From bsdfan666 at cyberdude.com  Sun Jan  4 23:42:28 2009
From: bsdfan666 at cyberdude.com (BSD Fanatic)
Date: Sun, 4 Jan 2009 17:42:28 -0500
Subject: Why has this project died?
Message-ID: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>

Does nobody care about it? really.. somebody save btpd! no other client 
comes close to the quality.. it needs someone willing to modernize it!

Please. Somebody.. HELP! :-(

-- 
Be Yourself @ mail.com!
Choose From 200+ Email Addresses
Get a Free Account at www.mail.com


From DylanV at semaphore.com  Mon Jan  5 06:33:39 2009
From: DylanV at semaphore.com (Dylan Vanderhoof)
Date: Sun, 4 Jan 2009 21:33:39 -0800
Subject: Why has this project died?
In-Reply-To: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
References: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
Message-ID: <EAC75A12BE260246A8145B73D26130870978AE01F4@exchange00.semaphore.lan>

Richard was theoretically doing some infrastructure upgrade to it around October that would allow newer features to be added more easily, but that's the last I've seen from him on this list.

Drag too, I agree, it's the best console client out there.

-D

-----Original Message-----
From: btpd-users-bounces at murmeldjur.se [mailto:btpd-users-bounces at murmeldjur.se] On Behalf Of BSD Fanatic
Sent: Sunday, January 04, 2009 2:42 PM
To: btpd-users at murmeldjur.se
Subject: Why has this project died?

Does nobody care about it? really.. somebody save btpd! no other client
comes close to the quality.. it needs someone willing to modernize it!

Please. Somebody.. HELP! :-(

--
Be Yourself @ mail.com!
Choose From 200+ Email Addresses
Get a Free Account at www.mail.com

_______________________________________________
btpd-users mailing list
btpd-users at murmeldjur.se
http://lists.stargirl.org/listinfo/btpd-users

From goatspice at gmail.com  Mon Jan  5 09:30:50 2009
From: goatspice at gmail.com (Yassen Roussev)
Date: Mon, 5 Jan 2009 03:30:50 -0500
Subject: Why has this project died?
In-Reply-To: <EAC75A12BE260246A8145B73D26130870978AE01F4@exchange00.semaphore.lan>
References: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
	<EAC75A12BE260246A8145B73D26130870978AE01F4@exchange00.semaphore.lan>
Message-ID: <34becb040901050030v3fce4b99m6ace09fab6dfb2c7@mail.gmail.com>

hi all,

yes, I'd be very keen to see btpd resureected too, as it is The Best by far.

Happy New Year to Richard and everybody on the list and may it bring us a
new btpd release ;-)

Yassen

2009/1/5 Dylan Vanderhoof <DylanV at semaphore.com>

> Richard was theoretically doing some infrastructure upgrade to it around
> October that would allow newer features to be added more easily, but that's
> the last I've seen from him on this list.
>
> Drag too, I agree, it's the best console client out there.
>
> -D
>
> -----Original Message-----
> From: btpd-users-bounces at murmeldjur.se [mailto:
> btpd-users-bounces at murmeldjur.se] On Behalf Of BSD Fanatic
> Sent: Sunday, January 04, 2009 2:42 PM
> To: btpd-users at murmeldjur.se
> Subject: Why has this project died?
>
> Does nobody care about it? really.. somebody save btpd! no other client
> comes close to the quality.. it needs someone willing to modernize it!
>
> Please. Somebody.. HELP! :-(
>
> --
> Be Yourself @ mail.com!
> Choose From 200+ Email Addresses
> Get a Free Account at www.mail.com
>
> _______________________________________________
> btpd-users mailing list
> btpd-users at murmeldjur.se
> http://lists.stargirl.org/listinfo/btpd-users
> _______________________________________________
> btpd-users mailing list
> btpd-users at murmeldjur.se
> http://lists.stargirl.org/listinfo/btpd-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.stargirl.org/pipermail/btpd-users/attachments/20090105/df98bb98/attachment.html 

From per at computer.org  Mon Jan  5 09:34:14 2009
From: per at computer.org (Per Jessen)
Date: Mon, 05 Jan 2009 09:34:14 +0100
Subject: Why has this project died?
In-Reply-To: <EAC75A12BE260246A8145B73D26130870978AE01F4@exchange00.semaphore.lan>
References: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
	<EAC75A12BE260246A8145B73D26130870978AE01F4@exchange00.semaphore.lan>
Message-ID: <4961C606.7030801@computer.org>

Dylan Vanderhoof wrote:
> Richard was theoretically doing some infrastructure upgrade to it around October that 
 >would allow newer features to be added more easily, but that's the 
last I've seen from him on this list.
> 
> Drag too, I agree, it's the best console client out there.
> 
> -D

I second that.


/Per Jessen, Zurich

From royalbee at gmail.com  Mon Jan  5 10:48:23 2009
From: royalbee at gmail.com (Eyal Bari)
Date: Mon, 5 Jan 2009 11:48:23 +0200
Subject: Why has this project died?
In-Reply-To: <4961C606.7030801@computer.org>
References: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
	<EAC75A12BE260246A8145B73D26130870978AE01F4@exchange00.semaphore.lan>
	<4961C606.7030801@computer.org>
Message-ID: <44f4144c0901050148r5310ab9dp76bd22742ed00d4a@mail.gmail.com>

me third

On Mon, Jan 5, 2009 at 10:34 AM, Per Jessen <per at computer.org> wrote:

> Dylan Vanderhoof wrote:
> > Richard was theoretically doing some infrastructure upgrade to it around
> October that
>  >would allow newer features to be added more easily, but that's the
> last I've seen from him on this list.
> >
> > Drag too, I agree, it's the best console client out there.
> >
> > -D
>
> I second that.
>
>
> /Per Jessen, Zurich
> _______________________________________________
> btpd-users mailing list
> btpd-users at murmeldjur.se
> http://lists.stargirl.org/listinfo/btpd-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.stargirl.org/pipermail/btpd-users/attachments/20090105/1ab94fc0/attachment.htm 

From rnyberg at murmeldjur.se  Mon Jan  5 13:26:26 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Mon, 5 Jan 2009 13:26:26 +0100 (CET)
Subject: Why has this project died?
In-Reply-To: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
References: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
Message-ID: <2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>

On Sun, January 4, 2009 23:42, BSD Fanatic wrote:
> Does nobody care about it? really.. somebody save btpd! no other client
> comes close to the quality.. it needs someone willing to modernize it!
>
> Please. Somebody.. HELP! :-(
>
Relax! It's not dead. Honest! Really!

Unfortunately I kind of took my development in a bad direction and I've
wasted a lot of effort, which got me a little bit tired of hacking on
btpd. Also real life and other projects have competed for time. They
still are, but I'm back on track again and am making progress. I think
a new release in February is very probable. :)

Btw, I've switched from subversion to git, since it fits me much better,
so the repo on the site is not up to date. I'll have to fix the site and
put up my git repo some time.

Regards,
        -Richard



From nemosoft at smcc.demon.nl  Mon Jan  5 16:13:09 2009
From: nemosoft at smcc.demon.nl (Nemosoft Unv.)
Date: Mon, 5 Jan 2009 16:13:09 +0100
Subject: Why has this project died?
In-Reply-To: <2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>
References: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
	<2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>
Message-ID: <200901051613.09673.nemosoft@smcc.demon.nl>

Hi,

On Monday 05 January 2009 13:26:26 Richard Nyberg wrote:
> On Sun, January 4, 2009 23:42, BSD Fanatic wrote:
> > Does nobody care about it? really.. somebody save btpd! no other client
> > comes close to the quality.. it needs someone willing to modernize it!
> >
> > Please. Somebody.. HELP! :-(
>
> Relax! It's not dead. Honest! Really!

Good to hear it! Happy new year, btw :-)

>> Unfortunately I kind of took my development in a bad direction and I've
> wasted a lot of effort, which got me a little bit tired of hacking on
> btpd. 

A bit of advice: if you happen to get stuck in a certain direction, consider 
abandoning the changes. Go back a few revisions, or take the latest official 
release, and work from there. I know from experience that sometimes, halfway 
through your coding, you realize that what you had in mind is not going to 
work. Then it's sometimes more efficient to abandon your changes and redo the 
things you want to do, in stead of trying to hammer the not-so-well working 
stuff in something you want. 

> Also real life and other projects have competed for time. They 
> still are, but I'm back on track again and am making progress. I think
> a new release in February is very probable. :)

Okay, nice to hear. Can you give us a hint of what we can expect?

> Btw, I've switched from subversion to git, since it fits me much better,
> so the repo on the site is not up to date. I'll have to fix the site and
> put up my git repo some time.

Funny, I'm busy switching from CVS to subversion :-)

 - Nemosoft


From DylanV at semaphore.com  Mon Jan  5 18:39:21 2009
From: DylanV at semaphore.com (Dylan Vanderhoof)
Date: Mon, 5 Jan 2009 09:39:21 -0800
Subject: Why has this project died?
In-Reply-To: <2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>
Message-ID: <EAC75A12BE260246A8145B73D261308709788DA4E7@exchange00.semaphore.lan>

Woo!  Good to hear you chime in!

Looking forward to the new release when it rolls around.

-D

> -----Original Message-----
> From: btpd-users-bounces at murmeldjur.se
> [mailto:btpd-users-bounces at murmeldjur.se] On Behalf Of Richard Nyberg
> Sent: Monday, January 05, 2009 4:26 AM
> To: btpd general discussion.
> Subject: Re: Why has this project died?
>
>
> On Sun, January 4, 2009 23:42, BSD Fanatic wrote:
> > Does nobody care about it? really.. somebody save btpd! no
> other client
> > comes close to the quality.. it needs someone willing to
> modernize it!
> >
> > Please. Somebody.. HELP! :-(
> >
> Relax! It's not dead. Honest! Really!
>
> Unfortunately I kind of took my development in a bad
> direction and I've
> wasted a lot of effort, which got me a little bit tired of hacking on
> btpd. Also real life and other projects have competed for time. They
> still are, but I'm back on track again and am making progress. I think
> a new release in February is very probable. :)
>
> Btw, I've switched from subversion to git, since it fits me
> much better,
> so the repo on the site is not up to date. I'll have to fix
> the site and
> put up my git repo some time.
>
> Regards,
>         -Richard
>
>
> _______________________________________________
> btpd-users mailing list
> btpd-users at murmeldjur.se
> http://lists.stargirl.org/listinfo/btpd-users
>

From axelgenus at gmail.com  Mon Jan  5 19:21:22 2009
From: axelgenus at gmail.com (=?UTF-8?Q?Alessandro_Calor=C3=AC?=)
Date: Mon, 5 Jan 2009 19:21:22 +0100
Subject: Why has this project died?
In-Reply-To: <2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>
References: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
	<2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>
Message-ID: <eeb7df810901051021k67b06f68y13fbd16726e17b80@mail.gmail.com>

Hi Richard,

2009/1/5 Richard Nyberg <rnyberg at murmeldjur.se>:
> I think a new release in February is very probable. :)

Wow... so unexpected! Can you tell us more about the next release?

> Btw, I've switched from subversion to git, since it fits me much better,
> so the repo on the site is not up to date. I'll have to fix the site and
> put up my git repo some time.

Good to know. ;)

> Regards,
>        -Richard

Bye, Alessandro.

From borg at uu3.net  Tue Jan  6 11:07:02 2009
From: borg at uu3.net (Unknown)
Date: Tue, 6 Jan 2009 11:07:02 +0100 (CET)
Subject: Why has this project died?
In-Reply-To: <2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>
References: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
	<2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>
Message-ID: <Pine.LNX.4.64.0901061105050.25122@cube>



On Mon, 5 Jan 2009, Richard Nyberg wrote:
> Btw, I've switched from subversion to git, since it fits me much better,
> so the repo on the site is not up to date. I'll have to fix the site and
> put up my git repo some time.
Git.. Git.. I really would like to check it. But Torvalds is Linuxish
asshole :) Git is not so portable as it should be..

Funny is that subversion guys went in bad direction too..
1.4.6 was very portable.. 1.5.x branch is b0rked..

So.. Im stick to 1.4.6 since it fits my current needs.

From goatspice at gmail.com  Tue Jan  6 11:24:33 2009
From: goatspice at gmail.com (Yassen Roussev)
Date: Tue, 6 Jan 2009 10:24:33 +0000
Subject: Why has this project died?
In-Reply-To: <2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>
References: <20090104224228.C4D5DBE502C@ws1-9.us4.outblaze.com>
	<2442.85.24.200.41.1231158386.squirrel@webmail.stargirl.org>
Message-ID: <34becb040901060224g3cbea877w22ea779dc8057416@mail.gmail.com>

2009/1/5 Richard Nyberg <rnyberg at murmeldjur.se>

> On Sun, January 4, 2009 23:42, BSD Fanatic wrote:
> > Does nobody care about it? really.. somebody save btpd! no other client
> > comes close to the quality.. it needs someone willing to modernize it!
> >
> > Please. Somebody.. HELP! :-(
> >
> Relax! It's not dead. Honest! Really!



