public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Chris Shoemaker <c.shoemaker@cox.net>
To: "Theodore Ts'o" <tytso@mit.edu>,
	Mark Watts <m.watts@eris.qinetiq.com>,
	Linux Kernel Mail List <linux-kernel@vger.kernel.org>
Subject: Re: ftp.kernel.org
Date: Fri, 28 May 2004 15:08:09 -0400	[thread overview]
Message-ID: <20040528190809.GB20758@cox.net> (raw)
In-Reply-To: <20040528150119.GE18449@thunk.org>

On Fri, May 28, 2004 at 11:01:19AM -0400, Theodore Ts'o wrote:
> On Fri, May 28, 2004 at 02:21:41AM -0400, Chris Shoemaker wrote:
> > > Agreed - fmirror is so much more reliable than rsync (imho) that it makes 
> > > rsync into a worst-case option for retrieving files.
> > 
> >   bug reports to rsync@lists.samba.org are appreciated...
> > 
> 
> The main problem with rsync that I can see is that it is fairly heavy
> weight on the server, so many servers (including mirrors.kernel.org)
> have a maximum number of connections set to something pathetically
> small, like say 5 connections.  

Do you mean w.r.t. memory usage or storage i/o?  I know that creating
the file list can take up a lot of memory for large lists.  

Five connections seems pretty low.  I've never personally hit any
connection limit, and I make moderate use of rsync.  On the server side 
I know of several rsync servers offering >1 million files.  Not sure how 
hard they work, but they're highly available.

> 
> I remember Tridge trying to get someone to implement checksum caching
> for rsync servers some 4+ years ago, which would surely help.  Did
> that ever get done?  If so, convincing the server admins that it's OK
> to up the maximum number of rsync connections would be the next step.
> 
> 						- Ted

I'm sure there are some things that can be done to make rsync 
lighter-weight.  Checksums aren't cached, but the problem is, the 
checksum seed is varible, so that might not be the best optimization.

Overall, I'd have to disagree with the parent-post saying that rsync is
worst-case option.  It's not perfect, but I much prefer rsync over
fmirror.  I think it's more convenient and, although I have no rigorous
measurements, but it seems faster, to boot.

-chris

  parent reply	other threads:[~2004-05-29  0:02 UTC|newest]

