|
Home > Archive > Anonymous Servers > October 2006 > Something wrong with remailers?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Something wrong with remailers?
|
|
| Anonymous via Runaway Remailer 2006-10-18, 1:13 pm |
| hey all,
what's going on? been sending test messages to alt.test through individual
remailers, seems none are getting through except for runaway. I update my
keys/stats from banana. is this a reliable sorce? thx
~~~~~~~~~~~~~~~~~~~~~
This message was posted via one or more anonymous remailing services.
The original sender is unknown. Any address shown in the From header
is unverified.
| |
|
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512
On 18 Oct 2006 14:35:21 -0000, Anonymous via Runaway Remailer wrote in
Message-Id: <M4RREVBT39008.6912152778@anonymous.poster>:
> hey all,
> what's going on? been sending test messages to alt.test through individual
> remailers, seems none are getting through except for runaway.
Nothing wrong with remailers at the moment, but those that deliver via
dizum and/or lcs are having trouble reaching Usenet. Check out:-
http://echolot.theremailer.net/side...list2-post.html
This is currently the only stats source that's based on Usenet posting
rather than email.
> I update my keys/stats from banana. is this a reliable sorce? thx
No, not in so far as posting to Usenet is concerned. If mail2news
gateways are down (as they are currently), your postings are likely to
go missing.
You might want to check out these (doctored) stats:-
http://stats.bananasplit.info/type2.list
http://stats.bananasplit.info/pubring.mix
Leave your mlist2 source as it is.
These stats strip the Post capability from remailers that are currently
failing to return News pings. If you use these with a Post directive,
you should see higher returns.
Note: You must use a Post directive for this to work effectively, not
hardcode mail2news gateways.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
iQEVAwUBRTZI22oLu9HNUqmMAQpDVwf/ceyH3eUjTDCZKnZali+tETs1R2jY4V9K
ogt2nGhg2vLz7koBO8QTfX9ka7BxrdfvQq44e8nQ
bmpDstAC1qzXJ2VjAaa2dN+U
pO7uvNRSXUmTSyWbTkonDmwgjTG8SRokOcOlWg7o
I475gdYqUyCMCWHbQIgqjsIx
d4Q1WIRTx13E5smOFPzt+9x2w7RHUTWIKihMdNZh
OW+sQ434GzOLeqAKcf/3utYG
v5AyoATJN5s28cQtSEm3d0Za9F+GCHiHok/j3PLQ0/fgqdiTP/eArE4bF1qr2sYq
Sh6MUmVImz2DH2QapkLYS/GKrOhzL74NR/kUCRubz+BQux/FJBiRCw==
=mvIc
-----END PGP SIGNATURE-----
--
pub 1024D/8ED57743 2003-07-08 Bananasplit Operator
Key fingerprint = 796F 67E0 E890 A0BB BDAE EBB4 94A6 7A09 8ED5 7743
uid Admin <admin.bananasplit.info>
| |
| Anonymous via Runaway Remailer 2006-10-18, 1:13 pm |
| Zax <admin@bananasplit.info> wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA512
>
>On 18 Oct 2006 14:35:21 -0000, Anonymous via Runaway Remailer wrote in
>Message-Id: <M4RREVBT39008.6912152778@anonymous.poster>:
>
>
>Nothing wrong with remailers at the moment, but those that deliver via
>dizum and/or lcs are having trouble reaching Usenet. Check out:-
>http://echolot.theremailer.net/side...list2-post.html
>This is currently the only stats source that's based on Usenet posting
>rather than email.
Thx for the info. for about the past week or so i've been posting via the
Anon-Post-To: method cuz it seemed the M2N gateways were pure crap and
nothing was getting through (seems many posts here confirm that), and I was
sending 3 copies, Ics, banana and uk.org. the APT method was working fine
with all rm's with Post: cap. up until yesterday, when it seems nothing
will get through using that method either (except for runaway).
>No, not in so far as posting to Usenet is concerned. If mail2news
>gateways are down (as they are currently), your postings are likely to
>go missing.
>
>You might want to check out these (doctored) stats:-
>http://stats.bananasplit.info/type2.list
>http://stats.bananasplit.info/pubring.mix
>Leave your mlist2 source as it is.