Brilliant news!

Looking forward to it!

Y.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.stargirl.org/pipermail/btpd-users/attachments/20090106/49b2d6a5/attachment.htm 

From rnyberg at murmeldjur.se  Wed Jan 14 00:58:56 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Wed, 14 Jan 2009 00:58:56 +0100 (CET)
Subject: git repo and stuff
Message-ID: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>

Hi users!

I've made a public git repo available at [1] so go ahead and clone it
if you're interested. I'll try to update the site soon, but it's not
my highest priority atm.

Anyway, main points so far are IPv6 support, replacing libevent with
my own event loop and solaris support.

btpd now uses both ipv4 and ipv6 by default. -4 and -6 exists to toggle
use of the respective versions. Ie. 'btpd -6' will only use ipv4. Any
thoughts on that?

Solaris support required some hacking but it's nice to have at least.

I've never really liked libevent. It's hard to have complete control of
it from an application, so I rolled my own simpler (from btpd's point
of view) event loop. Atm it can use kqueue, epoll or poll. If someone
wants to add support for /dev/poll please do so. :)

I've made so the initial process will wait until the daemon has started
(or failed to start) and returns appropriate error codes. Ie. you can
now do "btpd && echo it really started".

I intend to clean up the shutdown so that a btpd process just trying
to send stop messages to trackers before exiting doesn't use the btpd
dir. That is: "btcli kill && btpd" should be unproblematic. When that's
done it's probably btpd-0.14 time. :)

Regards,
        -Richard

[1] http://www.murmeldjur.se/btpd/btpd.git



From graue at oceanbase.org  Wed Jan 14 06:01:43 2009
From: graue at oceanbase.org (Scott Feeney)
Date: Wed, 14 Jan 2009 00:01:43 -0500
Subject: Bad data from peers (was Re: git repo and stuff)
In-Reply-To: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
Message-ID: <dqsrdPlG.1231909303.9770570.graue@pendrell.textdrive.com>

Richard,

I cloned the git repo to take a look. I'm really glad you're working on
btpd again.

Don't know if this is already on your todo list, but for me the most
urgent feature improvement would be banning peers who send bad data.
I've found torrents where the fastest uploaders seem to be sending
exclusively bad data, in some cases making it impractical to download
due to well over 200% overhead.

i.e., for every 1 MB of good data, I was downloading 2 MB of bad data.
Yes, this has happened to me, and then some...

Scott



On 1/13/2009, "Richard Nyberg" <rnyberg at murmeldjur.se> wrote:

>Hi users!
>
>I've made a public git repo available at [1] so go ahead and clone it
>if you're interested. I'll try to update the site soon, but it's not
>my highest priority atm.
>
>Anyway, main points so far are IPv6 support, replacing libevent with
>my own event loop and solaris support.
>
>btpd now uses both ipv4 and ipv6 by default. -4 and -6 exists to toggle
>use of the respective versions. Ie. 'btpd -6' will only use ipv4. Any
>thoughts on that?
>
>Solaris support required some hacking but it's nice to have at least.
>
>I've never really liked libevent. It's hard to have complete control of
>it from an application, so I rolled my own simpler (from btpd's point
>of view) event loop. Atm it can use kqueue, epoll or poll. If someone
>wants to add support for /dev/poll please do so. :)
>
>I've made so the initial process will wait until the daemon has started
>(or failed to start) and returns appropriate error codes. Ie. you can
>now do "btpd && echo it really started".
>
>I intend to clean up the shutdown so that a btpd process just trying
>to send stop messages to trackers before exiting doesn't use the btpd
>dir. That is: "btcli kill && btpd" should be unproblematic. When that's
>done it's probably btpd-0.14 time. :)
>
>Regards,
>        -Richard
>
>[1] http://www.murmeldjur.se/btpd/btpd.git
>
>
>_______________________________________________
>btpd-users mailing list
>btpd-users at murmeldjur.se
>http://lists.stargirl.org/listinfo/btpd-users

From icepic.dz at gmail.com  Wed Jan 14 08:27:38 2009
From: icepic.dz at gmail.com (Janne Johansson)
Date: Wed, 14 Jan 2009 08:27:38 +0100
Subject: git repo and stuff
In-Reply-To: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
Message-ID: <a10961760901132327r726ab493t73041290b9aca643@mail.gmail.com>

2009/1/14 Richard Nyberg <rnyberg at murmeldjur.se>

> Hi users!
>
> btpd now uses both ipv4 and ipv6 by default. -4 and -6 exists to toggle
> use of the respective versions. Ie. 'btpd -6' will only use ipv4. Any
> thoughts on that?
>

Last sentence came out bad. But yay for v6 support anyhow!

-- 
To our sweethearts and wives.  May they never meet. -- 19th century toast
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.stargirl.org/pipermail/btpd-users/attachments/20090114/2de6f277/attachment.htm 

From per at computer.org  Wed Jan 14 08:36:22 2009
From: per at computer.org (Per Jessen)
Date: Wed, 14 Jan 2009 08:36:22 +0100
Subject: git repo and stuff
In-Reply-To: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
Message-ID: <496D95F6.6060506@computer.org>

Richard Nyberg wrote:

> btpd now uses both ipv4 and ipv6 by default. -4 and -6 exists to toggle
> use of the respective versions. Ie. 'btpd -6' will only use ipv4. Any
> thoughts on that?

It's a little counter intuitive, I think.  Wouldn't it be better running 
in ipv6 mode by default, but having a switch (e.g. '-4') to reduce to 
ipv4 only?


/Per Jessen, Zurich

From goatspice at gmail.com  Wed Jan 14 08:53:30 2009
From: goatspice at gmail.com (Yassen Roussev)
Date: Wed, 14 Jan 2009 02:53:30 -0500
Subject: Bad data from peers (was Re: git repo and stuff)
In-Reply-To: <dqsrdPlG.1231909303.9770570.graue@pendrell.textdrive.com>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<dqsrdPlG.1231909303.9770570.graue@pendrell.textdrive.com>
Message-ID: <34becb040901132353u40688901u8aaa680bc97087b5@mail.gmail.com>

> Don't know if this is already on your todo list, but for me the most
> urgent feature improvement would be banning peers who send bad data.


Was keeping queit, due to excitement of btpd being active again, but looking
forward to one of the future releases when encryption will be available, and
defaulted to.

Best wishes,

Yassen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.stargirl.org/pipermail/btpd-users/attachments/20090114/68e3c39e/attachment.htm 

From axelgenus at gmail.com  Wed Jan 14 09:20:59 2009
From: axelgenus at gmail.com (=?UTF-8?Q?Alessandro_Calor=C3=AC?=)
Date: Wed, 14 Jan 2009 09:20:59 +0100
Subject: git repo and stuff
In-Reply-To: <496D95F6.6060506@computer.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<496D95F6.6060506@computer.org>
Message-ID: <eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>

2009/1/14 Per Jessen <per at computer.org>:
> It's a little counter intuitive, I think.  Wouldn't it be better running
> in ipv6 mode by default, but having a switch (e.g. '-4') to reduce to
> ipv4 only?

By default many tracker doesn't support IPv6 so it's better to have
default IPv4 support and a switch to enable IPv6 support.

Bye, Alessandro.

From per at computer.org  Wed Jan 14 10:00:06 2009
From: per at computer.org (Per Jessen)
Date: Wed, 14 Jan 2009 10:00:06 +0100
Subject: git repo and stuff
In-Reply-To: <eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>	<496D95F6.6060506@computer.org>
	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
Message-ID: <496DA996.7020002@computer.org>

Alessandro Calor? wrote:
> 2009/1/14 Per Jessen <per at computer.org>:
>> It's a little counter intuitive, I think.  Wouldn't it be better running
>> in ipv6 mode by default, but having a switch (e.g. '-4') to reduce to
>> ipv4 only?
> 
> By default many tracker doesn't support IPv6 so it's better to have
> default IPv4 support and a switch to enable IPv6 support.

I assumed any ipv6 support would automatically include ipv4 support.


/Per Jessen, Zurich


From axelgenus at gmail.com  Wed Jan 14 10:09:09 2009
From: axelgenus at gmail.com (=?UTF-8?Q?Alessandro_Calor=C3=AC?=)
Date: Wed, 14 Jan 2009 10:09:09 +0100
Subject: git repo and stuff
In-Reply-To: <496DA996.7020002@computer.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<496D95F6.6060506@computer.org>
	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
	<496DA996.7020002@computer.org>
Message-ID: <eeb7df810901140109k5bd5e2fas3b0dd665fa3d861f@mail.gmail.com>

2009/1/14 Per Jessen <per at computer.org>:
> I assumed any ipv6 support would automatically include ipv4 support.

On a IPv6 enabled computer with an IPv6 capable operating system the
IPv6 protocol includes IPv4 support. IPv6 BitTorrent clients (such as
uTorrent) can use the new protocol to connect to trackers and to peers
but both have to support IPv6 otherwise the connection won't be
established. IPv6 capable trackers are very few and if the tracker
doesn't support IPv6 it would send only peers' IPv4 addresses.

Bye, Alessandro.

From rnyberg at murmeldjur.se  Wed Jan 14 12:47:34 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Wed, 14 Jan 2009 12:47:34 +0100 (CET)
Subject: git repo and stuff
In-Reply-To: <496DA996.7020002@computer.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<496D95F6.6060506@computer.org>
	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
	<496DA996.7020002@computer.org>
Message-ID: <52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>

It seems we're all a little confused by this. I'm by no means
an expert on IPv6 so if you think something doesn't make sense,
please point it out.

On Wed, January 14, 2009 10:00, Per Jessen wrote:
> Alessandro Calor? wrote:
>> 2009/1/14 Per Jessen <per at computer.org>:
>>> It's a little counter intuitive, I think.  Wouldn't it be better
>>> running in ipv6 mode by default, but having a switch (e.g. '-4')
>>> to reduce to ipv4 only?
>>
>> By default many tracker doesn't support IPv6 so it's better to have
>> default IPv4 support and a switch to enable IPv6 support.

But what's the harm in having v6 enabled? If you don't disable v4 it
doesn't matter that a tracker doesn't support v6.

> I assumed any ipv6 support would automatically include ipv4 support.
>
No, they are treated separately. Having IPvX enabled in btpd
means
1) btpd listens on an IPvX socket
2) btpd will look for IPvX peers from tracker
3) btpd will be able to connect to IPvX peers and trackers

This is additive, so the default of having both IPv6 and IPv4
enabled means that btpd listens on two sockets, one for each
protocol, that btpd looks for both types of peers and is able
to connect to both type of peers and trackers.

Disabling both protocols is obviously useless, so btpd exit with
an error message at start.

I think it might make sense to change the options to have
-4 => use only ipv4 (i.e. turn off ipv6)
-6 => use only ipv6 (i.e. turn off ipv4)

Regards,
       -Richard



From rnyberg at murmeldjur.se  Wed Jan 14 13:01:05 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Wed, 14 Jan 2009 13:01:05 +0100 (CET)
Subject: Bad data from peers (was Re: git repo and stuff)
In-Reply-To: <34becb040901132353u40688901u8aaa680bc97087b5@mail.gmail.com>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<dqsrdPlG.1231909303.9770570.graue@pendrell.textdrive.com>
	<34becb040901132353u40688901u8aaa680bc97087b5@mail.gmail.com>
Message-ID: <60898.85.24.201.38.1231934465.squirrel@webmail.stargirl.org>

On Wed, January 14, 2009 08:53, Yassen Roussev wrote:
>
> Was keeping queit, due to excitement of btpd being active again, but
> looking forward to one of the future releases when encryption will
> be available, and defaulted to.
>
I really hate that stupid isp:s cause unnecessary code bloat that's
only there to nullify their restrictions. :(

However, message stream encryption will come. By default btpd will
probably not make encrypted connections but accept encrypted incoming
ones.

It'll take a little care so that the encrypted code path doesn't affect
the performance of the regular one.

        -Richard


From axelgenus at gmail.com  Wed Jan 14 13:01:45 2009
From: axelgenus at gmail.com (=?UTF-8?Q?Alessandro_Calor=C3=AC?=)
Date: Wed, 14 Jan 2009 13:01:45 +0100
Subject: git repo and stuff
In-Reply-To: <52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<496D95F6.6060506@computer.org>
	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
	<496DA996.7020002@computer.org>
	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>
Message-ID: <eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>

2009/1/14 Richard Nyberg <rnyberg at murmeldjur.se>:
> But what's the harm in having v6 enabled? If you don't disable v4 it
> doesn't matter that a tracker doesn't support v6.