Thread overview: 171+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-05-28  1:55 ftp.kernel.org Ricky Beam
2004-05-28  2:29 ` ftp.kernel.org Martin J. Bligh
2004-05-28  4:21   ` ftp.kernel.org Ricky Beam
2004-05-28  8:41     ` ftp.kernel.org Mark Watts
2004-05-28  6:21       ` ftp.kernel.org Chris Shoemaker
2004-05-28 15:01         ` ftp.kernel.org Theodore Ts'o
2004-05-28 16:32           ` ftp.kernel.org Andreas Dilger
2004-05-29 15:30             ` ftp.kernel.org Daniel Egger
2004-05-30  0:29               ` ftp.kernel.org H. Peter Anvin
2004-05-30  9:43                 ` ftp.kernel.org Andrew Walrond
2004-06-13 22:22                 ` ftp.kernel.org Pedro Larroy
2004-05-28 19:08           ` Chris Shoemaker [this message]
2004-05-28 22:15           ` ftp.kernel.org H. Peter Anvin
2004-05-28  8:55       ` ftp.kernel.org Jan-Benedict Glaw
2004-05-28 11:10         ` ftp.kernel.org Mark Watts
2004-05-28 13:57           ` ftp.kernel.org Keith Owens
2004-05-28 22:16           ` ftp.kernel.org H. Peter Anvin
2004-05-29 21:05         ` ftp.kernel.org Horst von Brand
2004-05-30  6:52           ` ftp.kernel.org H. Peter Anvin
2004-05-28 22:17 ` ftp.kernel.org H. Peter Anvin
  -- strict thread matches above, loose matches on Subject: below --
2004-06-18  0:09 Stop the Linux kernel madness 4Front Technologies
2004-06-18  0:20 ` Kyle McMartin
2004-06-18 14:51   ` Jesper Juhl
2004-06-18 14:54     ` Jesper Juhl
2004-06-19  2:36     ` Horst von Brand
2004-06-18  0:21 ` Martin J. Bligh
2004-06-18  0:27   ` 4Front Technologies
2004-06-18  0:31     ` Christoph Hellwig
2004-06-18  5:51     ` Martin J. Bligh
2004-06-18 10:39     ` Bernd Petrovitsch
2004-06-18 15:48     ` Andreas Gruenbacher
2004-06-18 17:30       ` Martin Schlemmer
2004-06-18 17:53       ` 4Front Technologies
2004-06-18 18:28         ` Timothy Miller
2004-06-18 18:23           ` 4Front Technologies
2004-06-18 18:54             ` Kyle Moffett
2004-06-18 20:31               ` Hannu Savolainen
2004-06-18 21:37                 ` Kyle Moffett
2004-06-18 19:08             ` Valdis.Kletnieks
2004-06-18 19:43             ` Timothy Miller
2004-06-18 19:02         ` Andreas Dilger
2004-06-18 20:12           ` 4Front Technologies
2004-06-18 19:36             ` Andreas Dilger
2004-06-18 19:40             ` Valdis.Kletnieks
2004-06-18 20:29               ` David Lang
2004-06-18 20:54                 ` Valdis.Kletnieks
2004-06-20 12:56                 ` Hannu Savolainen
2004-06-20 22:11                   ` David Lang
2004-06-21  1:16                     ` 4Front Technologies
2004-06-21  7:07                       ` Hannu Savolainen
2004-06-21  7:25                         ` Xavier Bestel
2004-06-21  8:27                           ` Hannu Savolainen
2004-06-21 13:12                             ` Denis Vlasenko
2004-06-22  2:06                       ` Andrea Arcangeli
2004-06-22  7:54                         ` Hannu Savolainen
2004-06-22 11:19                           ` Denis Vlasenko
2004-06-22 16:48                             ` 4Front Technologies
2004-06-22 17:46                           ` V13
2004-06-19 16:35       ` Jari Ruusu
2004-06-18 20:46     ` Sam Ravnborg
2004-06-18 20:59       ` 4Front Technologies
2004-06-18 22:59         ` Thomas Gleixner
2004-06-19 16:36       ` Jari Ruusu
2004-06-19 18:01         ` Roman Zippel
2004-06-19 21:12           ` Martin Schlemmer
2004-06-20 15:43           ` Jari Ruusu
2004-06-18  0:39 ` Andrew Morton
2004-06-18  1:29   ` Nicholas S. Wourms
2004-06-18  8:27   ` Jens Axboe
2004-06-18 14:43     ` Rik van Riel
2004-06-18 15:13       ` William Lee Irwin III
2004-06-18 15:33         ` Jan-Benedict Glaw
2004-06-18 17:17           ` Paul Jakma
2004-06-18 18:24             ` Jan-Benedict Glaw
2004-06-18 19:02           ` Tim Bird
2004-06-18 19:45             ` Timothy Miller
2004-06-18 19:46               ` Jan-Benedict Glaw
2004-06-18 20:05             ` Rik van Riel
2004-06-18 20:08               ` Jan-Benedict Glaw
2004-06-18 21:03                 ` jsimmons
2004-06-18 21:10                   ` Jan-Benedict Glaw
2004-06-18 21:13                     ` Jens Axboe
2004-06-18 21:38                       ` Jan-Benedict Glaw
2004-06-18 23:18                         ` jsimmons
2004-06-19  8:19                           ` Jan-Benedict Glaw
