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

Home -> Community -> Usenet -> c.d.o.server -> Re: Log switch query

Re: Log switch query

From: Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk>
Date: Sat, 7 Oct 2000 10:32:43 +0100
Message-ID: <970911646.26929.7.nnrp-04.9e984b29@news.demon.co.uk>

Have a look at the ARCH: archival started/stopped lines and their timestamps, I think this indicates that you have an automatic process somewhere kicking in every 5 minutes with the command 'archive log current'.

--

Jonathan Lewis
Yet another Oracle-related web site:  http://www.jlcomp.demon.co.uk

Howard J. Rogers wrote in message <39dd11cb_at_news.iprimus.com.au>...

>
>"Wolfgang Kremser" <wolfang.kremser_at_colt.at> wrote in message
>news:bk0D5.5$tZ4.120_at_stuart.coltnet.at...
>> in my opinion the checkpoint interval is too short ... its in the
>> initXXX.ora defined
>>
>> greetings Wolfgang
>
>
>That may well be, but issuing a checkpoint doesn't force a log switch
>(though the converse is of course true).
>
>As for the original question?
>
>Well, a quick glance at the alert log shows you switch redo logs every
three
>minutes or so, which is way, way, way too often. Your log members (all of
>them) to be much bigger to slow that down to something approaching a more
>reasonable logswitch per half hour, or per hour etc. Given that they are
>currently only 5Mb big (if I can read correctly, which is often
>problematic), you should be looking at re-creating your logfiles at around
>50 to 75Mb or even larger, and then dropping the old ones.
>
>When you're switching at a reasonable frequency, see what is happening to
>your archive sizes then.
>
>I can't think of anything that will force a log switch other than the extra
>things you mentioned, but when the redo subsystem is being put under stress
>like this, who knows what is going on. Sort that out first, and then see
>what's happening.
>
>Regards
>HJR
>
>
>
>>
>> Mike Hartas <michael.hartas_at_pancredit.com> wrote in message
>> news:8ri17s$2i7$1_at_nnrp1.deja.com...
>> > Hi,
>> > What can force a log switch other than redo log filling or an alter
>> > system switch logfile?. reason I asked is I'm looking at a list of
>> > achived logs and only every other one seems to be the size of a redo
>> > log. the alert log dosn`t show any alter system commands. what is going
>> > on ?
>> >
>> > Cheers, Mike.
>> >
>> > archive dir sample -
>> > -rw-r----- 1 oracle dba 5243392 Oct 5 13:38 PANL1_41717.dbf
>> > -rw-r----- 1 oracle dba 2008064 Oct 5 13:40 PANL1_41718.dbf
>> > -rw-r----- 1 oracle dba 5243392 Oct 5 13:43 PANL1_41719.dbf
>> > -rw-r----- 1 oracle dba 1871872 Oct 5 13:45 PANL1_41720.dbf
>> > -rw-r----- 1 oracle dba 5243392 Oct 5 13:49 PANL1_41721.dbf
>> > -rw-r----- 1 oracle dba 1413120 Oct 5 13:50 PANL1_41722.dbf
>> >
>> > alert log sample -
>> >
>> > Completed database checkpoint by background
>> > Thu Oct 5 13:35:03 2000
>> > ARCH: Archival stopped
>> > Thu Oct 5 13:35:03 2000
>> > Beginning database checkpoint by background
>> > Thu Oct 5 13:35:03 2000
>> > Thread 1 advanced to log sequence 41717
>> > Current log# 3 seq# 41717 mem#
>> > 0: /gbmdclp1/oradata4/PANL/redo/PANLlog_3_1.dbf
>> > Current log# 3 seq# 41717 mem#
>> > 1: /gbmdclp1/oradata5/PANL/redo/PANLlog_3_2.dbf
>> > Current log# 3 seq# 41717 mem#
>> > 2: /gbmdclp1/oradata6/PANL/redo/PANLlog_3_3.dbf
>> > Thu Oct 5 13:35:05 2000
>> > ARCH: Archival started
>> > Thu Oct 5 13:35:07 2000
>> > Completed database checkpoint by background
>> > Thu Oct 5 13:38:25 2000