>
>These stats strip the Post capability from remailers that are currently
>failing to return News pings. If you use these with a Post directive,
>you should see higher returns.
will do. but just curious, what exactly is it that sometimes cause this to
happen, or at least what caused it in this case? just seems so weird. one
day everything working fine, then the next day almost nothing working at
all. very bizarre
>
>Note: You must use a Post directive for this to work effectively, not
>hardcode mail2news gateways.
That's what I'm doing now, using Post or Anon Post To (with CP). Still
nothing. But will check into the info you gave me. Thx again.
>
>-----BEGIN PGP SIGNATURE-----
>Version: GnuPG v1.4.5 (GNU/Linux)
>
>iQEVAwUBRTZI22oLu9HNUqmMAQpDVwf/ceyH3eUjTDCZKnZali+tETs1R2jY4V9K
> ogt2nGhg2vLz7koBO8QTfX9ka7BxrdfvQq44e8nQ
bmpDstAC1qzXJ2VjAaa2dN+U
> pO7uvNRSXUmTSyWbTkonDmwgjTG8SRokOcOlWg7o
I475gdYqUyCMCWHbQIgqjsIx
> d4Q1WIRTx13E5smOFPzt+9x2w7RHUTWIKihMdNZh
OW+sQ434GzOLeqAKcf/3utYG
>v5AyoATJN5s28cQtSEm3d0Za9F+GCHiHok/j3PLQ0/fgqdiTP/eArE4bF1qr2sYq
>Sh6MUmVImz2DH2QapkLYS/GKrOhzL74NR/kUCRubz+BQux/FJBiRCw==
>=mvIc
>-----END PGP SIGNATURE-----
~~~~~~~~~~~~~~~~~~~~~
This message was posted via one or more anonymous remailing services.
The original sender is unknown. Any address shown in the From header
is unverified.
| |
|
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512
On 18 Oct 2006 16:00:00 -0000, Anonymous via Runaway Remailer wrote in
Message-Id: <CJC7II4D39008.75@anonymous.poster>:
> will do. but just curious, what exactly is it that sometimes cause this to
> happen, or at least what caused it in this case? just seems so weird. one
> day everything working fine, then the next day almost nothing working at
> all. very bizarre
Most remailers post via mail2news gateways, and the most popular choices
are Dizum and LCS. LCS failed about 11 days ago and Dizum went down
yesterday. As a result there are currently only 5 remailers averaging
over 95% success posting to Usenet:-
banana 112011221121 :32 ++++++++++++ 100.0% PR IN1
panta 4B3HHHHHEC32 18:55 ++++++++++++ 100.0% PRHG XATLEUIN9
frell AA603253352A 3:00 +++++++++++8 98.0% PR GO ATLEUIN9
eelbash 00A000000018 :13 +++5++++++++ 98.0% PRHG XATLEUIN0
bikikii C5??32C23325 2:48 ++00++++++++ 95.1% DPRHGOXATLEUIN1
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
iQEVAwUBRTZm/ 2oLu9HNUqmMAQo6uQf7BLETbBUxhl1R69JBQ0mRX
X3xyPxKKPg7
QJm/ tqELFDQuT1zH1JuJ8upvbYqR4Z3S4baOP5tGE8UI
8x3nnoFSWThkZsI/fHEA
DJb8xL9t2G43Nm5PV+W/nzvHWbB5+NU/DecZC9mKLjL1mFJxACCAjd8aGUgrLexu
HAoSMWXJCvkhEn7WT6ixGVCe+UaWb7pKMmwD9CUn
7D2XXEcirAzaJ8g3saJK2oBK
WNt+wvQXWrazTR6VrlvPC98a5LbifSxr5j+l2AF2
aClRRWWOZPCL7QkkIb+2kp7r
s0BT6ew+gW1n9omeeRsDuJp3e6zWR0D3xEu+oPKJ
EMpzGyLsp9R8LA==
=SJa4
-----END PGP SIGNATURE-----
--
pub 1024D/8ED57743 2003-07-08 Bananasplit Operator
Key fingerprint = 796F 67E0 E890 A0BB BDAE EBB4 94A6 7A09 8ED5 7743
uid Admin <admin.bananasplit.info>
| |
| Borked Pseudo Mailed 2006-10-18, 7:14 pm |
| In article <CJC7II4D39008.75@anonymous.poster>
Anonymous via Runaway Remailer <secret_anonymous_1@web.de> wrote:
>
> Zax <admin@bananasplit.info> wrote:
>
> Thx for the info. for about the past week or so i've been posting via the
> Anon-Post-To: method cuz it seemed the M2N gateways were pure crap and
> nothing was getting through (seems many posts here confirm that), and I was
> sending 3 copies, Ics, banana and uk.org. the APT method was working fine
> with all rm's with Post: cap. up until yesterday, when it seems nothing
> will get through using that method either (except for runaway).