2004-06-19  3:34                         ` Horst von Brand
2004-06-22 13:24                           ` Jan-Benedict Glaw
2004-06-18 22:05                   ` viro
2004-06-18 22:10                     ` Jan-Benedict Glaw
2004-06-19 12:42                   ` Francois Romieu
2004-06-19 13:55                     ` Jan-Benedict Glaw
2004-06-20 10:21                       ` Geert Uytterhoeven
2004-06-22 19:34                     ` jsimmons
2004-06-18 20:20               ` Tim Bird
2004-06-18 20:50                 ` Jan-Benedict Glaw
2004-06-18 20:05             ` Jan-Benedict Glaw
2004-06-18 20:03           ` Rik van Riel
2004-06-19 12:09           ` John Jasen
2004-06-19 12:29             ` lkml
     [not found]       ` <2c0942db040618100264ea6b7d@mail.gmail.com>
2004-06-18 17:27         ` Ray Lee
2004-06-18 20:51       ` Andrew Morton
2004-06-18 21:03         ` Rik van Riel
2004-06-18 21:26           ` Jan-Benedict Glaw
2004-06-18 21:17         ` Jens Axboe
2004-06-19 20:59           ` Lars Marowsky-Bree
2004-06-18  0:44 ` viro
2004-06-18  1:00   ` 4Front Technologies
2004-06-18  1:20     ` Roman Zippel
2004-06-18  1:33       ` 4Front Technologies
2004-06-18 10:12         ` Roman Zippel
2004-06-18 17:37           ` Hannu Savolainen
2004-06-18 20:26             ` Roman Zippel
2004-06-18 21:52             ` Jeff Garzik
2004-06-18 22:26               ` Matt Domsch
2004-06-18 23:13                 ` 4Front Technologies
2004-06-19  8:44               ` Hannu Savolainen
2004-06-19  2:59             ` Horst von Brand
2004-06-18  9:57       ` Petr Vandrovec
2004-06-18 13:47         ` Olaf Hering
2004-06-18 14:03           ` Petr Vandrovec
2004-06-18 14:59             ` Duncan Sands
2004-06-18  1:20     ` viro
2004-06-18  1:25     ` Thomas Gleixner
2004-06-18  1:31     ` Nick Piggin
2004-06-18  1:37       ` 4Front Technologies
2004-06-18  1:41         ` Nick Piggin
2004-06-18  2:17         ` Erik Harrison
2004-06-18  7:45         ` John Bradford
2004-06-18  9:05           ` Jan-Benedict Glaw
2004-06-18  1:34     ` Bernd Eckenfels
2004-06-18 10:51       ` Giuseppe Bilotta
2004-06-18  8:48     ` Flavio Stanchina
2004-06-18 10:25     ` Matthias Andree
2004-06-20  5:28       ` Ryan Anderson
2004-06-20  8:13         ` Jaroslav Kysela
2004-06-22 15:17         ` Matthias Andree
2004-06-22 18:22           ` Bill Davidsen
2004-06-18 15:42     ` Timothy Miller
2004-06-18  1:07 ` thinkliberty
2004-06-18  1:12   ` 4Front Technologies
2004-06-18  1:21     ` CaT
2004-06-18  1:37     ` Bastiaan Spandaw
2004-06-18  1:46       ` 4Front Technologies
2004-06-18  2:06         ` Thomas Gleixner
     [not found]           ` <40D259DD.6020604@opensound.com>
2004-06-18 22:53             ` Thomas Gleixner
2004-06-18  5:54         ` Jan-Benedict Glaw
2004-06-18  8:50         ` Helge Hafting
2004-06-18 11:05         ` Redeeman
2004-06-18 15:25         ` Jeff Garzik
2004-06-18 15:51         ` Timothy Miller
2004-06-18 16:53           ` Alex Goddard
2004-06-18 15:46     ` Timothy Miller
2004-06-18  6:40 ` Arjan van de Ven
2004-06-18  8:10 ` Clemens Schwaighofer
2004-06-18 15:34 ` Timothy Miller
     [not found] ` <mailman.1087541100.18231.linux-kernel2news@redhat.com>
2004-06-18 19:47   ` Pete Zaitcev
2004-06-19 16:35     ` Jari Ruusu
2004-06-19 20:52       ` Lars Marowsky-Bree
2004-06-20 15:43         ` Jari Ruusu
2004-06-20 20:44           ` Andreas Gruenbacher
2006-10-08  4:43 2.6.19-rc1: known regressions (v2) Trond Myklebust
2006-10-08  4:55 ` Adrian Bunk
     [not found]   ` <1160283948.10192.3.camel@lade.trondhjem.org>
2006-10-08  6:39     ` Adrian Bunk
2006-10-08  7:45       ` Pekka Enberg
2006-10-08 17:28         ` Adrian Bunk
2006-10-08 17:34           ` Jan-Benedict Glaw
2006-10-08 17:59             ` Adrian Bunk
2006-10-08 18:04               ` Jan-Benedict Glaw
2006-10-08 18:15                 ` Adrian Bunk
2006-10-08 18:22                   ` Jan-Benedict Glaw
2006-10-09  1:49             ` Horst H. von Brand

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=20040528190809.GB20758@cox.net \
    --to=c.shoemaker@cox.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=m.watts@eris.qinetiq.com \
    --cc=tytso@mit.edu \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox