- Previous message: Wil Cooley: "Re: Mailstore filesystem"
- Maybe in reply to: Søren Schimkat: "Mailstore filesystem"
- Next in thread: Phil Pennock: "Re: Mailstore filesystem"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
|
|
|
From: Michael Loftis (no email)
Date: Wed Jul 05 2006 - 17:43:42 EDT
It would be nice to have more details about version of ReiserFS, what hash
was being used, kernel version, hardware involved, and NFS or not
(especially kernel NFSd)...
That said we use ReiserFS on our mail and on our NFS servers running a
2.4.27 variant with about half a TB in NFS and about 120Gb in mail, no
issues whatsoever. We did have an issue with the CX200 that caused some
corruption in the NFS stores once, but that's resolved now. We were
running old FLARE that caused the array to spit out multiple drives when it
had a vault drive fail.
--On July 3, 2006 11:12:01 PM +0200 Phil Pennock
<> wrote:
> On 2006-07-03 at 15:44 +0200, Daniel Eckl wrote:
>> At the moment I have a temporary machine running the same system with
>> reiserfs. While I still have the problem of occasional high load, the
>> system now never freezes or hangs, just slowes down.
>
> On my private machine at home, I used ReiserFS until very recently; I
> had assumed that the negative publicity was scare-mongering based on
> historical problems and that by this time any problems would have been
> sorted out. I didn't believe the adage, "If you use ReiserFS and
> haven't lost data, then you haven't been using it for long enough."
>
> /home partition held regular /home, plus Cyrus IMAP content and a
> document repository updated nightly by rsync; the load is almost always
> trivial. Over-night, the load comes from:
>
> * occasional cron-jobs updating data; those should be read-only,
> storing to /var/, so irrelevant
> * the rsync of the documents
> * Cyrus IMAP mail deliveries and squatter, etc.
>
> It is rather disconcerting to see files which exist as directory entries
> but where the inode (or whatever its equivalent is) doesn't exist so
> that stat() fails. Race-conditions, fine, but this wasn't that.
>
> I bought an external HDD to spool the data off and migrated to xfs.
>
> Given this experience with a lightly-loaded system, I've joined the camp
> of "never use ReiserFS for production usage with data you value". I
> might consider using it for a Squid cache, I suppose.
> --
> "Everything has three factors: politics, money, and the right way to do
> it. In that order." -- Gary Donahue
> ----
> Cyrus Home Page: http://asg.web.cmu.edu/cyrus
> Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu
> List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
>
-- "Genius might be described as a supreme capacity for getting its possessors into trouble of all kinds." -- Samuel Butler ---- Cyrus Home Page: http://asg.web.cmu.edu/cyrus Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
|
|
|