scn: 0x0000.00000000 32sch: scn: 0x0000.00000000 mascn: (scn: 0x0000.00000000) ktrexf(): returning 9 on: 059864D0 scn: 0xffff.ffffffff xid: 0x0000.000.00000000 uba: 0x00000000.0000.00 scn: 0xffff.ffffffff sfl: 0 ktrgcm(): completed for block <0x0005 : 0x0140b095> objd: 0x00010fc5 ktrget2(): completed for block <0x0005 : 0x0140b095> objd: 0x00010fc5

From trace file, the query read leaf index block two times, it is very strange.

Regards
Jacky Received on Sat Apr 18 2009 - 23:47:30 CDT