Send Devl mailing list submissions to
devl at freenetproject.org
To subscribe or unsubscribe via the World Wide Web, visit
http://www.uprizer.com/mailman/listinfo/devl
or, via email, send a message with subject or body 'help' to
devl-request at freenetproject.org
You can reach the person managing the list at
devl-admin at freenetproject.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Devl digest..."
Today's Topics:
1. Re: Espra - what's that? (Mark J. Roberts)
2. RE: 0.3.7 windoze bugs (Ross Smith)
3. Re: Espra - what's that? (Steven Hazel)
4. Compression of Jars (Mr.Bad)
5. Re: Espra - what's that? (Mark J. Roberts)
6. Re: Espra - what's that? (Steven Hazel)
7. IMPORTANT: Where is the Freenet Windows release? (Ian Clarke)
8. RE: Killing Freenet (Re: [freenet-devl] Aardvark) (Benjamin Coates)
9. RE: Killing Freenet (Re: [freenet-devl] Aardvark) (Fred Salzer)
10. Re: Aardvark (Chris Anderson)
11. Re: Espra - what's that? (tav)
12. Re: Compression of Jars (Don Marti)
13. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Scott G. Miller)
14. Re: Compression of Jars (Steven Hazel)
--__--__--
Message: 1
Date: Mon, 5 Feb 2001 16:28:51 -0500 (EST)
From: "Mark J. Roberts" <[email protected]>
To: <devl at freenetproject.org>
Subject: Re: [freenet-devl] Espra - what's that?
Reply-To: devl at freenetproject.org
On 5 Feb 2001, Steven Hazel wrote:
> "tav" <tavlists at espnow.com> writes:
>
> > therefore, i am proud to hear that espra is in VAPOR!
> >
> > for the vapor shall condense and eventually become the
> > expansive watery oceans so vital to life!
>
> And I'll be happy to see it do so. I'd also like to see it backed by
> libfreenet, and I'm willing to put in a considerable amount of time
> helping to make that happen, because otherwise I'm going to have to
> write a libfreenet-backed GUI client myself.
>
> So let me know where to start, and I'll just dig right in.
I haven't been able to make libfreenet work. Requests fail with -9 and
inserts with -15.
Inserts fail in endtoend.c:twofish_init. The keystr string just before it
aborts equals "daaeba90a5b7d44be8e08ed9a72a74c4" for the key
"freenet:KSK at blah". Don't know if that's relevant though.
Haven't checked out requests yet. Any ideas?
--
Mark Roberts
mjr at statesmean.com
--__--__--
Message: 2
From: "Ross Smith" <[email protected]>
To: <devl at freenetproject.org>
Subject: RE: [freenet-devl] 0.3.7 windoze bugs
Date: Mon, 5 Feb 2001 14:56:33 -0800
Reply-To: devl at freenetproject.org
> -----Original Message-----
> [mailto:devl-admin at freenetproject.org]On Behalf Of Sebastian Spaeth
> > Ross Smith wrote:
>
> I do have a small utility called cfgclient which does this adaption to
> flaunch.ini and .fproxyrc it is however deleted after installation. We
> could keep it around telling the user to run it after he changed a port.
> I just don't want to litter the directory with normally useless
> utilities. What do you think?
Not up to me, but I would say, leave it around for the interpid.
> > > It seems a previous installer (0.3.3?) asked for the port address before
> > > generating the config files. Can we add
> this to the new
> > > installer?
>
> Do you really need to decide yourself on a certain port or is automatic
> random choice OK for you? It would not be much work to ask explicitely
> for a port.
I need to define a specific port as my network administrator has opened that
port in the firewall. If you dictate to me the port,
then I'll have to either 1) edit the three config files, or 2) ask my net admin
to reconfigure the firewall ... both our yukky.
> > > 3. Add/Remove Programs says Freenet 0.3.5 is installed.
>
> Small but ugly glitch, where I forgot to upate the installatioons
> script. Is fixed.
Thanks,
Ross
--__--__--
Message: 3
To: devl at freenetproject.org
Subject: Re: [freenet-devl] Espra - what's that?