Just a note:
The mail2news@remailer.org.uk (amigo) has been shut down for quite some
time. If you've been using that it's been doing you no good anyway.
| |
| Anonymous 2006-10-18, 7:14 pm |
| In article <CJC7II4D39008.75@anonymous.poster>
Anonymous via Runaway Remailer <secret_anonymous_1@web.de> wrote:
>
> Zax <admin@bananasplit.info> wrote:
>
>
> Thx for the info. for about the past week or so i've been posting via the
> Anon-Post-To: method cuz it seemed the M2N gateways were pure crap and
> nothing was getting through (seems many posts here confirm that), and I was
> sending 3 copies, Ics, banana and uk.org. the APT method was working fine
> with all rm's with Post: cap. up until yesterday, when it seems nothing
> will get through using that method either (except for runaway).
Other posts dispute that nothing was getting through and that M2N gateways
were pure crap. It's much more of a problem of users of the system not
taking the time to stay abreast of the situation, educate themselves and
use the system to their advantage. After all, it's so much easier to whine.
| |
| Anonymous 2006-10-19, 1:12 am |
| On Thu, 19 Oct 2006, Anonymous <cripto@ecn.org> wrote:
> It's much more of a problem of users of the system not
>taking the time to stay abreast of the situation, educate themselves and
>use the system to their advantage. After all, it's so much easier to whine.
by your own standards, you're whining about somebody else's supposed
"whining"
| |
| Ragnarok 2006-10-19, 1:12 am |
| In article
<0abba634d8b445b97f405a80261a365e@anon.mixmaster.mixmin.net>
Anonymous <nobody@mixmin.net> wrote:
>
> On Thu, 19 Oct 2006, Anonymous <cripto@ecn.org> wrote:
>
themselves and[vbcol=seagreen]
easier to whine.[vbcol=seagreen]
>
> by your own standards, you're whining about somebody else's
supposed
> "whining"
Well, I suppose he's as entitled to whine as anyone else.
But he's not contributing anything to the discussion.
The people who posted working m2n gateways, and the ones who
posted the fact that BOTH lcs and dizum are currently inop are
the ones who are contributing.
~~~~~~~~~~~~~~~~~~~~~
This message was posted via one or more anonymous remailing services.
The original sender is unknown. Any address shown in the From header
is unverified.
| |
| Anonymous 2006-10-19, 7:12 pm |
| In article <0abba634d8b445b97f405a80261a365e@anon.mixmaster.mixmin.net>
Anonymous <nobody@mixmin.net> wrote:
>
> On Thu, 19 Oct 2006, Anonymous <cripto@ecn.org> wrote:
>
>
> by your own standards, you're whining about somebody else's supposed
> "whining"
No, my standard of "whining" defines it as somebody continually making
blanket untrue statements like "nothing's working", calling the system crap
and generally acting poor poor pitiful them when the answers to their
problems are answered year after year, almost daily. Usually I'm one of
those who patiently answers.
Once upon a time I was one of those whiners. All it took was one of you
assholes to tell me "quit the XXXX whining" and I did, fixing my own
problems and lack of understanding. This resulted in near perfect
reliability for my posts to Usenet since. It would take the simultaneous
collapse of every mail2news plus the panta, frell, and banana remailers
before I might consider whining again, and then I wouldn't be able to do it.
| |
| Odious Ferret 2006-10-20, 1:12 am |
| In article <e72ccedfa607ddc3016b81f6e9871202@ecn.org>
Anonymous <cripto@ecn.org> wrote:
>
> In article <0abba634d8b445b97f405a80261a365e@anon.mixmaster.mixmin.net>
> Anonymous <nobody@mixmin.net> wrote:
>
> No, my standard of "whining" defines it as somebody continually making
> blanket untrue statements like "nothing's working", calling the system crap
> and generally acting poor poor pitiful them when the answers to their
> problems are answered year after year, almost daily. Usually I'm one of
> those who patiently answers.
>
> Once upon a time I was one of those whiners. All it took was one of you
> assholes to tell me "quit the XXXX whining" and I did, fixing my own
> problems and lack of understanding. This resulted in near perfect
> reliability for my posts to Usenet since. It would take the simultaneous
> collapse of every mail2news plus the panta, frell, and banana remailers
> before I might consider whining again, and then I wouldn't be able to do it.
More whining. You have a near perfect record for whining too.
| |
| Anonymous 2006-10-20, 1:12 am |
| In article <4538165c$0$17715$ed362ca5@nr3.newsreader.com>
Odious Ferret wrote:
>
> In article <e72ccedfa607ddc3016b81f6e9871202@ecn.org>
> Anonymous <cripto@ecn.org> wrote:
>
> More whining. You have a near perfect record for whining too.
Sure, why not. Seek perfection in everything.
| |
| madmax 2006-10-21, 1:13 am |
| On Thu, 19 Oct 2006, Anonymous <cripto@ecn.org> wrote:
>Once upon a time I was one of those whiners. All it took was one of you
>assholes to tell me "quit the XXXX whining" and I did, fixing my own
>problems and lack of understanding. This resulted in near perfect
>reliability for my posts to Usenet since. It would take the simultaneous
>collapse of every mail2news plus the panta, frell, and banana remailers
>before I might consider whining again, and then I wouldn't be able to do it.
hey, good joke
since you write well, why not compose a simple posting guide (client
independent) and post it once a week or so?
| |
| Anonymous 2006-10-21, 1:13 am |
| On 20 Oct 2006, madmax <anon@comments.header> wrote:
>On Thu, 19 Oct 2006, Anonymous <cripto@ecn.org> wrote:
>
>
>hey, good joke
>
>since you write well, why not compose a simple posting guide (client
>independent) and post it once a week or so?
Why don't you?
| |
| Anonymous 2006-10-21, 1:13 am |
| In article <9IIMV52S39010.662025463@twistycreek.com>
madmax <anon@comments.header> wrote:
>
> On Thu, 19 Oct 2006, Anonymous <cripto@ecn.org> wrote:
>
>
> hey, good joke
>
> since you write well, why not compose a simple posting guide (client
> independent) and post it once a week or so?
Maybe I will. But until then, my biggest pieces of advice are to stay
personally informed and flexible. Before going off on any posting spree,
update, *read* and understand your stats. Then take a look at
http://stats.bananasplit.info/, http://stats.bananasplit.info/failed.html
for obvious problems. http://www.cotse.net/users/drsnoid/results.htm is
great for historical trends if not up-to-the-day status.
Doing the above should take all of five minutes. There's a ton more too,
when you have the time.
Don't feel tied to any particular chain or posting method that has worked
forever for you. Mix it up a lot. See how things behave and contemplate
why. Then when your perusal of the stats and other information (as well as
the discussions here) indicate there might be bumps in the road, probable
workarounds will become second nature.
It would be nice if Zax's dream of type II remailing just working out of
the box were a reality. It doesn't though - it takes some personal
involvement. Mixmaster and your client can't do it all for you. In my
case that's what interests me so much about it. It can be a challenge, it
can take some expertise - but it's never broken and it's never crap.
Just get good at it.
| |
| Echeloff 2006-10-21, 1:17 pm |
| >It would be nice if Zax's dream of type II remailing just working out of
>the box were a reality. It doesn't though - it takes some personal
>involvement. Mixmaster and your client can't do it all for you. In my
>case that's what interests me so much about it. It can be a challenge, it
>can take some expertise - but it's never broken and it's never crap.
The only weak point I see is the single exit remailer you have to
depend on. With a pure random chain based on current stats, multiple
copies and a list of several m2n gateways you can add as much
redundancy as required to the rest of the delivery system.
>Just get good at it.
That can't be a serious recommendation for the masses.
Echeloff
| |
| Non scrivetemi 2006-10-21, 7:16 pm |
| >It would be nice if Zax's dream of type II remailing just working out of
>the box were a reality. It doesn't though - it takes some personal
>involvement. Mixmaster and your client can't do it all for you. In my
>case that's what interests me so much about it. It can be a challenge, it
>can take some expertise - but it's never broken and it's never crap.
The only weak point I see is the single exit remailer you have to
depend on. With a pure random chain based on current stats, multiple
copies and a list of several m2n gateways you can add as much
redundancy as required to the rest of the delivery system.
>Just get good at it.
That can't be a serious recommendation for the masses.
Echeloff
| |
|
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512
On 21 Oct 2006 18:11:35 -0000, Echeloff wrote in
Message-Id: <6DQKCPB039011.5497106481@anonymous.poster>:
> The only weak point I see is the single exit remailer you have to
> depend on. With a pure random chain based on current stats, multiple
> copies and a list of several m2n gateways you can add as much
> redundancy as required to the rest of the delivery system.
I think multiple copies is really a result of the reliability issue. If
eveything worked 100%, there would be no need for it. The big problem
as I see it is that the methodology used for producing stats is
different to the method in which the system is used. Stats are produced
by single remailer chains whilst proper usage dictates multiple remailer
chains. Bunker is a classic example of this problem. Hey, identifying
the problem is easy, I don't claim to know the fix. 
Another issue is latency. It takes a long time for a failed remailer to
drop sufficiently in stats that clients stop selecting it. It would be
a nice feature of mixmaster if it could bypass the pool for uptime
pings.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
iQEVAwUBRTqJXWoLu9HNUqmMAQrQNAf+MEG64MJs
C3ZrpGxTLtn284ud6k3aLrgU
rDX2UFw82qV+7/ kqd5ex05W6i9nZALqEOJPSpXZYgICcY9mPUaATCJ
hyfc7UOWm5
CUdxFJlZC+HFXs1FSEGbAH2JHekaz/0P6zYNVNbYNT2llP+AeVOnYx3vzTmUjKxR
IE2QjyIhBk1XSTLfaI9Mtf9fYujmiZ7I3Vuirk3U
D/HjMXT8tZnI8qEaQJVSIibH
CpjChmiKne816+NW12tlrn1w1P7idQHRVuGPOsFV
X3syt4PRUjbEsrPDr/in2HU8
1tpVpiBi22mqIPP1JQ3VOuWn6XXEzArioK3BsbTs
UWoPV22YtnAO1A==
=GS3T
-----END PGP SIGNATURE-----
--
pub 1024D/8ED57743 2003-07-08 Bananasplit Operator
Key fingerprint = 796F 67E0 E890 A0BB BDAE EBB4 94A6 7A09 8ED5 7743
uid Admin <admin.bananasplit.info>
| |
| Anonymous via Panta Rhei 2006-10-21, 7:16 pm |
| In article <6DQKCPB039011.5497106481@anonymous.poster>
Echeloff <echeloff@null.dev> wrote:
>
>
> That can't be a serious recommendation for the masses.
>
> Echeloff
Of course it is. What's the alternative? That we continually coddle those
who come here and say "the remailers suck" rather than the more accurate "I
suck at this"? I don't know about you, but I have better things to do than
hold hands with cluebies. Day after day after day. Somebody needs to have
tough love and say "Quit the XXXX whining. You suck at this. Fix it."
~~~~~~~~~~~~~~~~~~~~~
This message was posted via one or more anonymous remailing services.
The original sender is unknown. Any address shown in the From header
is unverified. You need a valid hashcash token to post to groups other
than alt.test and alt.anonymous.messages. Visit www.panta-rhei.eu.org
for abuse and hashcash info.
| |
| Echeloff 2006-10-22, 1:12 pm |
| Hi Zax!
Zax <admin@bananasplit.info> wrote:
>On 21 Oct 2006 18:11:35 -0000, Echeloff wrote in
>Message-Id: <6DQKCPB039011.5497106481@anonymous.poster>:
>
>
>I think multiple copies is really a result of the reliability issue.
Sure, it's all about reliability. And that reliability must not be
based on the reliability of single remailers, which can't be
guaranteed, but it has to derive from redundancy. That's how the whole
Internet works.
>If
>eveything worked 100%, there would be no need for it.
But then you were speaking of a different scenario, maybe with several
'free' anonymization remailers in the wild and a few sponsored final
hops around the world with 24/7 reliability, financed by donations,
which might bring on new types of problems ;-)
>The big problem
>as I see it is that the methodology used for producing stats is
>different to the method in which the system is used. Stats are produced
>by single remailer chains whilst proper usage dictates multiple remailer
>chains. Bunker is a classic example of this problem. Hey, identifying
>the problem is easy, I don't claim to know the fix.
