Debian Bug report logs -
#350580
slot allocation no longer works around shitty peers
Reported by: Robert Millan <rmh@aybabtu.com>
Date: Mon, 30 Jan 2006 14:33:02 UTC
Severity: normal
Tags: upstream
Found in version amule/2.1.0-1
Done: Sandro Tosi <morph@debian.org>
Bug is archived. No further changes may be made.
Toggle useless messages
Report forwarded to debian-bugs-dist@lists.debian.org, Julien Delange <julien@gunnm.org>:
Bug#350580; Package amule.
(full text, mbox, link).
Acknowledgement sent to Robert Millan <rmh@aybabtu.com>:
New Bug report received and forwarded. Copy sent to Julien Delange <julien@gunnm.org>.
(full text, mbox, link).
Message #5 received at submit@bugs.debian.org (full text, mbox, reply):
Package: amule
Version: 2.1.0-1
Severity: normal
After upgrading to 2.1.0, the slot allocation for upload connections stopped
working properly.
It often happens that clients who are assigned one of my slots to download
my files, have either a shitty or very busy connection, resulting in transfer
rate drop. Sometimes the drop is to 0.0 kB/s [1].
Then amule tries to compensate this by assigning more bandwidth to the other
slots.
When _all_ my slots are in this situation (i.e. assigned to shitty peers), amule
has to take a decision wether a new slot has to be added.
- amule versions prior to 2.1.0 added a new slot (I think) if the amount of
used bandwidth was inferior to the upload_bandwidth_limit. I think the code
looks like:
if (slot_allocation * slots_in_use < upload_bandwidth_limit)
[add new slot]
- amule 2.1.0 seems to have a different criteria. I'm not sure which it is
excatly, but it's a very bad one, because it doesn't add a new slot in
situations where it's badly needed. I suspect it's doing something like:
if ((slot_allocation + 1) * slots_in_use < upload_bandwidth_limit)
[add new slot]
My workaround for this problem is setting the slot_allocation parameter to be
1/4 of the upload_bandwidth_limit. This way _usually_ non-broken clients come
to the rescue and compensate the loss caused by broken ones. However, for
resource control reasons I would prefer setting slot_allocation to be as close
as possible to upload_bandwidth_limit (1/2 or maybe even 1/1).
[1] Btw, when this happens, I would suggest banning that client as it's just
wasting overhead bandwidth unnecessarily.
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable'), (500, 'testing')
Architecture: i386 (i686)
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.12-1-k7
Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ANSI_X3.4-1968) (ignored: LC_ALL set to C)
Versions of packages amule depends on:
hi amule-common 2.1.0-1 common files for aMule
ii libcrypto++5.2c2a 5.2.1c2a-2 General purpose cryptographic shar
ii libgcc1 1:4.0.2-5 GCC support library
ii libstdc++6 4.0.2-5 The GNU Standard C++ Library v3
ii libwxgtk2.6-0 2.6.1.2 wxWidgets Cross-platform C++ GUI t
ii zlib1g 1:1.2.3-9 compression library - runtime
Versions of packages amule recommends:
pn amule-daemon <none> (no description available)
pn amule-utils <none> (no description available)
-- no debconf information
Tags added: upstream
Request was from Julien Delange <soda@yoko.gunnm.org>
to control@bugs.debian.org.
(full text, mbox, link).
Message sent on
to Robert Millan <rmh@aybabtu.com>:
Bug#350580.
(Wed, 30 Dec 2009 23:36:13 GMT) (full text, mbox, link).
Message #10 received at 350580-submitter@bugs.debian.org (full text, mbox, reply):
Hello,
we recently uploaded in Debian unstable the new upstream release of
aMule: 2.2.6 .
We are contacting you because you reported a bug against aMule and
because we'd like to ask you to verify if your bug still applies to
the 2.2.6 version.
Since some of the bugs we are contacting are really old, there are
chances they were already fixed in a previous version (just for the
reference, Debian stable has version 2.2.1 while Debian testing has
(currently) version 2.2.5).
Your help is appreciated in screening your bug, so we close those that
are already fixed and address those that are not (yet).
Thanks in advance,
Sandro (on behalf of aMule Debian packaging team)
--
Sandro Tosi (aka morph, morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi
Information forwarded
to debian-bugs-dist@lists.debian.org, Debian aMule Team <pkg-amule-devel@lists.alioth.debian.org>:
Bug#350580; Package amule.
(Wed, 30 Dec 2009 23:39:04 GMT) (full text, mbox, link).
Acknowledgement sent
to Sandro Tosi <morph@debian.org>:
Extra info received and forwarded to list. Copy sent to Debian aMule Team <pkg-amule-devel@lists.alioth.debian.org>.
(Wed, 30 Dec 2009 23:39:05 GMT) (full text, mbox, link).
Reply sent
to Sandro Tosi <morph@debian.org>:
You have taken responsibility.
(Sun, 11 Jul 2010 13:18:03 GMT) (full text, mbox, link).
Notification sent
to Robert Millan <rmh@aybabtu.com>:
Bug acknowledged by developer.
(Sun, 11 Jul 2010 13:18:04 GMT) (full text, mbox, link).
Message #20 received at 350580-done@bugs.debian.org (full text, mbox, reply):
Hello,
these bugs are long-standing and received no reply after a ping
following 2.2.6 upload, hence I'm closing them now: please reopen if
some of them still apply.
Regards,
--
Sandro Tosi (aka morph, morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi
Bug archived.
Request was from Debbugs Internal Request <owner@bugs.debian.org>
to internal_control@bugs.debian.org.
(Mon, 09 Aug 2010 07:36:23 GMT) (full text, mbox, link).
Send a report that this bug log contains spam.
Debian bug tracking system administrator <owner@bugs.debian.org>.
Last modified:
Sun Jul 2 07:17:14 2023;
Machine Name:
bembo
Debian Bug tracking system
Debbugs is free software and licensed under the terms of the GNU
Public License version 2. The current version can be obtained
from https://bugs.debian.org/debbugs-source/.
Copyright © 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson,
2005-2017 Don Armstrong, and many other contributors.