From: Steven Hazel <[email protected]>
Date: 05 Feb 2001 17:02:28 -0600
Reply-To: devl at freenetproject.org
"Mark J. Roberts" <mjr at statesmean.com> writes:
> I haven't been able to make libfreenet work. Requests fail with -9 and
> inserts with -15.
>
> Inserts fail in endtoend.c:twofish_init. The keystr string just before it
> aborts equals "daaeba90a5b7d44be8e08ed9a72a74c4" for the key
> "freenet:KSK at blah". Don't know if that's relevant though.
>
> Haven't checked out requests yet. Any ideas?
The node is hanging up on you on request (-9 is "READ FAILED" -- see
protocol.h for a comprehensive list of error codes). That's probably
due to some deeper problem.
-15 (MAKEKEY FAILED) is probably the result of mis-compiled twofish
code. I just realized that my build instructions are wrong for x86
machines -- you'll need to compile the twofish code with "-D_M_IX86"
on the gcc command line.
Let me know if that doesn't fix the request problem as well.
My next release will have makefiles for the rijndael and twofish code,
so hopefully this won't be a problem in the near future.
-S
--__--__--
Message: 4
To: devl at freenetproject.org
From: Mr.Bad <[email protected]>
Organization: Pigdog Journal
Date: 05 Feb 2001 15:24:19 -0800
Subject: [freenet-devl] Compression of Jars
Reply-To: devl at freenetproject.org
So, here's my informal study: I built the 0.3.7 jar using the
following commands:
ant jar -Dbuild.compiler=jikes -Dcompress-jar=yes
-Djarfile=compressed-freenet.jar
...and:
ant jar -Dbuild.compiler=jikes -Dcompress-jar=no
-Djarfile=uncompressed-freenet.jar
"compress-jar" just tells ant to use its default for compressing the
jar. I checked that the two files did in fact contain the same data by
un-jarring them to separate directories and diffing -r the dirs. No
difference.
I then ran zip and gzip, with no options, on the two files. Here were
the results:
---8<---
evan priss:~/tmp/jarcompare$ vdir
total 2588
-rw-r--r-- 1 evan evan 448051 Feb 5 15:14 compressed-freenet.jar
-rw-r--r-- 1 evan evan 400769 Feb 5 15:16
compressed-freenet.jar.gz
-rw-r--r-- 1 evan evan 400904 Feb 5 15:17
compressed-freenet.jar.zip
-rw-r--r-- 1 evan evan 811264 Feb 5 15:14 uncompressed-freenet.jar
-rw-r--r-- 1 evan evan 274794 Feb 5 15:16
uncompressed-freenet.jar.gz
-rw-r--r-- 1 evan evan 274931 Feb 5 15:17
uncompressed-freenet.jar.zip
---8<---
My guess is that the jar tool uses much more conservative compression
than either zip (that's infozip, btw) or gzip. Once the file is
compressed, most of the useful redundancy has been removed, so it
won't compress much more.
Of course, I've been wrong before, but it's still interesting.
~Mr. Bad
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/\____/\ Mr. Bad <mr.bad at pigdog.org>
\ / Pigdog Journal | http://pigdog.org/ | *Stay*Real*Bad*
| (X \x)
( ((**) "If it's not bad, don't do it.
\ <vvv> If it's not crazy, don't say it." - Ben Franklin
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
--__--__--
Message: 5
Date: Mon, 5 Feb 2001 17:33:28 -0500 (EST)
From: "Mark J. Roberts" <[email protected]>
To: <devl at freenetproject.org>
Subject: Re: [freenet-devl] Espra - what's that?
Reply-To: devl at freenetproject.org
On 5 Feb 2001, Steven Hazel wrote:
> "Mark J. Roberts" <mjr at statesmean.com> writes:
>
> > I haven't been able to make libfreenet work. Requests fail with -9 and
> > inserts with -15.
> >
> > Inserts fail in endtoend.c:twofish_init. The keystr string just before it
> > aborts equals "daaeba90a5b7d44be8e08ed9a72a74c4" for the key
> > "freenet:KSK at blah". Don't know if that's relevant though.
> >
> > Haven't checked out requests yet. Any ideas?
>
> The node is hanging up on you on request (-9 is "READ FAILED" -- see
> protocol.h for a comprehensive list of error codes). That's probably
> due to some deeper problem.
