<html><head><style type='text/css'>body { font-family: 'Arial'; font-size: 12pt; color: #000000}</style></head><body><br>----- Original Message -----<br>From: "Antonio Varni" <avarni@estalea.com><br>To: centos@centos.org<br>Sent: Friday, November 16, 2007 9:06:52 AM (GMT+1000) Australia/Brisbane<br>Subject: [CentOS] IO causing major performance issues<br><br>Hello everyone.<br><br>I'm wondering what other people's experiences are WRT systems becoming<br>unresponsive (unable to ssh in, etc) for brief periods of time when<br>a large amount of IO is being performed. It's really starting to<br>cause a problem for us. We're on Dell PowerEdge 1955 blades - but this same<br>issue has caused us problems on PE1950, PE1850, PE1750 servers.<br><br>We're running Centos 4.5 right now. I know Centos 5 includes ionice, more<br>io scheduler/elevator selections like deadlock/etc. Perhaps that would<br>fix this issue. We're running the latest PERC firmware.<br><br>The specific issue I'm referring to at this point is on a system running<br>mysql. All mysql data files are on a netapp filer but mysql's tmp directory<br>is on local disk. Whenever a lot of temp tables are created (and thus<br>written and deleted from local disk quickly) we can't even log in to the<br>machine - and our monitoring system gets all freaked out and we get<br>lots of pages, etc... FYI this is two disks with hardware raid 1.<br><br>Is it just me? Or is this specific to Dell systems, or is this just<br>the state of the Linux kernel these days? Is there some magical patch<br>I can apply to make this issue go away :)<br><br><br>Thanks in advance for any insight into this issue.<br><br>Antonio<br><br><br><br>-- <br>_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _<br>antonio varni<br>[ technology ]<br><br>ESTALEA, L.P.<br>629 State Street #222<br>Santa Barbara, CA 93101<br>v 805.252.0115<br>f 805.899.2697<br>e avarni@estalea.com<br>w www.estalea.com<br>_______________________________________________<br>CentOS mailing list<br>CentOS@centos.org<br>http://lists.centos.org/mailman/listinfo/centos<br><br>-- <br>This message has been scanned for viruses and<br>dangerous content by MailScanner, and is<br>believed to be clean.<br><br><br><br><br>I have noticed similar behaviour on all sort of linuxes (in particular, ssh into the box is really slow when it's doing IO) and wondered why, but never really thought about investigating any further.<br><br>Unfortunately, I do a lot of work with solaris and the funny thing is that I have *never* seen a solaris kernel exhibit this sort of behaviour. Even if it is installed on normal IDE/SATA disks. And, in fact, even if installed on the exact same hardware.<br><br><br>Now I'm curious.....especially given that I'm right in the middle of pushing to get rid of solaris in favour of RHEL.<br><br><br><br><br><br><br><br><br><br><br></body><br />--
<br />This message has been scanned for viruses and
<br />dangerous content by
<a href="http://www.mailscanner.info/"><b>MailScanner</b></a>, and is
<br />believed to be clean.
</html>