<br>I had no idea that such a tool existed on Linux. Do anyone know of a graphical/web frontend for systemtap?<br><br>Regards,<br>Luis<br><blockquote class="replbq" style="border-left: 2px solid rgb(16, 16, 255); margin-left: 5px; padding-left: 5px;"><br><b><i>Ulf Zimmermann <ulf@atc-onlane.com></i></b> wrote:<blockquote class="replbq" style="border-left: 2px solid rgb(16, 16, 255); margin-left: 5px; padding-left: 5px;"> I will look at it. In the meanwhile I did find at least one of the<br>standby processes reading in bursts every 60-70 seconds like 400MB in<br>14.xx seconds from control01.ctl, even that file is only 94MB large.<br><br>> -----Original Message-----<br>> From: Andrew Phillips [mailto:Andrew.Phillips@betfair.com]<br>> Sent: Tuesday, February 19, 2008 01:43<br>> To: Ulf Zimmermann<br>> Cc: ocfs2-users@oss.oracle.com<br>> Subject: RE: [Ocfs2-users] Anyone have an idea how to find file<br>> i/othroughput?<br>> <br>> Ulf,<br>>
<br>> Have you considered using systemtap? There is a recipe here that<br>could<br>> be used to find out whats going on;<br>> <br>> http://sourceware.org/systemtap/wiki/WSDeviceMonitor?highlight=%28%<br>> 28WarStories%29%29<br>> <br>> I'm not sure how well that would work with ocfs2. Unlike dtrace,<br>> systemtap can be more "uneven" in coverage. Its also something that<br>> requires a bit of fiddling (installing debuginfo packages).<br>> <br>> The recipe above traps vfs_read and vfs_write so should work as a<br>> first stab at identifying the process id thats causing the I/O.<br>> <br>> I'd also advise some thought if its to be used on a production<br>> environment. Having said that, I've used it on a production oracle RAC<br>> database server and found it very valuable.<br>> <br>> I don't recall you mentioning the distribution, but RH, CentOS, and<br>> oracle's version of CentOS should all
work.<br>> <br>> As always, read the instructions on the label, etc...<br>> <br>> Andy<br>> <br>> <br>> On Mon, 2008-02-18 at 23:14 -0800, Ulf Zimmermann wrote:<br>> > Forgot to mention, this remote server is just Oracle. It has one<br>standby<br>> > database and one local database, the local one is suppose to be<br>idle,<br>> > i.e. nothing connecting to it, besides once in a while for available<br>> > check.<br>> ><br>> > While the primary database of the standby was down, I saw less disk<br>read<br>> > access, but every 5 minutes for about 60 seconds I would see<br>> > 50-60MB/sec. After the primary came back up, read access is as high<br>as<br>> > 160MB/sec.<br>> ><br>> > We are only seeing it on this single node of the remote standby. The<br>> > local standby (on EXT3) is not doing the same thing.<br>> ><br>> > > -----Original Message-----<br>> >
> From: Sunil Mushran [mailto:sunil.mushran@oracle.com]<br>> > > Sent: Monday, February 18, 2008 19:28<br>> > > To: Ulf Zimmermann<br>> > > Cc: ocfs2-users@oss.oracle.com<br>> > > Subject: Re: [Ocfs2-users] Anyone have an idea how to find file<br>i/o<br>> > > throughput?<br>> > ><br>> > > If a userspace process is behind the io surge, then strace should<br>> > help.<br>> > > But determining the process may require a bit of trial and error.<br>> > ><br>> > > Ulf Zimmermann wrote:<br>> > > > We got a remote Oracle 10g R2 standby running on OCFS2. Initial<br>when<br>> > we<br>> > > > started the standby, read I/O was < 5MB/sec on average. Since<br>then<br>> > it<br>> > > > has grown to over 40MB/sec (longer average, it peaks much<br>higher).<br>> > Here<br>> > > > is a graph showing this:<br>> > >
><br>> > > > http://www.alameda.net/~ulf/dbphx01.png<br>> > > ><br>> > > > We also have a local standby running (on EXT3) which is not<br>showing<br>> > the<br>> > > > same symptom. I am trying to find where all these reads are<br>> > happening.<br>> > > > Anyone have an idea how to figure that out on Linux?<br>> > > ><br>> > > > Ulf.<br>> > > ><br>> > > ><br>> > > > _______________________________________________<br>> > > > Ocfs2-users mailing list<br>> > > > Ocfs2-users@oss.oracle.com<br>> > > > http://oss.oracle.com/mailman/listinfo/ocfs2-users<br>> > > ><br>> ><br>> ><br>> > _______________________________________________<br>> > Ocfs2-users mailing list<br>> > Ocfs2-users@oss.oracle.com<br>> > http://oss.oracle.com/mailman/listinfo/ocfs2-users<br>>
<br>><br>________________________________________________________________________<br>> In order to protect our email recipients, Betfair Group use SkyScan<br>from<br>> MessageLabs to scan all Incoming and Outgoing mail for viruses.<br>> <br>><br>________________________________________________________________________<br><br>_______________________________________________<br>Ocfs2-users mailing list<br>Ocfs2-users@oss.oracle.com<br>http://oss.oracle.com/mailman/listinfo/ocfs2-users<br></blockquote><br><div> </div><hr size="1">Be a better friend, newshound, and know-it-all with Yahoo! Mobile. <a href="http://us.rd.yahoo.com/evt=51733/*http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ%20"> Try it now.</a></blockquote><br><p> 
<hr size=1>Never miss a thing. <a href="http://us.rd.yahoo.com/evt=51438/*http://www.yahoo.com/r/hs"> Make Yahoo your homepage.</a>