There are so many traps, that IMHO it's presumptuous to believe, they
all can be analyzed and taken care of in an exhaustive manner. Maybe
the established ones, e.g. remailers that don't deliver to certain m2n
gateways, can be dealt with. But there are unpredictable variable ones
as well. Unfortunately it isn't enough to set MINREL to 100, as the
outcome of a delivery queue is just not predictable by adding the
results of its single components. That's why I see no alternative to
redundancy, and therefore an adequate amount of random chains and m2n
gateways.
>Another issue is latency. It takes a long time for a failed remailer to
>drop sufficiently in stats that clients stop selecting it.
Even without any delay you're not able to foresee the future. There's
no guarantee, the selected '100%' remailers stay in operation till the
work is done. So statistics aren't a cure-all. They are nothing but an
indispensable tool to reduce the great variety of possibilities to the
most promising candidates.
>It would be
>a nice feature of mixmaster if it could bypass the pool for uptime
>pings.
Would you be so kind to explain, what you mean by that? I only could
imagine Mixmaster to weight failed pings differently dependent on how
long ago that happened. Maybe the client software can take care of
such calculations and then hand over a stats file, that was modified
accordingly. But such tactics wouldn't mean a relevant improvement and
thus aren't worth the effort.
It remains to be seen, how much progress Mixminion implies.
Kind regards
Echeloff
| |
| Alex de Joode 2006-10-22, 1:12 pm |
| Echeloff <echeloff@null.dev> wrote:
[..]
: But then you were speaking of a different scenario, maybe with several
: 'free' anonymization remailers in the wild and a few sponsored final
: hops around the world with 24/7 reliability, financed by donations,
: which might bring on new types of problems ;-)
donations I get about e5 per year in donations .. hehe so that's not an
option I guess.
| |
| Echeloff 2006-10-22, 1:12 pm |
| Alex,
>: But then you were speaking of a different scenario, maybe with several
>: 'free' anonymization remailers in the wild and a few sponsored final
>: hops around the world with 24/7 reliability, financed by donations,
>: which might bring on new types of problems ;-)
>
>donations I get about e5 per year in donations .. hehe so that's not an
>option I guess.
that was just an unrealistic model of how to achieve a nearly 100%
success rate. I'm not so sure, that there's no commercial demand for a
bulletproof and easy to use remailer system. However, as long as the
infrastructure isn't reliable enough, the remailer network won't
attract many regular users, who were potential sponsors and would
allow to improve reliability by their donations ...
To be realistic, privacy never was the focus of attention, besides
being stigmatized nowadays, where multi-billion $ companies can demand
whatever they want for their 'gimmicks reloaded' spyware products, and
after they irrevocably established absolute power, to top it all, are
believed to be philanthropists on the basis of nothing but some clever
moves of their PR departments. But as this is an asymmetric
competition, the money those firms spend for toilet paper alone might
suffice to brace remailers for a more optimistic future, as long as
the legislation isn't globally brought into line by the snoopers.
... Sorry, seem to be dreaming once again ;-)
Echeloff
| |
| CL@this.group 2006-10-22, 7:12 pm |
|
adejoode+usenet@sabotage.org (Alex de Joode) wrote in news:ehg1km$s0r$1
@heroin.sabotage.org:
> Echeloff <echeloff@null.dev> wrote:
> [..]
>: But then you were speaking of a different scenario, maybe with several
>: 'free' anonymization remailers in the wild and a few sponsored final
>: hops around the world with 24/7 reliability, financed by donations,
>: which might bring on new types of problems ;-)
>
> donations I get about e5 per year in donations .. hehe so that's not an
> option I guess.
Actually, it might be.
I've donated to causes in the past (usually one time in amounts
of $100.00 to $200.00), but if I were to want to donate to the
"remailer network" I wouldn't even know where to send the money.
Perhaps it's because I'm lazy and haven't tracked down any particular
address, but I suspect there is no centralized address.
Those operating remailers with good reliability deserve recompense.
Donations could be requested "officially" and a single address
or account could be specified for anonymous donations.
It could work, perhaps, but it sure would complicate matters. How would
you distribute the money, and to whom? Who would be in charge of the
money? What rules would determine who was ineligible even if they
operate a reliable remailer?
CL
| |
|
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512
On 22 Oct 2006 13:12:28 -0000, Echeloff wrote in
Message-Id: <KLPW95RZ39012.3419907407@anonymous.poster>:
>
> Sure, it's all about reliability. And that reliability must not be
> based on the reliability of single remailers, which can't be
> guaranteed, but it has to derive from redundancy. That's how the whole
> Internet works.
