LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Chris Snook <csnook@redhat.com>
To: Andre Noll <maan@systemlinux.org>
Cc: "adam radford" <aradford@gmail.com>,
	"Tony Battersby" <tonyb@cybernetics.com>,
	"Johannes Wörner" <johannes.woerner@tuebingen.mpg.de>,
	"James Bottomley" <James.Bottomley@HansenPartnership.com>,
	linux-scsi <linux-scsi@vger.kernel.org>,
	linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: Performance problems with 3ware 9500S-4LP and 2.6.25-rc3
Date: Wed, 27 Feb 2008 14:55:52 -0500	[thread overview]
Message-ID: <47C5C048.9080704@redhat.com> (raw)
In-Reply-To: <20080227101143.GC30767@skl-net.de>

Andre Noll wrote:
> On 13:33, Chris Snook wrote:
> 
>>> During the daily cron job that uses rsync to sync a 500G file system
>> >from another machine to the raid on the 3ware controller the load
>>> jumps up, and the machine becomes sluggish as hell. For example, an
>>> ssh login to that machine takes minutes to complete and ldap becomes
>>> unreliable while the rsync job is running. Even Nagios complains
>>> about the machine being down while rsync is running.
>> You're putting your box under astronomical load.  This is generally 
>> regarded as a bad idea, regardless of how well your storage controller 
>> is performing.
> 
> The machine becomes sluggish also when I write directly to the raid
> array. A simple
> 
> 	dd if=/dev/zero of=tmpfile
> 
> shouldn't push the load up to 4, right?

Actually, it's normal for pdflush to spawn up to 8 threads when you're 
dirtying memory faster than it can be written to disk.  Load going to 4 
is not abnormal.  ssh logins taking minutes is very abnormal.

>> Can you measure the single-threaded throughput (say, 
>> coping one huge file, and then syncing) to give us a baseline 
>> performance figure?
> 
> Single threaded throughput seems to be ok (140M/s). The problem is
> that the machine becomes unresponsive.

Does the machine become unresponsive during the single-threaded test, or 
only when doing the rsync?

	-- Chris

  reply	other threads:[~2008-02-27 19:57 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-02-26 17:43 Andre Noll
2008-02-26 17:54 ` Bernd Schubert
2008-02-27 10:10   ` Andre Noll
2008-02-26 18:33 ` Chris Snook
2008-02-27 10:11   ` Andre Noll
2008-02-27 19:55     ` Chris Snook [this message]
2008-02-28  9:46       ` Andre Noll
2008-02-26 23:07 ` Arnd Hannemann
2008-02-27 10:11   ` Andre Noll
2008-02-27 14:26     ` Tony Battersby
2008-02-27 17:05       ` Andre Noll
2008-02-27 20:06         ` adam radford
2008-02-28  9:46           ` Andre Noll

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=47C5C048.9080704@redhat.com \
    --to=csnook@redhat.com \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=aradford@gmail.com \
    --cc=johannes.woerner@tuebingen.mpg.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=maan@systemlinux.org \
    --cc=tonyb@cybernetics.com \
    --subject='Re: Performance problems with 3ware 9500S-4LP and 2.6.25-rc3' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).