[freenet-support] does freenet slow down the computer
There should be bandwidth limits somewhere in the config file. I'm using 0.5myself, I'm guessing you're on 0.7, but look around in freenet.conf or whatever configuration files you can find for some kind of input and output bandwidth limits. I suggest running a speed test on your connection (just google 'internet speed test', there's tons of them out there), and rounding the speeds down to the nearest 100kbps or so and setting that as your limits. I've actually found that lowering the OUTGOING bandwidth limit a bit helps my browsing speed a lot more than limiting the incoming bandwidth. On Oct 31, 2007 4:51 PM, its me wrote: > *After I installed freenet, when the computer is idle, it is sending out > 56338 packets and receiving 139442 packets even when no program is running. > * > ** > *Is this activity all from freenet and if so, is that why my computer is > very slow when I am trying to browse regular pages on the internet ? * > ** > *thanks for any help you can provide* > ** > *Brian* > > ___ > Support mailing list > Support at freenetproject.org > http://news.gmane.org/gmane.network.freenet.support > Unsubscribe at > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support > Or mailto:support-request at freenetproject.org?subject=unsubscribe > -- http://www.spreadfirefox.com/?q=affiliatesid=0t=57;>http://sfx-images.mozilla.org/affiliates/Buttons/180x60/blank.gif"/> -- next part -- An HTML attachment was scrubbed... URL: <https://emu.freenetproject.org/pipermail/support/attachments/20071102/c4dc620a/attachment.html>
[freenet-support] Unable to delete "Never Connected" peers
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, i've just noticed that my FProxy goes crazy when i try to delete "Never Connected" peers. If i select more than one it only deletes the top-most and never finishes loading the page, thus i have to stop and go to /friends/ manually. I think that recently (relatively recently) the new ability was added that forced the peer to remove you when you removed that peer... could this be a result of that? - Volodya - -- http://freedom.libsyn.com/ Voice of Freedom, Radical Podcast http://freeselfdefence.info/ Self-defence wiki http://www.kingstonstudents.org/ Kingston University students' forum "None of us are free until all of us are free."~ Mihail Bakunin -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHK1N2uWy2EFICg+0RAhhwAKCzBsGOLW27ctLma0ER42GaNWkgfACgmz+6 FMaryackz9L2GRUqUC8rjeU= =6nBZ -END PGP SIGNATURE-
[freenet-support] Req: official release of patch for 0.5
According to the open source -principle, there definitely should be a solution for this. The current dictatorship of "knowing better" than the users what is good for them and what is not is more like Apple than an open source -project. If users want to have an opennet with loads of peers, why not allow them to have it? Maybe the users are right - and if not, they will come back to the 30 connections. Of course, there could be a branching of freenets (only nets or also codes) where one would be code-customized by users and the other would be a technology development project by enthusiastic developers. A bit like Debian and its derivatives. However, this is hardly good for the cause or justified with a much smaller project where different lines can be easily combined within a single project with some flexibility towards the wants of users. It would be better to have, even at the loss of completely homogenous network, a version of freenet 0.7 that would resemble freenet 0.5 enough to close the old net down. That is, only opennet, and a freely configurable amount of connections. Very probably the increasing amount of nodes in the 0.7 network would outweigh the loss of rigidity in the network due to nodes with loads of connections. The developers are right in one thing though - it would probably not be good to have a gui -configurable option for more connections, because then also the basic Windows -(l)users would just tune it up because "bigger is better". However, having a line in the config file for this that is not that obvious would get rid of the problem. Label it "conntrack" an define the value in binary if you like - but just give those willing to test the ability to do so and involve more users in the development. -- Malkus Lindroos Nomen Nescio wrote: > (please excuse me if this is a duplicate) > > Toad, > > Recently a message by you was xposted to 0.5 about a crypto weakness you > folks discovered in 0.7 that 0.5 is also subject to. > > After some discussion, one individual came up with a patch and procedure for > fixing this on 0.5. While a lot will simply use that patch and build new > freenet.jar executables, there are some who aren't up to it for whatever > reason. > > Given the patch and the msg announcing it, would you be willing to create an > 'official' freenet 0.5 build 5108 and make it available at the same place > where the old 0.5 stuff is at http://downloads.freenetproject.org/ ? > > BTW- I know you and other freenet devs haven't been in the habit of doing so > but would you also be willing to include a digital signature of some kind? > At least an SHA-1 hash, preferably a detached pgp signature. > > Thanks > > Here's the message I mentioned, followed by the patch: > > > ...
[freenet-support] I almost forgot
In my earlier request for the crypto fix build for 0.5, I forgot to ask if you'd also be willing to incorporate the "painkiller" mods in 5108? Thanks again
[freenet-support] Req: official release of patch for 0.5
(please excuse me if this is a duplicate)
Toad,
Recently a message by you was xposted to 0.5 about a crypto weakness you folks
discovered in 0.7 that 0.5 is also subject to.
After some discussion, one individual came up with a patch and procedure for
fixing this on 0.5. While a lot will simply use that patch and build new
freenet.jar executables, there are some who aren't up to it for whatever reason.
Given the patch and the msg announcing it, would you be willing to create an
'official' freenet 0.5 build 5108 and make it available at the same place where
the old 0.5 stuff is at http://downloads.freenetproject.org/ ?
BTW- I know you and other freenet devs haven't been in the habit of doing so
but would you also be willing to include a digital signature of some kind? At
least an SHA-1 hash, preferably a detached pgp signature.
