At Mon, 03 Jul 2006 17:22:07 -0500,
Phil Brutsche wrote:
>
> Søren Schimkat wrote:
> >
> > Which filesystem would you recomend?
>
> ext3, hands down.
Or good old trusty FFS, as the case may be. :-)
(It is more than rock solid, and has been tested for decades on dozens
of different machine arc
thus Michael Loftis spake:
I'd be very careful with softupdates. If you shut down uncleanly for
any reason in the past it has caused *SEVERE* loss of data for me. Not
just once, but many many times. It has gotten much better but I haven't
put it to the test lately at all.
softupdates turn
I'd be very careful with softupdates. If you shut down uncleanly for any
reason in the past it has caused *SEVERE* loss of data for me. Not just
once, but many many times. It has gotten much better but I haven't put it
to the test lately at all.
--On July 10, 2006 10:17:57 AM -0400 Forrest
For the record... what about file systems for the mail spool on
non-linux systems, such as FreeBSD, et al.
I've read a number of documents that addressed (classic) Usenet-based
activity (applicable to cyrus) - most agree that FreeBSD UFS+SoftUpdates
performs very well.
Thanks.
Cyru
On 2006-07-05 at 15:43 -0600, Michael Loftis wrote:
> 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)...
I've never tried NFS in a production environment at work, so this was
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 1
On Wed, 2006-07-05 at 17:47 +0200, Marten Lehmann wrote:
> Hello
>
> > Really? Nice. But you would have to re-create the ext3 filesystem with
> > this newer 2.6 kernel, right? Or would tune2fs do the job given the
> > right options?
>
> I once enabled the index-option with tune2fs on an old ext3
On 05 Jul 2006, at 10:07, John Madden wrote:
True, but I do expect to reach this number on this machine in the next
couple of years. ...And reiserfs has been just fine so far. Then
again, I didn't even consider using ext3 at the time.
While I've not had performance problems with reiser3, I ha
> Hello,
>
>> How big? ext3 STILL only supports 32000 directories within a directory.
>> That gets to be quite a problem on large installs.
>
> it is true: Performance goes down on very large directories with ext3.
> But these cases should be very rare. How often would it happen to store
> million
Hello,
How big? ext3 STILL only supports 32000 directories within a directory.
That gets to be quite a problem on large installs.
it is true: Performance goes down on very large directories with ext3.
But these cases should be very rare. How often would it happen to store
millions of files
Hello
Really? Nice. But you would have to re-create the ext3 filesystem with
this newer 2.6 kernel, right? Or would tune2fs do the job given the
right options?
I once enabled the index-option with tune2fs on an old ext3 filesystem
at 2.4 kernel which didn't give much performance boost. But pa
> You can contain the fanout by enabling both fulldirhash and
> hashimapspool in imapd.conf (but you will need to stop the server and
> rehash everything if you do this on an existing system).
Of course -- but hashing only gets you so far.
> The 32000 limit applies to subdirectories, not to files
On Wed, Jul 05, 2006 at 08:48:23AM -0400, John Madden wrote:
> How big? ext3 STILL only supports 32000 directories within a directory.
> That gets to be quite a problem on large installs.
You can contain the fanout by enabling both fulldirhash and
hashimapspool in imapd.conf (but you will need t
On Tue, 2006-07-04 at 21:58 +0200, Marten Lehmann wrote:
> > I think this is because reiserfs is better handling a lot of small
> > files, while ext3 performes better with a few big files.
>
> Is it possible that you are working with a fairly old linux
> distribution? Today's ext3 shouldn't have
On Tue, Jul 04, 2006 at 09:58:50PM +0200, Marten Lehmann wrote:
> >I think this is because reiserfs is better handling a lot of small
> >files, while ext3 performes better with a few big files.
>
> Is it possible that you are working with a fairly old linux
> distribution? Today's ext3 shouldn't
Marten Lehmann wrote:
Is it possible that you are working with a fairly old linux
distribution? Today's ext3 shouldn't have these issues any longer,
otherwise Redhat wouldn't ship it as the only supported filesystem with
its Enterprise distributions. A modern 2.6er ext3 has indexes and
b-trees
I think this is because reiserfs is better handling a lot of small
files, while ext3 performes better with a few big files.
Is it possible that you are working with a fairly old linux
distribution? Today's ext3 shouldn't have these issues any longer,
otherwise Redhat wouldn't ship it as the on
Søren Schimkat wrote:
> Hi guys
>
> I'm about to migrate from Solaris with Sendmail / uw to Redhat
> Enterprise Linux with Postfix / Cyrus. Everything seems to work just
> fine, but one unsolved question remains: Which filesystem should I choose?
>
> I really would like to use ext3 .. because i
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 unt
That sounds promising, but sadly, CPU is not the problem. CPU usage I
can find with top and grepping the logfiles with the PID.
But the triggering client makes a high iowait, so load increases
dramatically while CPU usage stays low.
Best,
Daniel
Wesley Craig schrieb:
> On 03 Jul 2006, at 12:11,
On 03 Jul 2006, at 12:11, Daniel Eckl wrote:
I cannot identify the triggering client and/or the action it
starts
I have a set of patches to 2.2.x that adds per-user CPU logging. I'm
busying porting them forward to 2.3.x as we speak. Sadly, Linux
doesn't support getrusage(2) of more i
Well, I was surprised about the problem... I have only about 250
mailboxes on that server. But it's used extensively, because some users
have over 200 folders and an incredible mail count (mailboxes over 2 GB
of size).
I assume, my problem comes from Outlooks scanning all headers of a big
folder,
On 03 Jul 2006, at 11:35, Daniel Eckl wrote:
At this time I didn't change the scheduler. In fact, I actually
learned
about it a few days ago... So it was SuSE 9.3 default (might be
vanilla
default, don't know)
I think "anticipatory" is the default IO scheduler for SuSE 9.3 (from
Google).
At this time I didn't change the scheduler. In fact, I actually learned
about it a few days ago... So it was SuSE 9.3 default (might be vanilla
default, don't know)
What would you recommend for cyrus? (or for ext3 running cyrus)? And are
there prefered scheduler for other fs, too, especially reise
On Jul 3, 2006, at 9:44 AM, Daniel Eckl wrote:
I had this setup with ext3 before and I had severe problems with
extreme
load through high iowait and the system hangs for up to 20 seconds
while
the filesystem flushed unwritten data to disk.
This was on a SCSI hardware RAID 5.
Which IO schedu
I had this setup with ext3 before and I had severe problems with extreme
load through high iowait and the system hangs for up to 20 seconds while
the filesystem flushed unwritten data to disk.
This was on a SCSI hardware RAID 5.
At the moment I have a temporary machine running the same system with
26 matches
Mail list logo