>> > Beginning database checkpoint by background
>> > Thu Oct 5 13:38:25 2000
>> > Thread 1 advanced to log sequence 41718
>> > Current log# 2 seq# 41718 mem#
>> > 0: /gbmdclp1/oradata4/PANL/redo/PANLlog_2_1.dbf
>> > Current log# 2 seq# 41718 mem#
>> > 1: /gbmdclp1/oradata5/PANL/redo/PANLlog_2_2.dbf
>> > Current log# 2 seq# 41718 mem#
>> > 2: /gbmdclp1/oradata6/PANL/redo/PANLlog_2_3.dbf
>> > Thu Oct 5 13:38:26 2000
>> > Completed database checkpoint by background
>> > Thu Oct 5 13:40:02 2000
>> > ARCH: Archival stopped
>> > Thu Oct 5 13:40:03 2000
>> > Beginning database checkpoint by background
>> > Thu Oct 5 13:40:03 2000
>> > Thread 1 advanced to log sequence 41719
>> > Current log# 1 seq# 41719 mem#
>> > 0: /gbmdclp1/oradata4/PANL/redo/PANLlog_1_1.dbf
>> > Current log# 1 seq# 41719 mem#
>> > 1: /gbmdclp1/oradata5/PANL/redo/PANLlog_1_2.dbf
>> > Current log# 1 seq# 41719 mem#
>> > 2: /gbmdclp1/oradata6/PANL/redo/PANLlog_1_3.dbf
>> > Thu Oct 5 13:40:04 2000
>> > ARCH: Archival started
>> > Thu Oct 5 13:40:06 2000
>> > Completed database checkpoint by background
>> > Thu Oct 5 13:43:21 2000
>> > Beginning database checkpoint by background
>> > Thu Oct 5 13:43:21 2000
>> > Thread 1 advanced to log sequence 41720
>> > Current log# 3 seq# 41720 mem#
>> > 0: /gbmdclp1/oradata4/PANL/redo/PANLlog_3_1.dbf
>> > Current log# 3 seq# 41720 mem#
>> > 1: /gbmdclp1/oradata5/PANL/redo/PANLlog_3_2.dbf
>> > Current log# 3 seq# 41720 mem#
>> > 2: /gbmdclp1/oradata6/PANL/redo/PANLlog_3_3.dbf
>> > Thu Oct 5 13:43:25 2000
>> > Completed database checkpoint by background
>> > Thu Oct 5 13:45:02 2000
>> > ARCH: Archival stopped
>> > Thu Oct 5 13:45:03 2000
>> > Beginning database checkpoint by background
>> > Thu Oct 5 13:45:03 2000
>> > Thread 1 advanced to log sequence 41721
>> > Current log# 2 seq# 41721 mem#
>> > 0: /gbmdclp1/oradata4/PANL/redo/PANLlog_2_1.dbf
>> > Current log# 2 seq# 41721 mem#
>> > 1: /gbmdclp1/oradata5/PANL/redo/PANLlog_2_2.dbf
>> > Current log# 2 seq# 41721 mem#
>> > 2: /gbmdclp1/oradata6/PANL/redo/PANLlog_2_3.dbf
>> > Thu Oct 5 13:45:05 2000
>> > ARCH: Archival started
>> > Thu Oct 5 13:45:05 2000
>> > Completed database checkpoint by background
>> > Thu Oct 5 13:49:10 2000
>> > Beginning database checkpoint by background
>> > Thu Oct 5 13:49:10 2000
>> > Thread 1 advanced to log sequence 41722
>> > Current log# 1 seq# 41722 mem#
>> > 0: /gbmdclp1/oradata4/PANL/redo/PANLlog_1_1.dbf
>> > Current log# 1 seq# 41722 mem#
>> > 1: /gbmdclp1/oradata5/PANL/redo/PANLlog_1_2.dbf
>> > Current log# 1 seq# 41722 mem#
>> > 2: /gbmdclp1/oradata6/PANL/redo/PANLlog_1_3.dbf
>> > Thu Oct 5 13:49:12 2000
>> > Completed database checkpoint by background
>> > Thu Oct 5 13:50:02 2000
>> > ARCH: Archival stopped
>> > Thu Oct 5 13:50:03 2000
>> > Beginning database checkpoint by background
>> > Thu Oct 5 13:50:03 2000
>> > Thread 1 advanced to log sequence 41723
>> > Current log# 3 seq# 41723 mem#
>> > 0: /gbmdclp1/oradata4/PANL/redo/PANLlog_3_1.dbf
>> > Current log# 3 seq# 41723 mem#
>> > 1: /gbmdclp1/oradata5/PANL/redo/PANLlog_3_2.dbf
>> > Current log# 3 seq# 41723 mem#
>> > 2: /gbmdclp1/oradata6/PANL/redo/PANLlog_3_3.dbf
>> > Thu Oct 5 13:50:06 2000
>> > ARCH: Archival started
>> > Thu Oct 5 13:50:08 2000
>> > Completed database checkpoint by background
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> > Sent via Deja.com http://www.deja.com/
>> > Before you buy.
>>
>>
>
>
Received on Sat Oct 07 2000 - 04:32:43 CDT

Original text of this message

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