Thanks, Tanel.
Yes, I'll be at RMOUG.
I'll be in the front row for your presentation.
Again, thanks for the info.
Barb
- Tanel Poder <tanel.poder.003_at_mail.ee> wrote:
> Comments below.
>
> ----- Original Message -----
> To: "Multiple recipients of list ORACLE-L"
> <ORACLE-L_at_fatcity.com>
> Sent: Thursday, January 29, 2004 7:04 PM
>
>
> > Jonathan / Tanel:
> > I, however, AM having this problem. Didn't know
> where
> > to look till I saw this message. (I love this
> list!)
>
> So do I ;)
>
> >
> > I've yet to capture the sid (and therefore track
> back
> > to the table) where the 'direct path write'
> occurs.
> > Definitions for the tables were supplied by the
> > vendor. When I look at at the lobs, the
> definitions
> > are mixed. Most are nocache, logging yes. some
> are
> > no/no
> >
> > I don't see much on metalink about this -- just a
> > couple of generic articles on lob storage.
>
> If you are up to attending RMOUG Training days in
> few weeks, then I'll be
> speaking about LOB performance tuning there, you're
> most welcome ;)
>
> >
> > Should I change the lobs to cache/logging across
> the
> > board?
>
> Well, if you set all your LOBs to CACHE (cached LOBs
> are always logged),
> then you'll be generating more redo for current
> nocache nologging LOBs as
> well. Also, your buffer cache gets hit, especially
> if your LOBs are big.
>
> One possibility is setting event 10359 at level 1 to
> what Jonathan also
> referred, then your direct writes won't cause
> controlfile updates anymore.
> But when you have set this event, you can't rely on
> UNRECOVERABLE_CHANGE# in
> your V$DATAFILE anymore, for determining whether a
> datafile should be backed
> up due nologging operations in it.
>
> If you should ever need to restore a file containing
> NOLOGGING NOCACHE LOBs,
> then any "inconsistent" blocks will be marked
> corrupt (block sqn=0xFF) since
> there was no redo information logged for them. In
> that case you have update
> the corrupt LOBs to nulls for example to reclaim the
> space in LOB segment.
> Note that having few corrupt LOB items in a LOB
> segment won't affect other,
> healthy LOB items, you just have to have some kind
> of error handling
> mechanism implemented when you hit a corrupt item.
>
> Note that updates to LOB indexes are always cached
> and logged, despite the
> NOLOGGING setting.
>
> Tanel.
>
>
> >
> > Thanks for any insight.
> > Barb
> >
> > Table
> > Name Cach LOG
> > ---------------- ---- ---
> > CONTACTGROUP NO YES
> > CONTENT NO YES
> > IMAGEBINARY NO YES
> > IMAGEHEADER NO YES
> > IMAGEROW NO NO
> > IMPORTIN NO NO
> > TF_NEW_CONTENT NO YES
> > XMLS_DICTIONARY NO YES
> > XMLS_SALESSUPPORT NO NO
> > XMLS_STYLEPART YES YES
> >
> > --- Tanel Poder <tanel.poder.003_at_mail.ee> wrote:
> > > Hi!
> > >
> > > > Do you have any LOBs defined with
> > > > the nocache nologging attribute ?
> > >
> > > This was what I initially thought would be the
> > > problem, but then I thought,
> > > in case of nocache lobs you should see direct
> path
> > > writes (lob) a lot...
> > >
> > > Tanel.
> > >
> > >
> > >
> > >
> > > --
> > > Please see the official ORACLE-L FAQ:
> > > http://www.orafaq.net
> > > --
> > > Author: Tanel Poder
> > > INET: tanel.poder.003_at_mail.ee
> > >
> > > 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).
> >
> >
> > __________________________________
> > Do you Yahoo!?
> > Yahoo! SiteBuilder - Free web site building tool.
> Try it!
> > http://webhosting.yahoo.com/ps/sb/
> > --
> > 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: Tanel Poder
> INET: tanel.poder.003_at_mail.ee
>
> 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).
Do you Yahoo!?
Yahoo! SiteBuilder - Free web site building tool. Try it!
http://webhosting.yahoo.com/ps/sb/
--
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).
Received on Thu Jan 29 2004 - 21:04:25 CST