Hi,
I guess the sysadmin did not tell me everything when I
asked him if we got Quick IO. We do not have a
separate software, it's only an on/off toggle on vxfs
volume in volume manager .
So I guess I'll switch disk_asynch_io from true to
false.
Am I right ?
- ARUN K C <arun_k_c_at_hotmail.com> a écrit :
<HR>
<html><DIV>
<P>ave u enabled quick i/o on the files u want to
monitor</P>
<P>can u please let me know as to how u have enabled
the quick i/o</P>
<P>looks like it is not enabled.</P>
<P>there are two ways to enable the quick i/o</P>
<P>first enable the filesystem as quick i/o</P>
<P>2nd create quick i/o file and then create the
datafile based on that name.</P>
<P>or if u have already have a datafile and if u want
to enable quick i/o on that then u will have to
offline that particular file and then enable quick i/o
on it</P>
<P>the syntax goes like the below for datafiles</P>
<P>qiomkfile -h 8k -s 100M name_of_the_file</P>
<P>the above command creates 2 files one is a
.(dot)file and another a link which is linked to this
.(dot)file</P>
<P>or if u have already a datafile and u want to
enable this as a quick i/o file</P>
<P>first bring it offline</P>
<P>mv datafile .datafile(ie dotdatafile there is a dot
prior to the filename)</P>
<P>then create a link on that datafile</P>
<P>the scene goes as below</P>
<P>mv log1blink.ora .log1blink.ora</P>
<P>ln -s log1blink.ora .log1blink.ora</P>
<P>once this is done bring that file online and
voila</P>
<P>see the quick i/o working</P>
<P>if u feel it is not working right then</P>
<P>make that file offline rm the link and rename the
.file as a normal file.</P>
<P>Below is the truss on one of my logwriter which is
showing that aio write is going on</P>
<P> </P>
<P>994/1:
sigprocmask(SIG_BLOCK, 0xEFFFE740, 0x00000000) =
0<BR>994/1:
times(0xEFFFE670)
=
15661119<BR>994/1:
setitimer(ITIMER_REAL, 0xEFFFE670, 0x00000000) =
0<BR>994/1:
sigprocmask(SIG_UNBLOCK, 0xEFFFE6E0, 0x00000000) =
0<BR>994/1:
semop(131072, 0xEFFFE748,
1)
=
0<BR>994/1:
sigprocmask(SIG_BLOCK, 0xEFFFE740, 0x00000000) =
0<BR>994/1: !
!
setitimer(ITIMER_REAL, 0xEFFFE670, 0x00000000) =
0<BR>994/1:
sigprocmask(SIG_UNBLOCK, 0xEFFFE710, 0x00000000) =
0<BR>994/1:
kaio(AIOWRITE, 306, 0xA7918000, 512, 0x038BE400,
0x01082DE4) =
0<BR>994/1:
kaio(AIOWAIT,
0xEFFFE328)
=
17313252<BR>994/1:
semctl(131072, 232, 8,
1)
=
0<BR>994/1:
getcontext(0xEFFFE590)<BR>994/1:
sigprocmask(SIG_BLOCK, 0xEFFFE740, 0x0000!
!
0000) =
0<BR>994/1:
times(0xEFFFE670)
=
15661121<BR>994/1:
setitimer(ITIMER_REAL, 0xEFFFE670, 0x00000000) =
0<BR>994/1:
sigprocmask(SIG_UNBLOCK, 0xEFFFE6E0, 0x00000000) =
0<BR>994/1:
semop(131072, 0xEFFFE748,
1)
=
0<BR>994/1:
sigprocmask(SIG_BLOCK, 0xEFFFE740, 0x00000000) =
0<BR>994/1:
setitimer(ITIMER_REAL, 0xEFFFE670, 0x00000000) =
0<BR>994/1:&n!
!
bsp;
sigprocmask(SIG_UNBLOCK, 0xEFFFE710, 0x00000000) =
0<BR>994/1:
kaio(AIOWRITE, 306, 0xA7918200, 9216, 0x038BE600,
0x01082D50) =
0<BR>994/1:
kaio(AIOWAIT,
0xEFFFE328)
=
17313104<BR>994/1:
semctl(131072, 214, 8,
1)
=
0<BR>994/1:
getcontext(0xEFFFE590)<BR>994/1:
sigprocmask(SIG_BLOCK, 0xEFFFE740, 0x00000000) =
0<BR>994/1:
times(0xEFFFE670) &!
!
nbsp;
=
15661135<BR>994/1:
setitimer(ITIMER_REAL, 0xEFFFE670, 0x00000000) =
0<BR>994/1:
sigprocmask(SIG_UNBLOCK, 0xEFFFE6E0, 0x00000000) =
0<BR><BR></P></DIV>
<DIV></DIV>
<DIV></DIV>>From: paquette stephane
<STEPHANE_PAQUETTE_at_YAHOO.COM>
<DIV></DIV>>Reply-To: ORACLE-L_at_fatcity.com
<DIV></DIV>>To: Multiple recipients of list
ORACLE-L <ORACLE-L_at_FATCITY.COM>
<DIV></DIV>>Subject: Asynch IO and truss question
<DIV></DIV>>Date: Thu, 07 Dec 2000 19:29:44 -0800
<DIV></DIV>>
<DIV></DIV>>Hi all,
<DIV></DIV>>
<DIV></DIV>>Since the sys admin is available 15min
per week, I've
<DIV></DIV>>started to play with truss to make sure
asynch IO was
<DIV></DIV>>done.
<DIV></DIV>>This is on Solaris 2.7, Oracle 8.1.6
and Veritas Quick
<DIV></DIV>>IO.
<DIV></DIV>>I've done truss -fl on the the dbwr
process.
<DIV></DIV>>It's giving me
<DIV></DIV>>28529/1: Received signal #14, SIGALRM,
in
<DIV></DIV>>semop() [caught]
<DIV></DIV>>28529/1: semop(37224452,
0xFFFFFFFF7FFFDCAC, 1)
<DIV></DIV>> Err#91 ERESTART
<DIV></DIV>>28529/1: sigprocmask(SIG_BLOCK,
<DIV></DIV>>0xFFFFFFFF7FFFD770, 0x00000000) = 0
<DIV></DIV>>28529/1: times(0xFFFFFFFF7FFFD6A0)
<DIV></DIV>> = 243599308
<DIV></DIV>>28529/1: sigprocmask(SIG_UNBLOCK,
<DIV></DIV>>0xFFFFFFFF7FFFD770, 0x00000000) = 0
<DIV></DIV>>28529/1: getcontext(0xFFFFFFFF7FFFD3C0)
<DIV></DIV>>28529/1: setcontext(0xFFFFFFFF7FFFD3C0)
<DIV></DIV>>28529/1: sigprocmask(SIG_BLOCK,
<DIV></DIV>>0xFFFFFFFF7FFFDAE0, 0x00000000) = 0
<DIV></DIV>>28529/1: setitimer(ITIMER_REAL,
<DIV></DIV>>0xFFFFFFFF7FFFD9B0, 0x00000000) = 0
<DIV></DIV>>28529/1: sigprocmask(SIG_UNBLOCK,
<DIV></DIV>>0xFFFFFFFF7FFFDAB0, 0x00000000) = 0
<DIV></DIV>>28529/1: kaio(AIOWRITE, 402,
0x381C8C000,
<DIV></DIV>>16384, 0x0342000001E92B70) Err#48
ENOTSUP
<DIV></DIV>>28529/1:
lwp_cond_signal(0xFFFFFFFF7C8EBF70)
<DIV></DIV>> = 0
<DIV></DIV>>28529/140:
lwp_cond_wait(0xFFFFFFFF7C8EBF70,
<DIV></DIV>>0xFFFFFFFF7C8EBF80, 0x00000000) = 0
<DIV></DIV>>28529/140: pwrite(402,
"0202\0\0\080\r\b\00E
<DIV></DIV>>}1A".., 16384, 0x03420000) = 16384
<DIV></DIV>>28529/1: kaio(AIOWAIT,
0xFFFFFFFF7FFFD710)
<DIV></DIV>> = 1
<DIV></DIV>>28529/140: kaio(AIONOTIFY, -2184159392)
<DIV></DIV>> = 0
<DIV></DIV>>28529/1: getcontext(0xFFFFFFFF7FFFD8D0)
<DIV></DIV>>28529/1: sigprocmask(SIG_BLOCK,
<DIV></DIV>>0xFFFFFFFF7FFFDAE0, 0x00000000) = 0
<DIV></DIV>>28529/1: times(0xFFFFFFFF7FFFD9B0)
<DIV></DIV>> = 243599310
<DIV></DIV>>28529/1: setitimer(ITIMER_REAL,
<DIV></DIV>>0xFFFFFFFF7FFFD9B0, 0x00000000) = 0
<DIV></DIV>>28529/1: sigprocmask(SIG_UNBLOCK,
<DIV></DIV>>0xFFFFFFFF7FFFDA80, 0x00000000) = 0
<DIV></DIV>>28529/1: semop(37224452,
0xFFFFFFFF7FFFDCAC, 1)
<DIV></DIV>>(sleeping...)
