All of lore.kernel.org
 help / color / mirror / Atom feed
From: Matthias Schniedermeyer <ms@citd.de>
To: Shawn Bohrer <shawn.bohrer@gmail.com>
Cc: xfs@oss.sgi.com
Subject: Re: High latencies writing to a memory mapped file
Date: Tue, 27 Jul 2010 11:24:52 +0200	[thread overview]
Message-ID: <20100727092452.GA23307@citd.de> (raw)
In-Reply-To: <20100722144706.GA2840@BohrerMBP.rgmadvisors.com>

On 22.07.2010 09:47, Shawn Bohrer wrote:

I can't help with your exact problem, but this reminds me of a quite 
similar looking problem i have.

We have a linux-computer with samba to act as a fileserver for a few 
Windows-Clients and ever since kernel 2.6.26 (Never happend with 2.6.25) 
the server randomly "hangs" for a few seconds (Which is more of a 
problem that drives people crazy). It takes about 1-2 weeks for the 
symtoms to show and only a reboot of the computer "cures" it (for 
another 1-2 weeks).

Because of the long incubation period and the whole slew of subsystems 
(and samba) involded, i never had the drive to debug this, as a "quick" 
reboot every 1-2 weeks only takes about 2 minutes.

With "whole slew of subsystems" in mean:
Hardware: SATA on onboard-controller (Intel G41-Chipset(ICH7) with 
PIIX-driver, because BIOS is crippled and doesn't allow AHCI)
(Hardware itself is unlikly to be be at fault, because a few month ago 
the complete computer and HDDs were switched, but the problem persisted)

Kernel:
Currently 2.6.32(-5, Debian SID), first seen with 2.6.26 (vanilla)

Drivers/Filesystem:
md(RAID1), loop-aes, XFS
(I tried logbufs=8,logbsize=256k,nobarrier, no apparent changes. I'm 
not 100% sure if i tried other schedulers, but i think i did)

And finally Samba on top.


As i had to reboot the computer just yesterday, and will have to power 
down the server over the weekend because we are moving, it will be at 
least 2-3 weeks before i anticipate the problem to resurface.
In case anybody has an idea what i could do to identify and/or isolate 
the root-cause, i'm open for suggestions.




Bis denn

-- 
Real Programmers consider "what you see is what you get" to be just as 
bad a concept in Text Editors as it is in women. No, the Real Programmer
wants a "you asked for it, you got it" text editor -- complicated, 
cryptic, powerful, unforgiving, dangerous.

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

  parent reply	other threads:[~2010-07-27  9:22 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-07-22 14:47 High latencies writing to a memory mapped file Shawn Bohrer
2010-07-22 23:28 ` Dave Chinner
2010-07-26 22:09   ` Shawn Bohrer
2010-07-26 23:22     ` Dave Chinner
2010-08-03 22:03       ` Shawn Bohrer
2010-07-27  9:24 ` Matthias Schniedermeyer [this message]
2010-07-27 10:47   ` Emmanuel Florac
2010-07-27 11:27     ` Matthias Schniedermeyer
2010-07-27 13:06       ` Emmanuel Florac
2010-07-27 13:00     ` Stan Hoeppner
     [not found]   ` <4C4EDEFD.7000401@hardwarefreak.com>
2010-07-27 14:49     ` Matthias Schniedermeyer
2010-07-27 19:59       ` Kinzel, David
2010-07-27 21:02         ` Matthias Schniedermeyer

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=20100727092452.GA23307@citd.de \
    --to=ms@citd.de \
    --cc=shawn.bohrer@gmail.com \
    --cc=xfs@oss.sgi.com \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.