That's not what I was pointing at: there's no need for an IPv4 switch
because every tracker supports it; the IPv4 socket should always be
on. Also, keep in mind that you should add a compile flag to
enable/disable IPv6 support. ;)

> they are treated separately.

Good.

> This is additive, so the default of having both IPv6 and IPv4
> enabled means that btpd listens on two sockets, one for each
> protocol, that btpd looks for both types of peers and is able
> to connect to both type of peers and trackers.

How are you going to implement the peers' queue with mixed IPv4/6 addresses?

> Disabling both protocols is obviously useless, so btpd exit with
> an error message at start.

If you leave IPv4 always on by default and you implement only an IPv6
switch this situation could be avoided.

> Regards,
>       -Richard

Bye, Alessandro.

From rnyberg at murmeldjur.se  Wed Jan 14 13:11:57 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Wed, 14 Jan 2009 13:11:57 +0100 (CET)
Subject: Bad data from peers (was Re: git repo and stuff)
In-Reply-To: <dqsrdPlG.1231909303.9770570.graue@pendrell.textdrive.com>
References: <dqsrdPlG.1231909303.9770570.graue@pendrell.textdrive.com>
Message-ID: <43258.85.24.201.38.1231935117.squirrel@webmail.stargirl.org>

Hi Scott,

On Wed, January 14, 2009 06:01, Scott Feeney wrote:
> Richard,
>
> I cloned the git repo to take a look. I'm really glad you're working on
> btpd again.
>
> Don't know if this is already on your todo list, but for me the most
> urgent feature improvement would be banning peers who send bad data.
> I've found torrents where the fastest uploaders seem to be sending
> exclusively bad data, in some cases making it impractical to download
> due to well over 200% overhead.
>
> i.e., for every 1 MB of good data, I was downloading 2 MB of bad data.
> Yes, this has happened to me, and then some...

It's actually not for lack of effort btpd doesn't already do this. I've
thought very hard on how to this in a good way. The problem is that in
the general case it's impossible to know which peer sent the bad data.

I agree that something needs to be done though, and lacking perfect
solutions, I'll see if I can find one that's 'good enough'.

        -Richard



From rnyberg at murmeldjur.se  Wed Jan 14 14:00:51 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Wed, 14 Jan 2009 14:00:51 +0100 (CET)
Subject: git repo and stuff
In-Reply-To: <eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<496D95F6.6060506@computer.org>
	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
	<496DA996.7020002@computer.org>
	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>
	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>
Message-ID: <52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>

On Wed, January 14, 2009 13:01, Alessandro Calor?? wrote:
> 2009/1/14 Richard Nyberg <rnyberg at murmeldjur.se>:
>> But what's the harm in having v6 enabled? If you don't disable v4 it
>> doesn't matter that a tracker doesn't support v6.
>
> That's not what I was pointing at: there's no need for an IPv4 switch
> because every tracker supports it; the IPv4 socket should always be
> on.

Ah. But this is not about the trackers, it's about the machine btpd
is running on. There are IPv6 or IPv4 only machines. :)
Leaving both protocols enabled should be no problem what so ever on
most machines though.

> Also, keep in mind that you should add a compile flag to
> enable/disable IPv6 support. ;)

Why? I'd rather not, unless there's a very good reason to.

> How are you going to implement the peers' queue with mixed IPv4/6
> addresses?

I don't know which queue you mean, but the code is in the repo. I
really don't store the addresses anywhere, but it should only be
a simple matter of a union or struct sockaddr_storage if you don't
mind the wasted bytes.

Regards,
        -Richard



From per at computer.org  Wed Jan 14 18:23:23 2009
From: per at computer.org (Per Jessen)
Date: Wed, 14 Jan 2009 18:23:23 +0100
Subject: git repo and stuff
In-Reply-To: <52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>	<496D95F6.6060506@computer.org>	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>	<496DA996.7020002@computer.org>	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>
	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
Message-ID: <496E1F8B.2040607@computer.org>

Richard Nyberg wrote:
> On Wed, January 14, 2009 13:01, Alessandro Calor?? wrote:
>> 2009/1/14 Richard Nyberg <rnyberg at murmeldjur.se>:
>>> But what's the harm in having v6 enabled? If you don't disable v4 it
>>> doesn't matter that a tracker doesn't support v6.
>> That's not what I was pointing at: there's no need for an IPv4 switch
>> because every tracker supports it; the IPv4 socket should always be
>> on.
> 
> Ah. But this is not about the trackers, it's about the machine btpd
> is running on. There are IPv6 or IPv4 only machines. :)
> Leaving both protocols enabled should be no problem what so ever on
> most machines though.

Richard,
I'm curious (because I've got some code I need to make ipv6-aware 
myself) - my research so far has shown me that the easiest approach is 
to go all-IPV6 in the code and just map IPV4 addresses into the IPV6 
address space?  That way an application becomes IPV6 aware, amd there's 
no need to make special allowances for IPV4 (other than the mapping). 
Does that match what you've been doing to btpd ?

> 
>> Also, keep in mind that you should add a compile flag to
>> enable/disable IPv6 support. ;)
> 
> Why? I'd rather not, unless there's a very good reason to.

I don't see any reason. btpd becomes IPV6 aware and will use it when 
it's available.

>> How are you going to implement the peers' queue with mixed IPv4/6
>> addresses?
> 
> I don't know which queue you mean, but the code is in the repo. 

Cool, I'll be taking a look.


/Per


From DylanV at semaphore.com  Wed Jan 14 18:40:13 2009
From: DylanV at semaphore.com (Dylan Vanderhoof)
Date: Wed, 14 Jan 2009 09:40:13 -0800
Subject: Bad data from peers (was Re: git repo and stuff)
In-Reply-To: <43258.85.24.201.38.1231935117.squirrel@webmail.stargirl.org>
Message-ID: <EAC75A12BE260246A8145B73D261308709788DA5C0@exchange00.semaphore.lan>

That would be excellent, this is possibly my largest grip with btpd.  On a well connected server, I've seen 10-20 Mbps get chewed up requesting the final
piece from a client that's persistantly sending bad data.

YAY for IPv6 however, I'm thrilled to have that included finally, even if almost nobody is using it yet.

-Dylan

> -----Original Message-----
> From: btpd-users-bounces at murmeldjur.se
> [mailto:btpd-users-bounces at murmeldjur.se] On Behalf Of Richard Nyberg
> Sent: Wednesday, January 14, 2009 4:12 AM
> To: btpd general discussion.
> Subject: Re: Bad data from peers (was Re: git repo and stuff)
>
>
> Hi Scott,
>
> On Wed, January 14, 2009 06:01, Scott Feeney wrote:
> > Richard,
> >
> > I cloned the git repo to take a look. I'm really glad
> you're working on
> > btpd again.
> >
> > Don't know if this is already on your todo list, but for me the most
> > urgent feature improvement would be banning peers who send bad data.
> > I've found torrents where the fastest uploaders seem to be sending
> > exclusively bad data, in some cases making it impractical
> to download
> > due to well over 200% overhead.
> >
> > i.e., for every 1 MB of good data, I was downloading 2 MB
> of bad data.
> > Yes, this has happened to me, and then some...
>
> It's actually not for lack of effort btpd doesn't already do
> this. I've
> thought very hard on how to this in a good way. The problem is that in
> the general case it's impossible to know which peer sent the bad data.
>
> I agree that something needs to be done though, and lacking perfect
> solutions, I'll see if I can find one that's 'good enough'.
>
>         -Richard
>
>
> _______________________________________________
> btpd-users mailing list
> btpd-users at murmeldjur.se
> http://lists.stargirl.org/listinfo/btpd-users
>

From axelgenus at gmail.com  Wed Jan 14 21:15:21 2009
From: axelgenus at gmail.com (=?UTF-8?Q?Alessandro_Calor=C3=AC?=)
Date: Wed, 14 Jan 2009 21:15:21 +0100
Subject: git repo and stuff
In-Reply-To: <52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<496D95F6.6060506@computer.org>
	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
	<496DA996.7020002@computer.org>
	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>
	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>
	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
Message-ID: <eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>

2009/1/14 Richard Nyberg <rnyberg at murmeldjur.se>:
> Ah. But this is not about the trackers, it's about the machine btpd
> is running on. There are IPv6 or IPv4 only machines. :)
> Leaving both protocols enabled should be no problem what so ever on
> most machines though.
>
>> Also, keep in mind that you should add a compile flag to
>> enable/disable IPv6 support. ;)
>
> Why? I'd rather not, unless there's a very good reason to.

btpd won't compile on systems without IPv6 support.
I know almost every distribution out there is compiled with IPv6
support but for some embedded machines it would be useless to have
IPv6 support without the protocol being supported by the ISP. It
should really be "optional".

> I really don't store the addresses anywhere, but it should only be
> a simple matter of a union or struct sockaddr_storage if you don't
> mind the wasted bytes.

Uhm... I'm going to take a look at the code.

>
> Regards,
>        -Richard
>

Bye, Alessandro.

From icepic.dz at gmail.com  Thu Jan 15 08:12:29 2009
From: icepic.dz at gmail.com (Janne Johansson)
Date: Thu, 15 Jan 2009 08:12:29 +0100
Subject: Bad data from peers (was Re: git repo and stuff)
In-Reply-To: <EAC75A12BE260246A8145B73D261308709788DA5C0@exchange00.semaphore.lan>
References: <43258.85.24.201.38.1231935117.squirrel@webmail.stargirl.org>
	<EAC75A12BE260246A8145B73D261308709788DA5C0@exchange00.semaphore.lan>
Message-ID: <a10961760901142312p1ce65727k73c50d7ace29a59@mail.gmail.com>

2009/1/14 Dylan Vanderhoof <DylanV at semaphore.com>

> That would be excellent, this is possibly my largest grip with btpd.  On a
> well connected server, I've seen 10-20 Mbps get chewed up requesting the
> final
> piece from a client that's persistantly sending bad data.
>

Aww, that reminds me of that eDonkey or eMule whats-its-name which would
make peers send
as good as they could until a peer had only a small bit left, then it would
send slowly, in order to make that
almost-done peer share more to others.

-- 
To our sweethearts and wives.  May they never meet. -- 19th century toast
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.stargirl.org/pipermail/btpd-users/attachments/20090115/e32bc4d1/attachment.htm 

From per at computer.org  Thu Jan 15 08:40:50 2009
From: per at computer.org (Per Jessen)
Date: Thu, 15 Jan 2009 08:40:50 +0100
Subject: git repo and stuff
In-Reply-To: <eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>	<496D95F6.6060506@computer.org>	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>	<496DA996.7020002@computer.org>	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
	<eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>
Message-ID: <496EE882.6040308@computer.org>

Alessandro Calor? wrote:
> 2009/1/14 Richard Nyberg <rnyberg at murmeldjur.se>:
>> Ah. But this is not about the trackers, it's about the machine btpd
>> is running on. There are IPv6 or IPv4 only machines. :)
>> Leaving both protocols enabled should be no problem what so ever on
>> most machines though.
>>
>>> Also, keep in mind that you should add a compile flag to
>>> enable/disable IPv6 support. ;)
>> Why? I'd rather not, unless there's a very good reason to.
> 
> btpd won't compile on systems without IPv6 support.

autoconf should take care of that, but the code does need to be aware, 
that's true.

> I know almost every distribution out there is compiled with IPv6
> support but for some embedded machines it would be useless to have
> IPv6 support without the protocol being supported by the ISP. It
> should really be "optional".

Perhaps - if embedded devices are a real target audience for btpd. (I 
don't know if they are).


/Per Jessen, Zurich



From axelgenus at gmail.com  Thu Jan 15 09:29:34 2009
From: axelgenus at gmail.com (=?UTF-8?Q?Alessandro_Calor=C3=AC?=)
Date: Thu, 15 Jan 2009 09:29:34 +0100
Subject: git repo and stuff
In-Reply-To: <496EE882.6040308@computer.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<496D95F6.6060506@computer.org>
	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
	<496DA996.7020002@computer.org>
	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>
	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>
	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
	<eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>
	<496EE882.6040308@computer.org>
Message-ID: <eeb7df810901150029n453f7741n15424bb4c4057711@mail.gmail.com>

2009/1/15 Per Jessen <per at computer.org>:
>
>> btpd won't compile on systems without IPv6 support.
>
> autoconf should take care of that, but the code does need to be aware,
> that's true.
>

Yes, that's exactly what I meant.

>
>> I know almost every distribution out there is compiled with IPv6
>> support but for some embedded machines it would be useless to have
>> IPv6 support without the protocol being supported by the ISP. It
>> should really be "optional".
>
> Perhaps - if embedded devices are a real target audience for btpd. (I
> don't know if they are).
>

Well, btpd is perfect for embedded systems: it's small and fast and it
can be controlled remotely (via some Web interface or remote shell).
However IMHO it should be configurable at compile time: I use Gentoo
and I disabled IPv6 either in the kernel and in the toolchain because
my ISP doesn't support it. IPv6 is great but it's still not adopted by
the majority of ISP's (in the world).