Thanks
Here's the message I mentioned, followed by the patch:
- Jack O'Lantern - 2007.10.31 - 20:36:40GMT -
Happy Hallowe'en folks,
a certain Toad recently informed us that the Diffie-Hellman key exchange in
freenet 0.5 has been b0rked for about two years and we please bob for the
solution to the issue ourselves. So, here goes...
1. obtain a JDK (sun5 works), svn, ant and junit.
2. obtain the source:
$ svn co http://freenet.googlecode.com/svn/branches/legacy/stable/
freenet-0.5-5108
$ cd freenet-0.5-5108
$ svn co http://freenet.googlecode.com/svn/branches/legacy/contrib/ contrib
3. copy freenet-ext.jar from your freenet directory into the newly created
freenet-0.5-5108 directory (freenet-ext.jar remains unchanged)
4. obtain the patch I crafted:
freenet:CHK at
Sm50039W8Gt8kzLyMloGB6pvjkQNAwI,xYB4spgd2g1ZtJIYN0lfeg/freenet_dh.patch
5. Examine this patch closely. You don't know me. You never know beforehand if
you're about to be tricked or treated, or handled by an incompetent person.
This patch does the following things:
a. creates a method to check for weak Diffie-Hellman exponents (imported
from freenet 0.7)
b. there are *extremely* few weak exponentials in the number space, so
accidental creation of a weak exponential is *extremely* unlikely;
nevertheless, the patch adds code to prevent creating weak exponents (imported
from freenet 0.7)
c. adds code to abort a Diffie-Hellman key exchange if our peer ("Bob") uses
a weak key (logs an error) (two instances)
d. bumps the build number up to 5108. I feel it's justified that we users
hijack the build numbering scheme as the developers don't maintain it anymore.
6. If you're satisfied, copy this patch into the freenet-0.5-5108 directory.
7. Apply the patch:
$ patch -p1 < freenet_dh.patch
8. Build freenet.jar
$ CLASSPATH=freenet-ext.jar ant distclean dist
9. Make a backup of your old freenet.jar (freenet-ext.jar remains in place).
10. Copy the newly built freenet.jar to your freenet directory.
$ cp lib/freenet.jar /path/to/freenet/dir
11. Restart freenet
That's it. Enjoy your shiny new freenet build.
It would be great if someone in contact with the freenet 0.7 developers could
communicate the patch to them. I'm just a user, and some official
acknowledgement that the patch does indeed fix *all* instances of weak
Diffie-Hellman handling is appreciated because, you know, the comments in the
freenet source are not exactly abundant and it's not easy for a newcomer to
find one's way through.
EOM
Here's the patch:
diff -Naur freenet-0.5-5107/build.xml freenet-0.5-5108/build.xml
--- freenet-0.5-5107/build.xml 2007-10-31 18:12:46.0 +0100
+++ freenet-0.5-5108/build.xml 2007-10-31 18:22:13.0 +0100
@@ -22,8 +22,8 @@
-
-
+
+
http://securitytracker.com/alerts/2005/Aug/1014739.html
+* @see http://www.it.iitb.ac.in/~praj/acads/netsec/FinalReport.pdf
+*/
+ public static boolean checkDHExponentialValidity(Class caller,
BigInteger exponential) {
+ int onesCount=0, zerosCount=0;
+
+ // Ensure that we have at least 16 bits of each gender
+ for(int i=0; i < exponential.bitLength(); i++)
+ if(exponential.testBit(i))
+ onesCount++;
+ else
+ zerosCount++;
+ if((onesCount<16) || (zerosCount<16)) {
+ Core.logger.log(caller, "The provided exponential
contains "+zerosCount+" zeros and "+onesCount+" ones wich is unacceptable!",
Logger.ERROR);
+ return false;
+ }
+
+ // Ensure that g^x > 2^24
+ if(MIN_EXPONENTIAL_VALUE.compareTo(exponential) > -1) {
+ Core.logger.log(caller, "The provided exponential is
smaller than 2^24 which is unacceptable!", Logger.ERROR);
+ return false;
+ }
+ // Ensure that g^x < (p-2^24)
+
Re: [freenet-support] does freenet slow down the computer
There should be bandwidth limits somewhere in the config file. I'm using 0.5myself, I'm guessing you're on 0.7, but look around in freenet.conf or whatever configuration files you can find for some kind of input and output bandwidth limits. I suggest running a speed test on your connection (just google 'internet speed test', there's tons of them out there), and rounding the speeds down to the nearest 100kbps or so and setting that as your limits. I've actually found that lowering the OUTGOING bandwidth limit a bit helps my browsing speed a lot more than limiting the incoming bandwidth. On Oct 31, 2007 4:51 PM, its me [EMAIL PROTECTED] wrote: *After I installed freenet, when the computer is idle, it is sending out 56338 packets and receiving 139442 packets even when no program is running. * ** *Is this activity all from freenet and if so, is that why my computer is very slow when I am trying to browse regular pages on the internet ? * ** *thanks for any help you can provide* ** *Brian* ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:[EMAIL PROTECTED] -- HTML a href=http://www.spreadfirefox.com/?q=affiliatesamp;id=0amp;t=57;img border=0 alt=Get Firefox! title=Get Firefox! src= http://sfx-images.mozilla.org/affiliates/Buttons/180x60/blank.gif//a ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:[EMAIL PROTECTED]