>
> -15 (MAKEKEY FAILED) is probably the result of mis-compiled twofish
> code. I just realized that my build instructions are wrong for x86
> machines -- you'll need to compile the twofish code with "-D_M_IX86"
> on the gcc command line.
Now inserts fail with -9 as well.
Whiterose gives me "Error in crypto handshake" and Fred is weirder. The
first time I try, it gives me something like:
Accepted connection:tcp/127.0.0.1:2001
New connectionhandler with tcp/127.0.0.1:2001
Authenticating tcp/127.0.0.1:2001
tcp/127.0.0.1:2001 authenticated
Finished with connection - closing
and subsequent attempts fail like
Authenticating tcp/127.0.0.1:1977
Failed to initialize connection: java.net.SocketException: Connection reset by
peer: Connection reset by peer
presumably because the first handshake is still valid? Anyway the DH
params seem valid on Fred's side:
X:
117070260848926090478121567078125241186524020864447397449586038241578352523711094596234798550804231723699936313409552005344369011471025015797929522515563946378544526618435286424685352300392958179399253823096026532098336787239099266998655972315454581962701835089388667304253277895755816359143339663424859787160
Y:
64884489142937397493713427013475048956265588932163987768937429650402685636522468609085906613213965715574365696174407414363654194179779368075842307204959016994937447552395592348937422905709845922160341625098910833019439194730132761857663846282184246055329761146225184539248986751531674806039985314126288415040
K:
28718467441366289059137483429654897918962933116947401908907126460533449664703636933001281690798345252416809010662058643366168521765491502447001596778004865680429589348977698908702097589489937854279603757612951193880255450453936831323325723396924014112824534345127634978892065265548627093065346785945201821109
but I don't know libfreenet well enough just yet to see if it calculated
the same key.
> Let me know if that doesn't fix the request problem as well.
Nope.
> My next release will have makefiles for the rijndael and twofish code,
> so hopefully this won't be a problem in the near future.
>
> -S
>
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://www.uprizer.com/mailman/listinfo/devl
>
--
Mark Roberts
mjr at statesmean.com
--__--__--
Message: 6
To: devl at freenetproject.org
Subject: Re: [freenet-devl] Espra - what's that?
From: Steven Hazel <[email protected]>
Date: 05 Feb 2001 19:28:49 -0600
Reply-To: devl at freenetproject.org
"Mark J. Roberts" <mjr at statesmean.com> writes:
> Now inserts fail with -9 as well.
Just to eliminate some variables, what platform are you trying this
on? And what version of openssl are you using?
> Whiterose gives me "Error in crypto handshake" and Fred is weirder. The
> first time I try, it gives me something like:
>
> Accepted connection:tcp/127.0.0.1:2001
> New connectionhandler with tcp/127.0.0.1:2001
> Authenticating tcp/127.0.0.1:2001
> tcp/127.0.0.1:2001 authenticated
> Finished with connection - closing
>
> and subsequent attempts fail like
>
> Authenticating tcp/127.0.0.1:1977
> Failed to initialize connection: java.net.SocketException: Connection reset
> by peer: Connection reset by peer
>
> presumably because the first handshake is still valid?
So far as I can tell, it hasn't made it to handshaking at this point
-- it's dying in the initial key exchange, which doesn't maintain
validity over connections. There are some debugging printfs that I
deliberately left in protocol.c which should print out every message
sent or received by libfreenet -- if you're not seeing those, messages
aren't being sent.
> Anyway the DH params seem valid on Fred's side:
>
> but I don't know libfreenet well enough just yet to see if it calculated
> the same key.
You can insert these snippets of code at these lines in protocol.c to
see what those values are (all line numbers are before you make any
modifications, so mark the lines for yourself before you insert any
newlines):
line 263:
fprintf(stderr, "Y: ");
BN_print_fp(stderr, y);
fprintf(stderr, "\n");
line 308:
fprintf(stderr, "X: ");
BN_print_fp(stderr, pub_key);
fprintf(stderr, "\n");
line 320:
fprintf(stderr, "K: ");
BN_print_fp(stderr, k);
fprintf(stderr, "\n");
-S
--__--__--
Message: 7
Date: Mon, 5 Feb 2001 17:55:24 -0800
From: Ian Clarke <[email protected]>
To: devl at freenetproject.org
Subject: [freenet-devl] IMPORTANT: Where is the Freenet Windows release?