>
> /Per Jessen, Zurich
>

Bye, Alessandro.

From per at computer.org  Thu Jan 15 09:46:01 2009
From: per at computer.org (Per Jessen)
Date: Thu, 15 Jan 2009 09:46:01 +0100
Subject: IPv6 (was: git repo and stuff)
In-Reply-To: <eeb7df810901150029n453f7741n15424bb4c4057711@mail.gmail.com>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>	<496D95F6.6060506@computer.org>	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>	<496DA996.7020002@computer.org>	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>	<eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>	<496EE882.6040308@computer.org>
	<eeb7df810901150029n453f7741n15424bb4c4057711@mail.gmail.com>
Message-ID: <496EF7C9.2000804@computer.org>

Alessandro Calor? wrote:

> However IMHO it should be configurable at compile time: I use Gentoo
> and I disabled IPv6 either in the kernel and in the toolchain because
> my ISP doesn't support it. IPv6 is great but it's still not adopted by
> the majority of ISP's (in the world).

Well, we're going OT here, but that your ISP doesn't support it isn't 
really a reason for not having IPv6 enabled on your local system.  We 
don't currently have external IPV6 support either, but internally 
everything has it.


/Per Jessen, Zurich


From axelgenus at gmail.com  Thu Jan 15 09:54:24 2009
From: axelgenus at gmail.com (=?UTF-8?Q?Alessandro_Calor=C3=AC?=)
Date: Thu, 15 Jan 2009 09:54:24 +0100
Subject: IPv6 (was: git repo and stuff)
In-Reply-To: <496EF7C9.2000804@computer.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
	<496DA996.7020002@computer.org>
	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>
	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>
	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
	<eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>
	<496EE882.6040308@computer.org>
	<eeb7df810901150029n453f7741n15424bb4c4057711@mail.gmail.com>
	<496EF7C9.2000804@computer.org>
Message-ID: <eeb7df810901150054l35022935md9e8e545508ab559@mail.gmail.com>

2009/1/15 Per Jessen <per at computer.org>:
> Well, we're going OT here, but that your ISP doesn't support it isn't
> really a reason for not having IPv6 enabled on your local system.  We
> don't currently have external IPV6 support either, but internally
> everything has it.

Well, if you can't use IPv6 externally what's the reason to keep it
enabled? However, that was an example: the point is that you should
not take for granted the IPv6 support when engineering a piece of
software on *NIX systems. ;)

>
> /Per Jessen, Zurich
>

Bye, Alessandro.

From per at computer.org  Thu Jan 15 10:13:36 2009
From: per at computer.org (Per Jessen)
Date: Thu, 15 Jan 2009 10:13:36 +0100
Subject: IPv6
In-Reply-To: <eeb7df810901150054l35022935md9e8e545508ab559@mail.gmail.com>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>	<496DA996.7020002@computer.org>	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>	<eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>	<496EE882.6040308@computer.org>	<eeb7df810901150029n453f7741n15424bb4c4057711@mail.gmail.com>	<496EF7C9.2000804@computer.org>
	<eeb7df810901150054l35022935md9e8e545508ab559@mail.gmail.com>
Message-ID: <496EFE40.9040601@computer.org>

Alessandro Calor? wrote:
> 2009/1/15 Per Jessen <per at computer.org>:
>> Well, we're going OT here, but that your ISP doesn't support it isn't
>> really a reason for not having IPv6 enabled on your local system.  We
>> don't currently have external IPV6 support either, but internally
>> everything has it.
> 
> Well, if you can't use IPv6 externally what's the reason to keep it
> enabled? 

A few less changes.  Plus, at some point we will want use IPv6 externaly 
too. Again, less changes when that time comes.

> However, that was an example: the point is that you should
> not take for granted the IPv6 support when engineering a piece of
> software on *NIX systems. ;)

Yep, good point.


/Per


From rnyberg at murmeldjur.se  Thu Jan 15 11:45:27 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Thu, 15 Jan 2009 11:45:27 +0100 (CET)
Subject: git repo and stuff
In-Reply-To: <eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<496D95F6.6060506@computer.org>
	<eeb7df810901140020q4674d43dl9998a29826750b4b@mail.gmail.com>
	<496DA996.7020002@computer.org>
	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>
	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>
	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
	<eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>
Message-ID: <3818.85.24.200.106.1232016327.squirrel@webmail.stargirl.org>

On Wed, January 14, 2009 21:15, Alessandro Calor? wrote:
> 2009/1/14 Richard Nyberg <rnyberg at murmeldjur.se>:
>>>
>>> Also, keep in mind that you should add a compile flag to
>>> enable/disable IPv6 support. ;)
>>
>> Why? I'd rather not, unless there's a very good reason to.
>
> btpd won't compile on systems without IPv6 support.
> I know almost every distribution out there is compiled with IPv6
> support but for some embedded machines it would be useless to have
> IPv6 support without the protocol being supported by the ISP. It
> should really be "optional".
>
Any system supporting the api:s btpd uses just have to have the
IPv6 specific flags and structs as well. If you want btpd to support
non modern posix systems as well you'll have to add code that uses
the old IPv4-only api:s.

Anyway, patches > words. If someone cares enough about some system to
port btpd to it, great! Send me the patches and I'll see if I can use
them. I have enough trouble as it is between *BSD, Linux and Solaris.

Regards,
        -Richard



From icepic.dz at gmail.com  Fri Jan 16 07:26:24 2009
From: icepic.dz at gmail.com (Janne Johansson)
Date: Fri, 16 Jan 2009 07:26:24 +0100
Subject: IPv6
In-Reply-To: <496EFE40.9040601@computer.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<52687.85.24.201.38.1231933654.squirrel@webmail.stargirl.org>
	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>
	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
	<eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>
	<496EE882.6040308@computer.org>
	<eeb7df810901150029n453f7741n15424bb4c4057711@mail.gmail.com>
	<496EF7C9.2000804@computer.org>
	<eeb7df810901150054l35022935md9e8e545508ab559@mail.gmail.com>
	<496EFE40.9040601@computer.org>
Message-ID: <a10961760901152226v3a2e4544v347a48724961eef8@mail.gmail.com>

>> However, that was an example: the point is that you should
>> not take for granted the IPv6 support when engineering a piece of
>> software on *NIX systems. ;)
>
> Yep, good point.

"IPv6 support" in software or systems is not the same as "I have working IPv6
traffic at this moment". Having v6 support in software running on a
system without
any v6 setup shouldn't cause any real troubles.

-- 
 To our sweethearts and wives.  May they never meet. -- 19th century toast

From axelgenus at gmail.com  Fri Jan 16 08:30:56 2009
From: axelgenus at gmail.com (=?UTF-8?Q?Alessandro_Calor=C3=AC?=)
Date: Fri, 16 Jan 2009 08:30:56 +0100
Subject: IPv6
In-Reply-To: <a10961760901152226v3a2e4544v347a48724961eef8@mail.gmail.com>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
	<eeb7df810901140401u5f6699f7k93be5e40ee24e096@mail.gmail.com>
	<52842.85.24.201.38.1231938051.squirrel@webmail.stargirl.org>
	<eeb7df810901141215h3abfa100l88720ca3af0b0c35@mail.gmail.com>
	<496EE882.6040308@computer.org>
	<eeb7df810901150029n453f7741n15424bb4c4057711@mail.gmail.com>
	<496EF7C9.2000804@computer.org>
	<eeb7df810901150054l35022935md9e8e545508ab559@mail.gmail.com>
	<496EFE40.9040601@computer.org>
	<a10961760901152226v3a2e4544v347a48724961eef8@mail.gmail.com>
Message-ID: <eeb7df810901152330n6c79a5b4j415c8f8a9d8cbcf3@mail.gmail.com>

2009/1/16 Janne Johansson <icepic.dz at gmail.com>:
> "IPv6 support" in software or systems is not the same as "I have working IPv6
> traffic at this moment". Having v6 support in software running on a
> system without any v6 setup shouldn't cause any real troubles.

That's because I made an example: I meant NO support for IPv6 in the
software at all (i.e. disabling it in the kernel and/or in the compile
flag of the toolchain).

Bye, Alessandro.

From d.chirkov at perm.stream.ru  Fri Jan 23 07:31:29 2009
From: d.chirkov at perm.stream.ru (=?UTF-8?B?ItCn0LjRgNC60L7QsiDQlC7Qki4i?=)
Date: Fri, 23 Jan 2009 11:31:29 +0500
Subject: BTPD bind
Message-ID: <49796441.2010008@perm.stream.ru>

Hello, all. Host have two external ip. How to manually bind BTPD with 
one of the IP. For examples, "-b" option in telnet or ssh.

Chirkov Dmitry

From partiz at gmail.com  Fri Jan 23 09:36:21 2009
From: partiz at gmail.com (=?KOI8-R?B?88XSx8XKIPfPzMvP1w==?=)
Date: Fri, 23 Jan 2009 11:36:21 +0300
Subject: BTPD bind
In-Reply-To: <49796441.2010008@perm.stream.ru>
References: <49796441.2010008@perm.stream.ru>
Message-ID: <3a27c1550901230036o304f45a8s25541cf4f13cbc2@mail.gmail.com>

2009/1/23 "?????? ?.?." <d.chirkov at perm.stream.ru>

> Hello, all. Host have two external ip. How to manually bind BTPD with
> one of the IP. For examples, "-b" option in telnet or ssh.
>

--ip addr
        Make other peers use the given address, instead of the one
        the tracker perceives as this peer's address, when contacting
        this peer.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.stargirl.org/pipermail/btpd-users/attachments/20090123/a138d2a1/attachment.htm 

From rnyberg at murmeldjur.se  Fri Jan 23 09:36:28 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Fri, 23 Jan 2009 09:36:28 +0100 (CET)
Subject: Bad torrents
In-Reply-To: <492DC1CE.5080105@gmail.com>
References: <492DC1CE.5080105@gmail.com>
Message-ID: <49387.85.24.202.247.1232699788.squirrel@webmail.stargirl.org>

Sorry about the late reply. I just noticed this amongst the spam
in the ml admin interace.

On Wed, November 26, 2008 22:38, Volodymyr Kostyrko wrote:
> I have some torrents which cannot be downloaded with btpd as of version
> 0.13. Trying to add any of them I'll have:
>
> btcli: error loading 'xxxxxxx.torrent': Invalid argument
>
> There's also nothing in log.
>
> The files are valid torrents, they are good for uTorrent or mldonkey.
>
> What can I do to send more detailed report?

Possibly they lack trackers. btpd doesn't support dht yet. Otherwise,
if you think it's ok you can send me one of them and I'll take a look.

Regards,
       -Richard


From rnyberg at murmeldjur.se  Fri Jan 23 09:44:47 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Fri, 23 Jan 2009 09:44:47 +0100 (CET)
Subject: BTPD bind
In-Reply-To: <3a27c1550901230036o304f45a8s25541cf4f13cbc2@mail.gmail.com>
References: <49796441.2010008@perm.stream.ru>
	<3a27c1550901230036o304f45a8s25541cf4f13cbc2@mail.gmail.com>
Message-ID: <45348.85.24.202.247.1232700287.squirrel@webmail.stargirl.org>

On Fri, January 23, 2009 09:36, ?????? ?????? wrote:
> 2009/1/23 "?????? ?.?." <d.chirkov at perm.stream.ru>
>
>> Hello, all. Host have two external ip. How to manually bind BTPD with
>> one of the IP. For examples, "-b" option in telnet or ssh.
>>

Not at the moment, no. But in general it's good that it listen on all
IPs since the tracker will distribute your source address to other peers
and that depends on the route from your host to the tracker. Of course,
you could set what address the tracker should distribute with the
option below.

> --ip addr
>         Make other peers use the given address, instead of the one
>         the tracker perceives as this peer's address, when contacting
>         this peer.

This is not quite the same thing. btpd still listens on all IPs.

Regards,
        -Richard



From jue at jue.li  Fri Jan 23 10:44:39 2009
From: jue at jue.li (Juergen Daubert)
Date: Fri, 23 Jan 2009 10:44:39 +0100
Subject: Bad torrents
In-Reply-To: <492DC1CE.5080105@gmail.com>
References: <492DC1CE.5080105@gmail.com>
Message-ID: <20090123094438.GA9233@jue.netz>

On Wed, Nov 26, 2008 at 11:38:22PM +0200, Volodymyr Kostyrko wrote:
> I have some torrents which cannot be downloaded with btpd as of version 
> 0.13. Trying to add any of them I'll have:
> 
> btcli: error loading 'xxxxxxx.torrent': Invalid argument
> 
> There's also nothing in log.
> 
> The files are valid torrents, they are good for uTorrent or mldonkey.
> 
> What can I do to send more detailed report?

Possibly the same problem as reported here:
http://lists.stargirl.org/pipermail/btpd-users/2007-June/000250.html


-- 
Juergen Daubert  |  mailto:jue at jue.li  
Korb, Germany    |  http://jue.li/crux


From rnyberg at murmeldjur.se  Fri Jan 23 10:53:25 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Fri, 23 Jan 2009 10:53:25 +0100 (CET)
Subject: Bad torrents
In-Reply-To: <20090123094438.GA9233@jue.netz>
References: <492DC1CE.5080105@gmail.com> <20090123094438.GA9233@jue.netz>
Message-ID: <48717.85.24.202.247.1232704405.squirrel@webmail.stargirl.org>

On Fri, January 23, 2009 10:44, Juergen Daubert wrote:
> On Wed, Nov 26, 2008 at 11:38:22PM +0200, Volodymyr Kostyrko wrote:
>> I have some torrents which cannot be downloaded with btpd as of version
>> 0.13. Trying to add any of them I'll have:
>>
>> btcli: error loading 'xxxxxxx.torrent': Invalid argument
>>
>> There's also nothing in log.
>>
>> The files are valid torrents, they are good for uTorrent or mldonkey.
>>
>> What can I do to send more detailed report?
>
> Possibly the same problem as reported here:
> http://lists.stargirl.org/pipermail/btpd-users/2007-June/000250.html
>
Nah. That's about parsing http, this is about parsing the torrent file.
Btw, the http fix will be in the upcoming release.

Regards,
       -Richard



From goatspice at gmail.com  Fri Jan 23 13:32:46 2009
From: goatspice at gmail.com (Yassen Roussev)
Date: Fri, 23 Jan 2009 12:32:46 +0000
Subject: future feature?
Message-ID: <34becb040901230432w5e66d6ady2e2c0bc561343676@mail.gmail.com>

hi Richard and all,
I have had on a few occasions the following scenario.

1 btpd running with many torrents seeded from various partitions on an
external hard-disk.

2 said hard-disk becomes offline for whatever reason without any knowledge
of mine.

3 btpd realises that torrents cannot be seeded any more, so it initiates
leeching to the
main computer volume (OS X land) recreating the relative location of the
previously existing files.

On other clients e.g. utorrent, transmission the behavior is that if
location of a seeded torrent becomes unavailable
they require user interaction to locate the files etc

It'd be great if we had the option of turning the following behaviour on -
to automatically pause torrents that lose their location for seeded or
leeched files.

How big of a job would be this to implement?

Best wishes,

Yassen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.stargirl.org/pipermail/btpd-users/attachments/20090123/0d4ac101/attachment.htm 

From bsdfan666 at cyberdude.com  Sat Jan 24 03:49:24 2009
From: bsdfan666 at cyberdude.com (BSD Fanatic)
Date: Fri, 23 Jan 2009 21:49:24 -0500
Subject: git repo and stuff
Message-ID: <20090124024924.DABD9478088@ws1-5.us4.outblaze.com>

Richard Nyberg wrote:
> Hi users!
> 
> I've made a public git repo available at [1] so go ahead and clone it
> if you're interested. I'll try to update the site soon, but it's not
> my highest priority atm.

Thanks for the revival! this list is so alive now!

I keep checking the website; kinda instinctual.. looking 
for progress reports.. perhaps you need a blog? :-)

As for the git thing, for months I tracked your svn repo.. 
I really have no interest in installing the GPL mess that is git, 
would you be willing to mirror git changes to the svn server perodically? 
or perhaps.. release snapshot tarballs?

> 
> Anyway, main points so far are IPv6 support, replacing libevent with
> my own event loop and solaris support.
> 

That's cool; but.. considering less then 1% of all trackers support IPv6, was 
it really a high priority? encryption and DHT are slightly more important.

I'm awaiting for the new release though, no more tracking development for me.. 
the bits of git will never rot on my hard drives.

-- 
Be Yourself @ mail.com!
Choose From 200+ Email Addresses
Get a Free Account at www.mail.com


From bsdfan666 at cyberdude.com  Sat Jan 24 04:49:01 2009
From: bsdfan666 at cyberdude.com (BSD Fanatic)
Date: Fri, 23 Jan 2009 22:49:01 -0500
Subject: Generating configure script with latest git code..
Message-ID: <20090124034901.39444326774@ws1-8.us4.outblaze.com>

A friend of mine made a tarball of the changes in git, but it doesn't have a 
Makefile or a configure script generated..

The old commands I used were:

aclocal
(lots of warnings.. but completes I think..)
automake -ac
(Makefile.am: C objects in subdir but `AM_PROG_CC_C_O' not in `configure.ac')
autoconf
(configure.ac:10: error: possibly undefined macro: AC_USE_SYSTEM_EXTENSIONS)

Don't work anymore.. under OpenBSD 4.4. :(

automake-1.9.6p2
autoconf-2.59p3

How do I fix it?

-- 
Be Yourself @ mail.com!
Choose From 200+ Email Addresses
Get a Free Account at www.mail.com


From rnyberg at murmeldjur.se  Sat Jan 24 11:08:24 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Sat, 24 Jan 2009 11:08:24 +0100 (CET)
Subject: git repo and stuff
In-Reply-To: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
References: <40577.85.24.200.233.1231891136.squirrel@webmail.stargirl.org>
Message-ID: <1202.85.24.201.173.1232791704.squirrel@webmail.stargirl.org>

On Wed, January 14, 2009 00:58, Richard Nyberg wrote:
>
> I intend to clean up the shutdown so that a btpd process just trying
> to send stop messages to trackers before exiting doesn't use the btpd
> dir.
I've improved the tracker code and how btpd deals with stopping torrents
and shutdown. Much nicer behaviour now. As I said, a btpd that is stopping
and only tries to talk with trackers will "disconnect" from the btpd dir
and you can start a new btpd without them interfering with each other.
The old btpd will exit after a while.

>That is: "btcli kill && btpd" should be unproblematic.
But I decided to skip this, the old btpd may still be there for
a second or so.

> When that's done it's probably btpd-0.14 time. :)
I'll put some code that tries to deal with peers with bad data
first. Then it's release time.

        -Richard


From rnyberg at murmeldjur.se  Sat Jan 24 11:19:23 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Sat, 24 Jan 2009 11:19:23 +0100 (CET)
Subject: future feature?
In-Reply-To: <34becb040901230432w5e66d6ady2e2c0bc561343676@mail.gmail.com>
References: <34becb040901230432w5e66d6ady2e2c0bc561343676@mail.gmail.com>
Message-ID: <1259.85.24.201.173.1232792363.squirrel@webmail.stargirl.org>

On Fri, January 23, 2009 13:32, Yassen Roussev wrote:
> hi Richard and all,
> I have had on a few occasions the following scenario.
>
> 1 btpd running with many torrents seeded from various partitions on
> an external hard-disk.
>
> 2 said hard-disk becomes offline for whatever reason without any
> knowledge of mine.
>
> 3 btpd realises that torrents cannot be seeded any more, so it
> initiates leeching to the main computer volume (OS X land) recreating
> the relative location of the previously existing files.
>
I think the disks must have disappeared before the torrents were started.
Otherwise btpd should have got an error on read and stopped seeding the
torrents.

> On other clients e.g. utorrent, transmission the behavior is that if
> location of a seeded torrent becomes unavailable they require user
> interaction to locate the files etc
>
> It'd be great if we had the option of turning the following behaviour
> on - to automatically pause torrents that lose their location fo
> seeded or leeched files.
>
> How big of a job would be this to implement?

I'll have a look at it after the next release sometime. I think it
might be a good idea.

Regards,
        -Richard


From rnyberg at murmeldjur.se  Sat Jan 24 11:36:14 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Sat, 24 Jan 2009 11:36:14 +0100 (CET)
Subject: git repo and stuff
In-Reply-To: <20090124024924.DABD9478088@ws1-5.us4.outblaze.com>
References: <20090124024924.DABD9478088@ws1-5.us4.outblaze.com>
Message-ID: <51346.85.24.201.173.1232793374.squirrel@webmail.stargirl.org>

On Sat, January 24, 2009 03:49, BSD Fanatic wrote:
> Richard Nyberg wrote:
>> Hi users!
>>
>> I've made a public git repo available at [1] so go ahead and clone it
>> if you're interested. I'll try to update the site soon, but it's not
>> my highest priority atm.
>
> Thanks for the revival! this list is so alive now!
>
> I keep checking the website; kinda instinctual.. looking
> for progress reports.. perhaps you need a blog? :-)

Perhaps I do. :)

> As for the git thing, for months I tracked your svn repo..
> I really have no interest in installing the GPL mess that is git,
> would you be willing to mirror git changes to the svn server
> perodically? or perhaps.. release snapshot tarballs?

Snapshot tarballs is not a bad idea actually. Though I hope to keep
changes between releases down after this one, so we'll see if it'll
be needed.

>> Anyway, main points so far are IPv6 support, replacing libevent with
>> my own event loop and solaris support.
>>
> That's cool; but.. considering less then 1% of all trackers support
> IPv6, was it really a high priority? encryption and DHT are slightly
> more important.
Agreed, but low hanging fruit you know. Also sometimes what's most
prioritized is not actually what I feel like doing at a certain time.
However, my plan is to work on dht after this release.

> I'm awaiting for the new release though, no more tracking development
> for me..
> the bits of git will never rot on my hard drives.

Hey! Linus finally writes some good software, and you're dissing it? ;)

Regards,
        -Richard



From rnyberg at murmeldjur.se  Sat Jan 24 11:45:08 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Sat, 24 Jan 2009 11:45:08 +0100 (CET)
Subject: Generating configure script with latest git code..
In-Reply-To: <20090124034901.39444326774@ws1-8.us4.outblaze.com>
References: <20090124034901.39444326774@ws1-8.us4.outblaze.com>
Message-ID: <60406.85.24.201.173.1232793908.squirrel@webmail.stargirl.org>

On Sat, January 24, 2009 04:49, BSD Fanatic wrote:
> A friend of mine made a tarball of the changes in git, but it doesn't have
> a
> Makefile or a configure script generated..
>
> The old commands I used were:
>
> aclocal
> (lots of warnings.. but completes I think..)
> automake -ac
> (Makefile.am: C objects in subdir but `AM_PROG_CC_C_O' not in
> `configure.ac')
> autoconf
> (configure.ac:10: error: possibly undefined macro:
> AC_USE_SYSTEM_EXTENSIONS)
>
> Don't work anymore.. under OpenBSD 4.4. :(
>
> automake-1.9.6p2
> autoconf-2.59p3
>
> How do I fix it?
>
Arrgh! This autofoo stuff is driving me crazy.
The order of incantation should be like this:
aclocal && autoconf && automake -ac

I happen to use autoconf 2.61 and automake 1.10. It's very possible
these versions are required, but I have no idea really.

Also, it seems that I use the wrong macro, so change "AC_PROG_CC_C_O"
to "AM_PROG_CC_C_O".

Hope this helps,
        -Richard



From d.m.berry at verizon.net  Sat Jan 24 16:45:49 2009
From: d.m.berry at verizon.net (Duane Berry)
Date: Sat, 24 Jan 2009 10:45:49 -0500
Subject: May I donate time?
Message-ID: <C605E4CD-1B28-4EA9-B4A7-140CE448049E@verizon.net>


Hello everyone,

I am new to btpd but not new to programming.  Are there any tasks or  
subtasks that I may help with?

I checked the Trac page but didn't see any tickets.


Duane


From bsdfan666 at cyberdude.com  Sat Jan 24 17:29:06 2009
From: bsdfan666 at cyberdude.com (BSD Fanatic)
Date: Sat, 24 Jan 2009 11:29:06 -0500
Subject: Generating configure script with latest git code..
Message-ID: <20090124162907.07F83326774@ws1-8.us4.outblaze.com>

Richard Nyberg wrote:
> Arrgh! This autofoo stuff is driving me crazy.
> The order of incantation should be like this:
> aclocal && autoconf && automake -ac
> 
> I happen to use autoconf 2.61 and automake 1.10. It's very possible
> these versions are required, but I have no idea really.
> 
> Also, it seems that I use the wrong macro, so change "AC_PROG_CC_C_O"
> to "AM_PROG_CC_C_O".
> 
> Hope this helps,
>          -Richard

That worked! Thanks! :-)

automake-1.9.6p2
autoconf-2.61p3

I hate this stuff too, classic Makefiles are much more fun.

Note; it seems the configure script checks for -pthread.. but 
doesn't actually use it when linking, I had to modify AM_LDFLAGS 
in Makefile myself.

Take care, you rock man! sorry for sounding harsh before. :-)

-- 
Be Yourself @ mail.com!
Choose From 200+ Email Addresses
Get a Free Account at www.mail.com


From john at jcald.com  Mon Jan 26 20:09:46 2009
From: john at jcald.com (John Caldwell)
Date: Mon, 26 Jan 2009 11:09:46 -0800
Subject: btpd: content.c:272: cm_put_bytes: Assertion `!has_bit(bf, begin
	/ (1 << 14))' failed.
Message-ID: <497E0A7A.80708@jcald.com>

I am glad to see some activity here... I have recently found BTPD and I 
am really, really impressed with it!  I can't believe how low the 
resource usage is.

I'm receiving the following error when downloading a large torrent at 
reasonably high speeds (500k/s down, 50k/s up).

btpd: content.c:272: cm_put_bytes: Assertion `!has_bit(bf, begin / (1 << 
14))' failed.

This kills btpd completely.  This is running on a Linksys NSLU2 with the 
most recent version of the Openslug/BE linux distro.  I compiled btpd 
natively on the box with GCC 4.1.2.  It works great besides this error, 
which is killing btpd with pretty high regularity :(

Build info:

# uname -a
Linux Widget 2.6.21.7 #1 PREEMPT Mon Dec 24 23:12:24 UTC 2007 armv5teb 
unknown unknown GNU/Linux

# gcc -v
Using built-in specs.
Target: armeb-linux
Configured with: 
/home/slug/releases/slugos-4.8-beta/nslu2be.tmp/work/armv5teb-linux/gcc-4.1.2-r10/gcc-4.1.2/configure 
--build=i686-linux --host=armeb-linux --target=armeb-linux --prefix=/usr 
--exec_prefix=/usr --bindir=/usr/bin --sbindir=/usr/sbin 
--libexecdir=/usr/libexec --datadir=/usr/share --sysconfdir=/etc 
--sharedstatedir=/usr/com --localstatedir=/var --libdir=/usr/lib 
--includedir=/usr/include --oldincludedir=/usr/include 
--infodir=/usr/share/info --mandir=/usr/share/man 
--enable-clocale=generic --with-gnu-ld --enable-shared 
--enable-target-optspace --enable-languages=c,c++ --enable-threads=posix 
--enable-multilib --enable-c99 --enable-long-long --enable-symvers=gnu 
--enable-libstdcxx-pch --program-prefix=armeb-linux- 
--with-local-prefix=/usr/local 
--with-gxx-include-dir=/usr/include/c++/4.1.2 --with-float=soft 
--disable-libssp --enable-__cxa_atexit --disable-nls
Thread model: posix
gcc version 4.1.2

# openssl version
OpenSSL 0.9.7g 11 Apr 2005

Any ideas?

-John



From rnyberg at murmeldjur.se  Mon Jan 26 22:56:47 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Mon, 26 Jan 2009 22:56:47 +0100 (CET)
Subject: rough roadmap
Message-ID: <54965.85.24.201.198.1233007007.squirrel@webmail.stargirl.org>

Hi users!

I've decided to try put in some protection against peers with bad
data. When that's done I'll release 0.14.

DHT is probably the next thing I'll work on but other stuff in the
pipe are udp tracker support and message stream encryption.

Then I'll probably work on improving the btpd control protocol and
the user interface.

Finally btpd might even reach 1.0. We'll see. :)

        -Richard

PS. To ease the wait for 0.14 I put up a snapshot at
    <http://www.murmeldjur.se/btpd/btpd-20090126.tar.gz>.

PPS. Anything is subject to change. :)


From rnyberg at murmeldjur.se  Mon Jan 26 23:04:12 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Mon, 26 Jan 2009 23:04:12 +0100 (CET)
Subject: May I donate time?
In-Reply-To: <C605E4CD-1B28-4EA9-B4A7-140CE448049E@verizon.net>
References: <C605E4CD-1B28-4EA9-B4A7-140CE448049E@verizon.net>
Message-ID: <38953.85.24.201.198.1233007452.squirrel@webmail.stargirl.org>

On Sat, January 24, 2009 16:45, Duane Berry wrote:
>
> Hello everyone,

Hi Duane.

> I am new to btpd but not new to programming.  Are there any tasks or
> subtasks that I may help with?

Well, two relatively straight forward tasks that I wouldn't mind if
someone else helped with are
1) udp tracker support
2) extend evloop with /dev/poll support

I'll send an update if I think of something else. Let me know if you
start on anything and feel free to ask questions.

> I checked the Trac page but didn't see any tickets.

I only ever used it for its subversion browser and now there's no
point to that either. I'll remove it.

Regards,
        -Richard


From rnyberg at murmeldjur.se  Mon Jan 26 23:09:11 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Mon, 26 Jan 2009 23:09:11 +0100 (CET)
Subject: Generating configure script with latest git code..
In-Reply-To: <20090124162907.07F83326774@ws1-8.us4.outblaze.com>
References: <20090124162907.07F83326774@ws1-8.us4.outblaze.com>
Message-ID: <37603.85.24.201.198.1233007751.squirrel@webmail.stargirl.org>

On Sat, January 24, 2009 17:29, BSD Fanatic wrote:
> Richard Nyberg wrote:
>
> Note; it seems the configure script checks for -pthread.. but
> doesn't actually use it when linking, I had to modify AM_LDFLAGS
> in Makefile myself.
>
Could you try the snapshot I put up (see rough roadmap mail), without
editing the Makefile. If it doesn't do the right thing please send me
the output from make.

> Take care, you rock man! sorry for sounding harsh before. :-)

No worries. :)

Regards,
        -Richard



From barone_rosso2 at yahoo.it  Tue Jan 27 00:32:54 2009
From: barone_rosso2 at yahoo.it (barone rosso)
Date: Tue, 27 Jan 2009 00:32:54 +0100
Subject: rough roadmap
In-Reply-To: <54965.85.24.201.198.1233007007.squirrel@webmail.stargirl.org>
References: <54965.85.24.201.198.1233007007.squirrel@webmail.stargirl.org>
Message-ID: <F43BD16C-24C7-4CC8-B401-B38D0B8603A3@yahoo.it>


Il giorno 26/gen/09, alle ore 22:56, Richard Nyberg ha scritto:

> Hi users!
>
> I've decided to try put in some protection against peers with bad
> data. When that's done I'll release 0.14.
>
> DHT is probably the next thing I'll work on but other stuff in the
> pipe are udp tracker support and message stream encryption.
>
> Then I'll probably work on improving the btpd control protocol and
> the user interface.
>
> Finally btpd might even reach 1.0. We'll see. :)
>
>        -Richard
>
> PS. To ease the wait for 0.14 I put up a snapshot at
>    <http://www.murmeldjur.se/btpd/btpd-20090126.tar.gz>.
>
> PPS. Anything is subject to change. :)
>

problem compiling on MacOsX 10.5.6.

It's a bug from Apple; it seems we haven't clock_gettime()


I findo online someone suggest to use  gettimeofday() instead of  
clock_gettime()

--
Cordialmente, vostro

barone rosso




From queueram at gmail.com  Tue Jan 27 00:46:57 2009
From: queueram at gmail.com (Marq Schneider)
Date: Mon, 26 Jan 2009 17:46:57 -0600
Subject: rough roadmap
In-Reply-To: <F43BD16C-24C7-4CC8-B401-B38D0B8603A3@yahoo.it>
References: <54965.85.24.201.198.1233007007.squirrel@webmail.stargirl.org>
	<F43BD16C-24C7-4CC8-B401-B38D0B8603A3@yahoo.it>
Message-ID: <d9b431e0901261546p2e1bbf6dh32e9e312b3b223fb@mail.gmail.com>

On Mon, Jan 26, 2009 at 17:32, barone rosso <barone_rosso2 at yahoo.it> wrote:
>
> Il giorno 26/gen/09, alle ore 22:56, Richard Nyberg ha scritto:
>
>> Hi users!
>>
>> I've decided to try put in some protection against peers with bad
>> data. When that's done I'll release 0.14.
>>
>> DHT is probably the next thing I'll work on but other stuff in the
>> pipe are udp tracker support and message stream encryption.
>>
>> Then I'll probably work on improving the btpd control protocol and
>> the user interface.
>>
>> Finally btpd might even reach 1.0. We'll see. :)
>>
>>        -Richard
>>
>> PS. To ease the wait for 0.14 I put up a snapshot at
>>    <http://www.murmeldjur.se/btpd/btpd-20090126.tar.gz>.
>>
>> PPS. Anything is subject to change. :)
>>
>
> problem compiling on MacOsX 10.5.6.
>
> It's a bug from Apple; it seems we haven't clock_gettime()
>
>
> I findo online someone suggest to use  gettimeofday() instead of
> clock_gettime()
>
> --
> Cordialmente, vostro
>
> barone rosso

I found the attached compatibility static inline function as part of
OpenBSM [1].  No guarantees since i don't have OS X myself, but it
seems to make sense.

-Marq

[1] http://www.trustedbsd.org/openbsm.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: clock_gettime.h
Type: text/x-chdr
Size: 2117 bytes
Desc: not available
Url : http://lists.stargirl.org/pipermail/btpd-users/attachments/20090126/be344bbc/attachment.h 

From barone_rosso2 at yahoo.it  Tue Jan 27 09:28:37 2009
From: barone_rosso2 at yahoo.it (barone rosso)
Date: Tue, 27 Jan 2009 09:28:37 +0100
Subject: rough roadmap
In-Reply-To: <d9b431e0901261546p2e1bbf6dh32e9e312b3b223fb@mail.gmail.com>
References: <54965.85.24.201.198.1233007007.squirrel@webmail.stargirl.org>
	<F43BD16C-24C7-4CC8-B401-B38D0B8603A3@yahoo.it>
	<d9b431e0901261546p2e1bbf6dh32e9e312b3b223fb@mail.gmail.com>
Message-ID: <DD6A554B-1717-4794-A6BE-7F49E84BA1F2@yahoo.it>


Il giorno 27/gen/09, alle ore 00:46, Marq Schneider ha scritto:

> On Mon, Jan 26, 2009 at 17:32, barone rosso <barone_rosso2 at yahoo.it>  
> wrote:
>>
>> problem compiling on MacOsX 10.5.6.
>>
>> It's a bug from Apple; it seems we haven't clock_gettime()
>>
>>
>> I findo online someone suggest to use  gettimeofday() instead of
>> clock_gettime()
>>
>> --
>> Cordialmente, vostro
>>
>> barone rosso
>
> I found the attached compatibility static inline function as part of
> OpenBSM [1].  No guarantees since i don't have OS X myself, but it
> seems to make sense.
>
>


it support CLOCK_REALTIME not  CLOCK_MONOTONIC.

problem also in btpd/net

btpd/net.c:695: error: ?AI_NUMERICSERV? undeclared (first use in this  
function)



--
Cordialmente, vostro

barone rosso




From rnyberg at murmeldjur.se  Tue Jan 27 15:35:29 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Tue, 27 Jan 2009 15:35:29 +0100 (CET)
Subject: rough roadmap
In-Reply-To: <DD6A554B-1717-4794-A6BE-7F49E84BA1F2@yahoo.it>
References: <54965.85.24.201.198.1233007007.squirrel@webmail.stargirl.org>
	<F43BD16C-24C7-4CC8-B401-B38D0B8603A3@yahoo.it>
	<d9b431e0901261546p2e1bbf6dh32e9e312b3b223fb@mail.gmail.com>
	<DD6A554B-1717-4794-A6BE-7F49E84BA1F2@yahoo.it>
Message-ID: <42889.85.24.202.250.1233066929.squirrel@webmail.stargirl.org>

On Tue, January 27, 2009 09:28, barone rosso wrote:
> Il giorno 27/gen/09, alle ore 00:46, Marq Schneider ha scritto:
>> On Mon, Jan 26, 2009 at 17:32, barone rosso <barone_rosso2 at yahoo.it>
>> wrote:
>>>
>>> problem compiling on MacOsX 10.5.6.
>>>
>>> It's a bug from Apple; it seems we haven't clock_gettime()

Ok. I want monotonic time, so stuff using gettimeofday won't do,
but this looks quite simple to do anyway. However, I'll need ssh
access to a MacOS machine where I can build btpd.

> problem also in btpd/net
>
> btpd/net.c:695: error: ?AI_NUMERICSERV? undeclared (first use in this
> function)
>
That flag is completely unnecessary so I'll just remove it.

Regards,
       -Richard




From rnyberg at murmeldjur.se  Tue Jan 27 16:04:47 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Tue, 27 Jan 2009 16:04:47 +0100 (CET)
Subject: btpd: content.c:272: cm_put_bytes: Assertion `!has_bit(bf,
	begin / (1 << 14))' failed.
In-Reply-To: <497E0A7A.80708@jcald.com>
References: <497E0A7A.80708@jcald.com>
Message-ID: <41526.85.24.202.250.1233068687.squirrel@webmail.stargirl.org>

Hi John!