<DIV></DIV>>28529/2: kaio(6, 0x00000000,
0x00000000,
<DIV></DIV>>0x00000000, 0x00000000, 0x00000000)
(sleeping...)
<DIV></DIV>>28529/3:
lwp_cond_wait(0xFFFFFFFF7D58DF70,
<DIV></DIV>>0xFFFFFFFF7D58DF80, 0x00000000)
(sleeping...)
<DIV></DIV>>28529/4:
lwp_cond_wait(0xFFFFFFFF7D57BF70,
<DIV></DIV>>0xFFFFFFFF7D57BF80, 0x00000000)
(sleeping...)
<DIV></DIV>>...
<DIV></DIV>>
<DIV></DIV>>I've read on metalink that the error 48
is because of
<DIV></DIV>>the vol_maxio buffer being smaller than
<DIV></DIV>>db_file_direct_io_count*db_block_size.
<DIV></DIV>>Anybody can confirm ?
<DIV></DIV>>
<DIV></DIV>>Also, should I be concerned with the 91
error ?
<DIV></DIV>>
<DIV></DIV>>TIA
<DIV></DIV>>
<DIV></DIV>>
<DIV></DIV>>=====
<DIV></DIV>>Stephane Paquette
<DIV></DIV>>DBA Oracle
<DIV></DIV>>stephane_paquette_at_yahoo.com
<DIV></DIV>>
<DIV></DIV>>__________________________________________________
<DIV></DIV>>Do You Yahoo!?
<DIV></DIV>>Yahoo! Shopping - Thousands of Stores.
Millions of Products.
<DIV></DIV>>http://shopping.yahoo.com/
<DIV></DIV>>--
<DIV></DIV>>Please see the official ORACLE-L FAQ:
http://www.orafaq.com
<DIV></DIV>>--
<DIV></DIV>>Author:
=?iso-8859-1?q?paquette=20stephane?=
<DIV></DIV>> INET: stephane_paquette_at_yahoo.com
<DIV></DIV>>
<DIV></DIV>>Fat City Network Services -- (858)
538-5051 FAX: (858) 538-5051
<DIV></DIV>>San Diego, California -- Public
Internet access / Mailing Lists
<DIV></DIV>>--------------------------------------------------------------------
<DIV></DIV>>To REMOVE yourself from this mailing
list, send an E-Mail message
<DIV></DIV>>to: ListGuru_at_fatcity.com (note EXACT
spelling of 'ListGuru') and in
<DIV></DIV>>the message BODY, include a line
containing: UNSUB ORACLE-L
<DIV></DIV>>(or the name of mailing list you want
to be removed from). You may
<DIV></DIV>>also send the HELP command for other
information (like subscribing).
<DIV></DIV><br clear=all><hr>Get more from the Web.
FREE MSN Explorer download : <a
href="http://explorer.msn.com">http://explorer.msn.com</a><br></p></html>
--
Please see the official ORACLE-L FAQ:
http://www.orafaq.com
--
Author: ARUN K C
INET: arun_k_c_at_hotmail.com
Fat City Network Services -- (858) 538-5051 FAX:
(858) 538-5051
San Diego, California -- Public Internet access
/ Mailing Lists
--------------------------------------------------------------------
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
Received on Fri Dec 15 2000 - 08:26:40 CST