Reply-To: devl at freenetproject.org
--GPJrCs/72TxItFYR
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
The Freenet Windows release has gone missing (it was there this-morning)
- can whoever has a copy please replace it?!
Ian.
--GPJrCs/72TxItFYR
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE6f1mMQtgxRWSmsqwRAtedAJ4nDK3G5HtvMl/6D0OQKYyvEPcc5QCeOKGG
o+nKN+u620gVLygl+FzLj0c=
=lJX2
-----END PGP SIGNATURE-----
--GPJrCs/72TxItFYR--
--__--__--
Message: 8
Date: Mon, 5 Feb 2001 22:02:29 -0500
From: Benjamin Coates <[email protected]>
To: devl at freenetproject.org
Subject: RE: Killing Freenet (Re: [freenet-devl] Aardvark)
Reply-To: devl at freenetproject.org
>From Peter Todd <retep2 at home.com>
> I delibrately set my
>nodes informDelay to 0 and restart it every 24 hours for this reason
>and to stop memory and hd leaks. (garbage collection isn't perfect)
I've found that 0.3.7 and some post-0.3.6 trunk builds no longer have this
problem, and stay in the 20 MB ballpark even when run for long stretches of
time. It also seems to stop responding less often now, but it's hard to tell.
--
Benjamin Coates
--__--__--
Message: 9
From: "Fred Salzer" <[email protected]>
To: <devl at freenetproject.org>
Subject: RE: Killing Freenet (Re: [freenet-devl] Aardvark)
Date: Mon, 5 Feb 2001 19:40:29 -0800
Reply-To: devl at freenetproject.org
The corrupted store_<port> problem apparently was fixed. I haven't had that
problem since Sebastian's first version of "0.3.6.1" and 0.3.7 on Win2K/JDK
1.3 runs roughly at 20MB for me also, even when requesting or inserting
files. However, I've had 0.3.7 go to 99% CPU twice but have not determined
under what conditions.
Fred
-----Original Message-----
On February 05, 2001 19:02 Benjamin Coates wrote:
I've found that 0.3.7 and some post-0.3.6 trunk builds no longer have this
problem, and stay in the 20 MB ballpark even when run for long stretches of
time. It also seems to stop responding less often now, but it's hard to
tell.
--__--__--
Message: 10
Date: Mon, 5 Feb 2001 23:04:23 -0500 (EST)
From: Chris Anderson <[email protected]>
To: devl at freenetproject.org
Subject: Re: [freenet-devl] Aardvark
Reply-To: devl at freenetproject.org
On Mon, 5 Feb 2001, I wrote:
> With a 4.2 mb mp3 file it takes 11.96 secs for RequestClient to transfer
> it by CHK.
>
> Without rijndael cipher streams it takes 6.65 secs
>
> Without rijndael and without twofish, but with twofish's cipherstream
> (disable twofish.encipher), it takes 5.29 secs
>
> Without any cipherstreams, it take 4.16 secs
>
> Without the sha1 verify stream it takes 1.99 secs
>
> I'm still digesting...
>
Looks like it may be possible to knock the time down to 8.86 secs without
too much work, but there's a catch.
It turns out that the synchronized flag is fairly expensive on the
PCFBMode.[en|de]cipher methods and on the SHA1.update method. The
CipherStream's have it as a private member, so the only way PCFBMode
can be clobbered is if two threads write to the same stream at the
same time, and my guess is that this may happen if a message needs to
break a Conduit. The SHA1 update sync flag is another story, they
are used on so many places...
Here's a summary of the time in modules from the above run:
rijndael: 1.52 * 2
cipher-stream: 1.13 * 3
twofish: 1.36 * 1
hash-verify: 2.17 * 1
other: 1.99
Just removing the sync flags cuts the cipher-stream and hash-verify times
down by about .3 secs. I ended up doing some monster inlining on the
SHA1 transform to cut another .55 secs off of the verify stream. So, now
we have this:
rijndael: 1.52 * 2
cipher-stream: .71 * 3
twofish: 1.36 * 1
hash-verify: 1.30 * 1
other: 1.99
I don't see much that can be done with the two ciphers, but the 1.99 other
category which is mostly at the client, looks mighty suspicious...
--__--__--
Message: 11
From: "tav" <[email protected]>
To: <devl at freenetproject.org>
Subject: Re: [freenet-devl] Espra - what's that?
Date: Tue, 6 Feb 2001 04:18:04 -0000
Organization: esp worldwide ltd
Reply-To: devl at freenetproject.org
some people wanted it, so here's a screenshot of espra...
http://espra.net/screenshots/espra01
comments / thoughts much appreciated. thanks
--
best regards, tav
tav at espnow.com
--__--__--
Message: 12
Date: Mon, 5 Feb 2001 20:19:07 -0800
From: Don Marti <[email protected]>
To: devl at freenetproject.org
Subject: Re: [freenet-devl] Compression of Jars
Reply-To: devl at freenetproject.org
On Mon, Feb 05, 2001 at 03:24:19PM -0800, Mr. Bad wrote:
> My guess is that the jar tool uses much more conservative compression
> than either zip (that's infozip, btw) or gzip. Once the file is
> compressed, most of the useful redundancy has been removed, so it
> won't compress much more.
My guess is that it uses compress-then-archive (like BRU) instead of
archive-then-compress (like .tar.gz) Compress-then-archive makes it more
likely that you can recover all but one file from a corrupted archive.
Even Sun isn't retarded enough to pick a compression algorithm that
makes files twice as big as zip or gzip.
--
Don Marti "I've never sent or received a GIF in my life."
dmarti at zgp.org -- Bruce Schneier, Secrets and Lies, p. 246.
http://zgp.org/~dmarti/ (Free the Web: http://burnallgifs.org/)
--__--__--
Message: 13
Date: Tue, 6 Feb 2001 00:27:32 -0500
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
From: "Scott G. Miller" <[email protected]>
Reply-To: devl at freenetproject.org
--YiEDa0DAkWCtVeE4
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Mon, Feb 05, 2001 at 10:02:29PM -0500, Benjamin Coates wrote:
> >From Peter Todd <retep2 at home.com>
>=20
> > I delibrately set my
> >nodes informDelay to 0 and restart it every 24 hours for this reason
> >and to stop memory and hd leaks. (garbage collection isn't perfect)
>=20
> I've found that 0.3.7 and some post-0.3.6 trunk builds no longer have thi=
s=20
> problem, and stay in the 20 MB ballpark even when run for long stretches =
of=20
> time. It also seems to stop responding less often now, but it's hard to =
tell.
Try the current CVS. I may have found the 'stop responding' problem.
Scott
--YiEDa0DAkWCtVeE4
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE6f4tDr9IW4v3mHtQRAkLxAJwLUYqg/Mh3FTHoZDbqJ9F1r2V7mACgkvfz
/7PQaZEPiLyzI4CsDk2EIHc=
=WYh1
-----END PGP SIGNATURE-----
--YiEDa0DAkWCtVeE4--
--__--__--
Message: 14
To: devl at freenetproject.org
Subject: Re: [freenet-devl] Compression of Jars
From: Steven Hazel <[email protected]>
Date: 06 Feb 2001 00:28:58 -0600
Reply-To: devl at freenetproject.org
Don Marti <dmarti at zgp.org> writes:
> My guess is that it uses compress-then-archive (like BRU) instead of
> archive-then-compress (like .tar.gz) Compress-then-archive makes it
> more likely that you can recover all but one file from a corrupted
> archive. Even Sun isn't retarded enough to pick a compression
> algorithm that makes files twice as big as zip or gzip.
Compress-then-archive will compress considerably less well than
archive-then-compress, since compress works better when applied to
more data. And unless pkzip, infozip, and winzip can all decode
compress-then-archive files without any special indications that they
need to do so, that's not how jars work.
-S
--__--__--
_______________________________________________
Devl mailing list
Devl at freenetproject.org
http://www.uprizer.com/mailman/listinfo/devl
End of Devl Digest