On Mon, January 26, 2009 20:09, John Caldwell wrote:
> I am glad to see some activity here... I have recently found BTPD and I
> am really, really impressed with it!  I can't believe how low the
> resource usage is.

I'm rather the other way around. I can't believe how much resources some
other clients use. ;)

> I'm receiving the following error when downloading a large torrent at
> reasonably high speeds (500k/s down, 50k/s up).
>
> btpd: content.c:272: cm_put_bytes: Assertion `!has_bit(bf, begin / (1 <<
> 14))' failed.
>
> This kills btpd completely.  This is running on a Linksys NSLU2 with the
> most recent version of the Openslug/BE linux distro.  I compiled btpd
> natively on the box with GCC 4.1.2.  It works great besides this error,
> which is killing btpd with pretty high regularity :(
>
Is it always that error?

> Any ideas?
>
I've got some suspicions at least. Would you be willing to help?
If so, please try the snapshot I put up yesterday. It'll probably
fail in the same way, but it'd be good to know. If it doesn't work
I'll have a modified btpd for you to try.

Regards,
        -Richard



From john at jcald.com  Tue Jan 27 16:27:55 2009
From: john at jcald.com (John Caldwell)
Date: Tue, 27 Jan 2009 07:27:55 -0800
Subject: btpd: content.c:272: cm_put_bytes: Assertion `!has_bit(bf,
	begin / (1 << 14))' failed.
In-Reply-To: <41526.85.24.202.250.1233068687.squirrel@webmail.stargirl.org>
References: <497E0A7A.80708@jcald.com>
	<41526.85.24.202.250.1233068687.squirrel@webmail.stargirl.org>
Message-ID: <2E421296-802E-41DF-817A-8FAB893032C6@jcald.com>





On Jan 27, 2009, at 7:04 AM, "Richard Nyberg" <rnyberg at murmeldjur.se>  
wrote:

> Hi John!
>
> On Mon, January 26, 2009 20:09, John Caldwell wrote:
>> I am glad to see some activity here... I have recently found BTPD  
>> and I
>> am really, really impressed with it!  I can't believe how low the
>> resource usage is.
>
> I'm rather the other way around. I can't believe how much resources  
> some
> other clients use. ;)

Haha! I wish more devs thought this way. It is tough to keep lots of  
stuff running on the NSLU2 as it only has 32mb of memory.

>
>
>> I'm receiving the following error when downloading a large torrent at
>> reasonably high speeds (500k/s down, 50k/s up).
>>
>> btpd: content.c:272: cm_put_bytes: Assertion `!has_bit(bf, begin /  
>> (1 <<
>> 14))' failed.
>>
>> This kills btpd completely.  This is running on a Linksys NSLU2  
>> with the
>> most recent version of the Openslug/BE linux distro.  I compiled btpd
>> natively on the box with GCC 4.1.2.  It works great besides this  
>> error,
>> which is killing btpd with pretty high regularity :(
>>
> Is it always that error?
>

Yup, always the same.


>> Any ideas?
>>
> I've got some suspicions at least. Would you be willing to help?

Absolutely!  I do java/perl/php development, so I may be able to do  
some code additions as well... One feature I would really like is a  
delimited, combined output of btcli list and btcli stat, which I think  
I can tackle.  I'm using both for a web interface I'm writing and  
currently i am using some ugly regexes to parse it.

>
> If so, please try the snapshot I put up yesterday. It'll probably
> fail in the same way, but it'd be good to know. If it doesn't work
> I'll have a modified btpd for you to try.
>

I will try the snapshot and report back.  I tried pulling out the  
assertion as well as just trapping the error and neither one got me  
very far.  Commenting it out causes it to hang and stop responding to  
requests from btcli.  Using the error handling code to catch it and  
return cause it to throw the IO errors in the loop below and die upon  
the next call to cm_put_bytes.

I will let you know what I find from using the snapshot.  Thanks,

-John


>

From bsdfan666 at cyberdude.com  Wed Jan 28 03:07:16 2009
From: bsdfan666 at cyberdude.com (BSD Fanatic)
Date: Tue, 27 Jan 2009 21:07:16 -0500
Subject: Generating configure script with latest git code..
Message-ID: <20090128020716.1A3DF326774@ws1-8.us4.outblaze.com>

Richard wrote:
> I wrote:
> > Note; it seems the configure script checks for -pthread.. but
> > doesn't actually use it when linking, I had to modify AM_LDFLAGS
> > in Makefile myself.
> >
> Could you try the snapshot I put up (see rough roadmap mail), without
> editing the Makefile. If it doesn't do the right thing please send me
> the output from make.

Hi,

I tried the snapshot tarball, it works fine.. no linking issues, but the 
latest "git" still produces errors related to pthreads.

Thanks.

-- 
Be Yourself @ mail.com!
Choose From 200+ Email Addresses
Get a Free Account at www.mail.com


From nemosoft at smcc.demon.nl  Wed Jan 28 18:51:44 2009
From: nemosoft at smcc.demon.nl (Nemosoft Unv.)
Date: Wed, 28 Jan 2009 18:51:44 +0100
Subject: [announce] btmon version 0.1
Message-ID: <200901281851.44102.nemosoft@smcc.demon.nl>

Hello all,

I just wanted to let you know that I've written a small tool to display the 
statistics of BTPD as a nicely formatted HTML page. I find this very useful 
to (remotely) watch the progress while downloading a torrent, or to see how 
your torrents are doing in general. The HTML page reads a lot easier than the 
text outpuf of btcli, IMO.

The tool can be found at http://www.smcc.demon.nl/btmon

Please let me know what you think; comments & bug reports are welcome!

Regards,

 - Nemosoft Unv

From partiz at gmail.com  Wed Jan 28 19:42:19 2009
From: partiz at gmail.com (=?KOI8-R?B?88XSx8XKIPfPzMvP1w==?=)
Date: Wed, 28 Jan 2009 21:42:19 +0300
Subject: php-btpd (BTPD Web control panel)
Message-ID: <3a27c1550901281042gc11d533m3b8c6e5371276b3a@mail.gmail.com>

Hi All!

http://code.google.com/p/php-btpd/ - my web-based control panel for BTPD.
maybe someone find it useful...

WBR,
Serge.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.stargirl.org/pipermail/btpd-users/attachments/20090128/0bfb16dc/attachment.htm 

From axelgenus at gmail.com  Thu Jan 29 09:14:08 2009
From: axelgenus at gmail.com (=?UTF-8?Q?Alessandro_Calor=C3=AC?=)
Date: Thu, 29 Jan 2009 09:14:08 +0100
Subject: [announce] btmon version 0.1
In-Reply-To: <200901281851.44102.nemosoft@smcc.demon.nl>
References: <200901281851.44102.nemosoft@smcc.demon.nl>
Message-ID: <eeb7df810901290014y645cf3dfje4e34f9a10f7f8c7@mail.gmail.com>

2009/1/28 Nemosoft Unv. <nemosoft at smcc.demon.nl>:
> Hello all,
>
> I just wanted to let you know that I've written a small tool to display the
> statistics of BTPD as a nicely formatted HTML page. I find this very useful
> to (remotely) watch the progress while downloading a torrent, or to see how
> your torrents are doing in general. The HTML page reads a lot easier than the
> text outpuf of btcli, IMO.
>
> The tool can be found at http://www.smcc.demon.nl/btmon
>
> Please let me know what you think; comments & bug reports are welcome!

Nice! Did you try to make btmon to communicate with the daemon instead
of reformatting the btcli output to HTML?
However, AFAIK there's another Web interface for btpd
(http://code.google.com/p/btpd-webui/).

> Regards,
>
>  - Nemosoft Unv

Bye,
Alessandro.

From rnyberg at murmeldjur.se  Thu Jan 29 17:17:18 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Thu, 29 Jan 2009 17:17:18 +0100 (CET)
Subject: Generating configure script with latest git code..
In-Reply-To: <20090128020716.1A3DF326774@ws1-8.us4.outblaze.com>
References: <20090128020716.1A3DF326774@ws1-8.us4.outblaze.com>
Message-ID: <39252.85.24.200.117.1233245838.squirrel@webmail.stargirl.org>

On Wed, January 28, 2009 03:07, BSD Fanatic wrote:
>
> Hi,
>
> I tried the snapshot tarball, it works fine.. no linking issues, but the
> latest "git" still produces errors related to pthreads.
>
You need automake >= 1.10 to get a working Makefile. configure.ac now
contains versioning checks.

Regards,
       -Richard



From rnyberg at murmeldjur.se  Thu Jan 29 17:22:28 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Thu, 29 Jan 2009 17:22:28 +0100 (CET)
Subject: new snapshot
Message-ID: <35276.85.24.200.117.1233246148.squirrel@webmail.stargirl.org>

http://www.murmeldjur.se/btpd/btpd-20090129.tar.gz

btpd should work on MacOS again.
Thanks for the login access barone rosso!

        -Richard


From nemosoft at smcc.demon.nl  Thu Jan 29 19:04:44 2009
From: nemosoft at smcc.demon.nl (Nemosoft Unv.)
Date: Thu, 29 Jan 2009 19:04:44 +0100
Subject: [announce] btmon version 0.1
In-Reply-To: <eeb7df810901290014y645cf3dfje4e34f9a10f7f8c7@mail.gmail.com>
References: <200901281851.44102.nemosoft@smcc.demon.nl>
	<eeb7df810901290014y645cf3dfje4e34f9a10f7f8c7@mail.gmail.com>
Message-ID: <200901291904.44271.nemosoft@smcc.demon.nl>

On Thursday 29 January 2009 09:14:08 Alessandro Calor? wrote:
Hi,

> 2009/1/28 Nemosoft Unv. <nemosoft at smcc.demon.nl>:
> > Hello all,
> >
> >
> > Please let me know what you think; comments & bug reports are welcome!
>
> Nice! Did you try to make btmon to communicate with the daemon instead
> of reformatting the btcli output to HTML?

Nope; although I suppose it's possible, but then I would have to do some 
socket stuff and, more importantly, implement the protocol and variable 
wrap/unwrap code that btpd uses. Plus, it would require access to the btpd 
socket, which leads to all kinds of security issues (remember, the webserver 
runs with a different user than btpd, and any webpage/script served by that 
server will have the same user ID)

> However, AFAIK there's another Web interface for btpd
> (http://code.google.com/p/btpd-webui/).

Yes, I saw the announcement, but I haven't looked at it. Besides, Python isn't 
my strongest point :-)

 - Nemosoft

From john at jcald.com  Thu Jan 29 19:15:20 2009
From: john at jcald.com (John Caldwell)
Date: Thu, 29 Jan 2009 10:15:20 -0800
Subject: [announce] btmon version 0.1
In-Reply-To: <200901291904.44271.nemosoft@smcc.demon.nl>
References: <200901281851.44102.nemosoft@smcc.demon.nl>	<eeb7df810901290014y645cf3dfje4e34f9a10f7f8c7@mail.gmail.com>
	<200901291904.44271.nemosoft@smcc.demon.nl>
Message-ID: <4981F238.3030009@jcald.com>

I'm currently working on a perl cgi based monitor similar to this.  I'm 
a PHP developer, but the hardware that I'm running btpd on (Linksys 
NSLU2) is a little too small for a full PHP installation :)  I'm also 
working on a modified version of btcli that outputs a bit more "raw" 
output, and combines the output of list and stat into one list, as such:

name|id|status|content_got|content_size|downloaded|rate_down|uploaded|rate_up|tot_up|peers|pieces_seen|torrent_pieces|tr_good
xxxxxxxxxxxxxxxxxxxxx|5|S|55544267|55544267|16384|0|0|0|0|0|848|848|0

I'm using this currently and it is working quite well.  It allows a bit 
more fine-grained control  of output formatting, etc.  I previously used 
regular expressions to parse it all out, but they are a bit cumbersome.  
I've also added in file upload functionality to the script, and 
management of the status of torrents (stop/pause/kill).  At this point 
there is no security and the code is very rough, but some time next week 
I will post my code to the list, including the btcli patch.  I added a 
new flag (-r) to 'btcli stat' for this raw output.