Hi there, sorry for not responding sooner, I got distracted by other
things yesterday, such as puppies eating my kitchen furniture! :-)
It would be nice to have this redundancy but it would take a major
change to the current design in order to support multiple exit points.
Exit nodes would have to communicate with each other to share info on
who has delivered what messages.
[vbcol=seagreen]
> But then you were speaking of a different scenario, maybe with several
> 'free' anonymization remailers in the wild and a few sponsored final
> hops around the world with 24/7 reliability, financed by donations,
> which might bring on new types of problems ;-)
I don't think the problem is really that bad. At the moment you have a
number of exit-nodes that easily achieve 99% availability. Many
business systems fail to achieve such levels. It's reaction time to
failures that causes the biggest problem.
[vbcol=seagreen]
> There are so many traps, that IMHO it's presumptuous to believe, they
> all can be analyzed and taken care of in an exhaustive manner. Maybe
> the established ones, e.g. remailers that don't deliver to certain m2n
> gateways, can be dealt with.
Your example isn't really a remailer reliability issue. It's a problem
with how remailers interact with external systems like mail2news
gateways. Last year I did some analysis of messages hitting the banana
mail2news gateway. About 20% of messages arriving from remailers were
missing a Newsgroups header. To the sender, this appears as remailer
failure when in reality the remailers have done the job fine. I'm
guessing messages that make it to the entry node in a chain arrive
at the exit node with very high reliability. The majority of problems
are external to the remailer process.
> But there are unpredictable variable ones as well. Unfortunately it
> isn't enough to set MINREL to 100, as the outcome of a delivery queue
> is just not predictable by adding the results of its single
> components. That's why I see no alternative to redundancy, and
> therefore an adequate amount of random chains and m2n gateways.
True, but the focal point of all that redundancy is still the exit
remailer. Doesn't matter how many copies are used or how many recipient
mail2news are specified, if that node fails, messages are lost.
Take the recent closure of bigapple as an example. It's an exit
remailer that's been running with high availability for a long time.
Then suddenly it stops working. Stats (that run on auto-pilot) take
hours, if not days to drop uptime below most users random selection
threshold. No multiple copies or multiple mail2news recipients solve
this.
>
> Even without any delay you're not able to foresee the future. There's
> no guarantee, the selected '100%' remailers stay in operation till the
> work is done. So statistics aren't a cure-all. They are nothing but an
> indispensable tool to reduce the great variety of possibilities to the
> most promising candidates.
Agreed. Stats can never react instantly to remailer failure, but they
could react much quicker than they do now.
>
> Would you be so kind to explain, what you mean by that?
Pings are currently just like any other remailer message, they are sent
using a mixmaster client, through a single node chain (the remailer
being pinged) with a recipient of the pinger. In this manner they are
subjected to the same rules as any other message and spend long periods
of time sat in a Mixmaster pool. Take Frell as an example; it will take
on average 5 hours to get a single ping response from it. If pings
could bypass the pool, the pinger would receive a near-instant response.
> It remains to be seen, how much progress Mixminion implies.
The code for pinging is scheduled for 0.0.8 I believe. This will be a
big improvement as ping functionality is built into the remailer.
You might want to check out http://blog.bananasplit.info/?p=116 for some
ideas I had on improving mixmaster pings.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
iQEVAwUBRTygpGoLu9HNUqmMAQorawf/aAKE16VubNrgDOjw8qM/Xh3BoeNDqJGe
yGVGwCbPFFZ1PiEpAtF98HibdXtAfo+DtMFp03B3
UHZ1yYLON/p2xv2Vlah1rLYx
VbTwcNnPjU8IBl++sbqe1YmAE89+fhy4S3eCy7EQ
usI5DpNRcmTREvv9PGZCmtVj
MHA3jFhC3Z8qhtwi/ gGsrXASE4XN5oRvHpJO5kv5oQq8NZT06PoRvHFKe
xWozIeb
T8K11YEWoegbmOZp3++FIkP3l7g/VlykX+cKL+1l+NONjfQT//4L+EFuSGibXm6A
Kafhxw8poHDiyAkzoorgr1A+DLOgc23K1XYGuJ4O
AWRtV2lQoJZwPg==
=xzya
-----END PGP SIGNATURE-----
--
pub 1024D/8ED57743 2003-07-08 Bananasplit Operator
Key fingerprint = 796F 67E0 E890 A0BB BDAE EBB4 94A6 7A09 8ED5 7743
uid Admin <admin.bananasplit.info>
| |
| Echeloff 2006-10-23, 1:12 pm |
| Hi Steve,
many thanks for your response.