-John


Nemosoft Unv. wrote:
> On Thursday 29 January 2009 09:14:08 Alessandro Calor? wrote:
> Hi,
>
>   
>> 2009/1/28 Nemosoft Unv. <nemosoft at smcc.demon.nl>:
>>     
>>> Hello all,
>>>
>>>
>>> Please let me know what you think; comments & bug reports are welcome!
>>>       
>> Nice! Did you try to make btmon to communicate with the daemon instead
>> of reformatting the btcli output to HTML?
>>     
>
> Nope; although I suppose it's possible, but then I would have to do some 
> socket stuff and, more importantly, implement the protocol and variable 
> wrap/unwrap code that btpd uses. Plus, it would require access to the btpd 
> socket, which leads to all kinds of security issues (remember, the webserver 
> runs with a different user than btpd, and any webpage/script served by that 
> server will have the same user ID)
>
>   
>> However, AFAIK there's another Web interface for btpd
>> (http://code.google.com/p/btpd-webui/).
>>     
>
> Yes, I saw the announcement, but I haven't looked at it. Besides, Python isn't 
> my strongest point :-)
>
>  - Nemosoft
> _______________________________________________
> btpd-users mailing list
> btpd-users at murmeldjur.se
> http://lists.stargirl.org/listinfo/btpd-users
>   


From john at jcald.com  Thu Jan 29 21:12:18 2009
From: john at jcald.com (John Caldwell)
Date: Thu, 29 Jan 2009 12:12:18 -0800
Subject: btpd: content.c:272: cm_put_bytes: Assertion `!has_bit(bf, begin
	/ (1 << 14))' failed.
In-Reply-To: <2E421296-802E-41DF-817A-8FAB893032C6@jcald.com>
References: <497E0A7A.80708@jcald.com>
	<41526.85.24.202.250.1233068687.squirrel@webmail.stargirl.org>
	<2E421296-802E-41DF-817A-8FAB893032C6@jcald.com>
Message-ID: <49820DA2.5020109@jcald.com>

John Caldwell wrote:
> I will try the snapshot and report back.  I tried pulling out the 
> assertion as well as just trapping the error and neither one got me 
> very far.  Commenting it out causes it to hang and stop responding to 
> requests from btcli.  Using the error handling code to catch it and 
> return cause it to throw the IO errors in the loop below and die upon 
> the next call to cm_put_bytes.
>
> I will let you know what I find from using the snapshot.  

I tried a couple of things, with mixed success.  I did a 'git clone' on 
the 27th and compiled it, and it worked well on the NSLU2.  Large 
torrents no longer throw the cm_put_bytes assertion.  There is one 
caveat though.  Torrents off of a particular site would not start.  At 
first I thought this was due to the version number, but I changed that 
and it still doesn't work.  I believe they're using a custom tracker of 
some sort.  What is interesting is that 0.13 does not have any problem 
with these torrents and starts working immediately.

I have also tested the most recent snapshot (20090129), and it is able 
to connect to the tracker without any issues.  However, I did get the 
following error while downloading:

btpd: btpd/download.c:18: dl_on_piece_ann: Assertion `pc != ((void *)0)' 
failed.


-John


From bsdfan666 at cyberdude.com  Thu Jan 29 22:55:24 2009
From: bsdfan666 at cyberdude.com (BSD Fanatic)
Date: Thu, 29 Jan 2009 16:55:24 -0500
Subject: Generating configure script with latest git code..
Message-ID: <20090129215524.2DADABE4CE4@ws1-9.us4.outblaze.com>

> You need automake >= 1.10 to get a working Makefile. configure.ac now
> contains versioning checks.
> 
> Regards,
>         -Richard

Darn, that's not available in OpenBSD 4.4... this sucks big time.

I guess I'll go back to the last SVN version.

* no IPv6 bloat.. libevent beauty.. 
* no huge ass peer/seed locate times.. (takes forever to download now).
* no git..
* no compile problems..

Bye Richard; I miss the old btpd.

-- 
Be Yourself @ mail.com!
Choose From 200+ Email Addresses
Get a Free Account at www.mail.com


From nemosoft at smcc.demon.nl  Thu Jan 29 22:58:45 2009
From: nemosoft at smcc.demon.nl (Nemosoft Unv.)
Date: Thu, 29 Jan 2009 22:58:45 +0100
Subject: [announce] btmon version 0.1
In-Reply-To: <4981F238.3030009@jcald.com>
References: <200901281851.44102.nemosoft@smcc.demon.nl>
	<200901291904.44271.nemosoft@smcc.demon.nl>
	<4981F238.3030009@jcald.com>
Message-ID: <200901292258.45921.nemosoft@smcc.demon.nl>

Hi,

On Thursday 29 January 2009 19:15:20 John Caldwell wrote:
> I'm currently working on a perl cgi based monitor similar to this.  I'm
> a PHP developer, but the hardware that I'm running btpd on (Linksys
> NSLU2) is a little too small for a full PHP installation :)  I'm also
> working on a modified version of btcli that outputs a bit more "raw"
> output, and combines the output of list and stat into one list, as such:
>
> name|id|status|content_got|content_size|downloaded|rate_down|uploaded|rate_
>up|tot_up|peers|pieces_seen|torrent_pieces|tr_good
> xxxxxxxxxxxxxxxxxxxxx|5|S|55544267|55544267|16384|0|0|0|0|0|848|848|0

Hmm, that's interesting. However, I found out that the set of fields that I 
choose for btmon are the most useful ones. But I suppose all these fields can 
be displayed in a 'detailed' view or something similar.

BTW, you did escape '|' symbols in the name, didn't you? ;)

> I'm using this currently and it is working quite well.  It allows a bit
> more fine-grained control  of output formatting, etc.  I previously used
> regular expressions to parse it all out, but they are a bit cumbersome.
> I've also added in file upload functionality to the script, and
> management of the status of torrents (stop/pause/kill).  At this point
> there is no security and the code is very rough, but some time next week
> I will post my code to the list, including the btcli patch.  I added a
> new flag (-r) to 'btcli stat' for this raw output.

Richard, any chance of getting this flag in? Might be useful for at least 2 of 
us. :-)

 - Nemosoft

From rnyberg at murmeldjur.se  Fri Jan 30 00:45:03 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Fri, 30 Jan 2009 00:45:03 +0100 (CET)
Subject: Generating configure script with latest git code..
In-Reply-To: <20090129215524.2DADABE4CE4@ws1-9.us4.outblaze.com>
References: <20090129215524.2DADABE4CE4@ws1-9.us4.outblaze.com>
Message-ID: <46629.85.24.200.117.1233272703.squirrel@webmail.stargirl.org>

On Thu, January 29, 2009 22:55, BSD Fanatic wrote:
> * no huge ass peer/seed locate times.. (takes forever to download now).

Yes, there's definitely a problem. This is not apparent with all
trackers though, so I had missed it. More detail in reports is
always good, though I realize you're more interested in ranting than
helping.



From bsdfan666 at cyberdude.com  Fri Jan 30 01:53:43 2009
From: bsdfan666 at cyberdude.com (BSD Fanatic)
Date: Thu, 29 Jan 2009 19:53:43 -0500
Subject: Generating configure script with latest git code..
Message-ID: <20090130005343.2BD3D1BF2C4@ws1-10.us4.outblaze.com>

Richard Nyberg wrote
> Yes, there's definitely a problem. This is not apparent with all
> trackers though, so I had missed it. 

So I'm not imagining things then.

> More detail in reports is always good, though I realize you're more 
> interested in ranting than helping.

I'm emotionally unstable, what sort of detail do you expect from
a novice?

:-) hi.

-- 
Be Yourself @ mail.com!
Choose From 200+ Email Addresses
Get a Free Account at www.mail.com


From mattf at bauchan.org  Fri Jan 30 04:32:53 2009
From: mattf at bauchan.org (Matt Fretwell)
Date: Fri, 30 Jan 2009 03:32:53 +0000
Subject: Generating configure script with latest git code..
In-Reply-To: <20090130005343.2BD3D1BF2C4@ws1-10.us4.outblaze.com>
References: <20090130005343.2BD3D1BF2C4@ws1-10.us4.outblaze.com>
Message-ID: <20090130033253.5d6d0403.mattf@bauchan.org>

On Thu, 29 Jan 2009 19:53:43 -0500
"BSD Fanatic" <bsdfan666 at cyberdude.com> wrote:

> > More detail in reports is always good, though I realize you're more 
> > interested in ranting than helping.
> 
> I'm emotionally unstable, what sort of detail do you expect from
> a novice?


 The same as from anyone else. :)

 If you think things ought to be done differently, btw, put your point
across. Don't just strop off because you don't like something.

 With regards to the details bit, I would say that the more info you
can supply with regards to a specific problem you are encountering, the
better. For example, any certain conditions, i.e: clients/hosts,
applicable to the problem which may help give an idea or possible clue
as to where a problem may lie. The more information you can supply, the
easier tracing a bug may become.


Matt

From rnyberg at murmeldjur.se  Fri Jan 30 15:04:40 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Fri, 30 Jan 2009 15:04:40 +0100 (CET)
Subject: new snapshot
In-Reply-To: <35276.85.24.200.117.1233246148.squirrel@webmail.stargirl.org>
References: <35276.85.24.200.117.1233246148.squirrel@webmail.stargirl.org>
Message-ID: <48425.85.24.201.138.1233324280.squirrel@webmail.stargirl.org>

And another one, which should get rid of some tracker problems.

http://www.murmeldjur.se/btpd/btpd-20090130.tar.gz

o Use only IPv4 by default. Some sites have useless IPv6 trackers
  that get in the way.
o People usually don't create tracker tiers correctly so don't assume
  they do.

I should probably make the IP version option per torrent in the future.

        -Richard


From rnyberg at murmeldjur.se  Fri Jan 30 15:19:16 2009
From: rnyberg at murmeldjur.se (Richard Nyberg)
Date: Fri, 30 Jan 2009 15:19:16 +0100 (CET)
Subject: btpd: content.c:272: cm_put_bytes: Assertion `!has_bit(bf,
	begin / (1 << 14))' failed.
In-Reply-To: <49820DA2.5020109@jcald.com>
References: <497E0A7A.80708@jcald.com>
	<41526.85.24.202.250.1233068687.squirrel@webmail.stargirl.org>
	<2E421296-802E-41DF-817A-8FAB893032C6@jcald.com>
	<49820DA2.5020109@jcald.com>
Message-ID: <34143.85.24.201.138.1233325156.squirrel@webmail.stargirl.org>

On Thu, January 29, 2009 21:12, John Caldwell wrote:
> John Caldwell wrote:
> I tried a couple of things, with mixed success.  I did a 'git clone' on
> the 27th and compiled it, and it worked well on the NSLU2.  Large
> torrents no longer throw the cm_put_bytes assertion.  There is one
> caveat though.  Torrents off of a particular site would not start.  At
> first I thought this was due to the version number, but I changed that
> and it still doesn't work.  I believe they're using a custom tracker of
> some sort.  What is interesting is that 0.13 does not have any problem
> with these torrents and starts working immediately.

Is it a multi tracker enabled torrent with multiple trackers in the same
tier? Ie. btinfo shows more then one tracker inside the inner brackets:

Tracker URLs: [ [ tracker1 tracker2 ] ]

Did you get messages about tracker failure in the log?

I've fixed one tracker related problem that could be what happened
for you.

> I have also tested the most recent snapshot (20090129), and it is able
> to connect to the tracker without any issues.  However, I did get the
> following error while downloading:
>
> btpd: btpd/download.c:18: dl_on_piece_ann: Assertion `pc != ((void *)0)'
> failed.

The root cause is very likely the same as for the previous error.
I'll send you a modified btpd to try. Don't bother with snapshots
or git clones for now.

Thanks!
        -Richard



From john at jcald.com  Fri Jan 30 17:41:38 2009
From: john at jcald.com (John Caldwell)
Date: Fri, 30 Jan 2009 08:41:38 -0800
Subject: btpd: content.c:272: cm_put_bytes: Assertion `!has_bit(bf, begin
	/ (1 << 14))' failed.
In-Reply-To: <34143.85.24.201.138.1233325156.squirrel@webmail.stargirl.org>
References: <497E0A7A.80708@jcald.com>	<41526.85.24.202.250.1233068687.squirrel@webmail.stargirl.org>	<2E421296-802E-41DF-817A-8FAB893032C6@jcald.com>	<49820DA2.5020109@jcald.com>
	<34143.85.24.201.138.1233325156.squirrel@webmail.stargirl.org>
Message-ID: <49832DC2.9060800@jcald.com>

Richard Nyberg wrote:
> Is it a multi tracker enabled torrent with multiple trackers in the same
> tier? Ie. btinfo shows more then one tracker inside the inner brackets:
>
> Tracker URLs: [ [ tracker1 tracker2 ] ]
>   
No, it is a single tracker.

> Did you get messages about tracker failure in the log?
>   

That is the weird thing- there are no messages in the log.  Eventually 
the torrent's status changes to "I", but that is the only indication 
that something is wrong (besides zero bytes downloaded, of course!)

> The root cause is very likely the same as for the previous error.
> I'll send you a modified btpd to try. Don't bother with snapshots
> or git clones for now.
>   

Compiling it now :)  I will let you know how it goes.

-John