Zax <admin@bananasplit.info> wrote:
>On 22 Oct 2006 13:12:28 -0000, Echeloff wrote in
>Message-Id: <KLPW95RZ39012.3419907407@anonymous.poster>:
>
>
>Hi there, sorry for not responding sooner, I got distracted by other
>things yesterday, such as puppies eating my kitchen furniture! :-)
A non-virtual damage, oh dear! <g>
>It would be nice to have this redundancy but it would take a major
>change to the current design in order to support multiple exit points.
>Exit nodes would have to communicate with each other to share info on
>who has delivered what messages.
Right. This isn't simply done by installing two servers at different
locations and letting them share their keys. After receiving copies of
a message, they would have to engage some locking mechanism to prevent
from double posting. With Usenet postings it might help to preserve
the Message-ID coming from the originator, but that's no solution for
the delivery of mails.
>
>
>I don't think the problem is really that bad. At the moment you have a
>number of exit-nodes that easily achieve 99% availability. Many
>business systems fail to achieve such levels. It's reaction time to
>failures that causes the biggest problem.
ACK, which may be infinite with mails, as they lack any feedback
channel. In this respect nym accounts behave more predictable.
>
>
>Your example isn't really a remailer reliability issue. It's a problem
>with how remailers interact with external systems like mail2news
>gateways. Last year I did some analysis of messages hitting the banana
>mail2news gateway. About 20% of messages arriving from remailers were
>missing a Newsgroups header.
Good heavens! That's an extremely high number. Let's hope, with an
effective automation in place, those avoidable errors can be
eradicated.
> To the sender, this appears as remailer
>failure when in reality the remailers have done the job fine. I'm
>guessing messages that make it to the entry node in a chain arrive
>at the exit node with very high reliability. The majority of problems
>are external to the remailer process.
>
>
>True, but the focal point of all that redundancy is still the exit
>remailer. Doesn't matter how many copies are used or how many recipient
>mail2news are specified, if that node fails, messages are lost.
>
>Take the recent closure of bigapple as an example. It's an exit
>remailer that's been running with high availability for a long time.
>Then suddenly it stops working. Stats (that run on auto-pilot) take
>hours, if not days to drop uptime below most users random selection
>threshold. No multiple copies or multiple mail2news recipients solve
>this.
Yep, such an incident makes you think.
>
>Agreed. Stats can never react instantly to remailer failure, but they
>could react much quicker than they do now.
>
>
>Pings are currently just like any other remailer message, they are sent
>using a mixmaster client, through a single node chain (the remailer
>being pinged) with a recipient of the pinger. In this manner they are
>subjected to the same rules as any other message and spend long periods
>of time sat in a Mixmaster pool. Take Frell as an example; it will take
>on average 5 hours to get a single ping response from it. If pings
>could bypass the pool, the pinger would receive a near-instant response.
Now I got it, a short circuit for specifically marked ping messages.
>
>The code for pinging is scheduled for 0.0.8 I believe. This will be a
>big improvement as ping functionality is built into the remailer.
Right. It won't cure the exit uncertainty, but problems are indicated
much earlier.
>You might want to check out http://blog.bananasplit.info/?p=116 for some
>ideas I had on improving mixmaster pings.
I did, thanks. Interesting stuff. So please allow some questions from
a stats novice, asked without much reflection:
What data are currently expressed in your stats? Is it only about ping
messages, or is there some early warning system as well, which
analyzes the response to e.g. remailer-help mails, and may indicate a
complete breakdown almost immediately?
Do the pinger messages contain unique markers to allow an assignment
to the time of their transmission, and, if so, are there several
periodically sent messages simultaneously on their way through
remailers with a high latency like frell or bunker? And finally, what
time interval between those messages is it about?
Kind regards
Echeloff
| |
| Cyberiade.it Anonymous Remailer 2006-10-25, 7:12 am |
| >I've donated to causes in the past (usually one time in amounts
>of $100.00 to $200.00), but if I were to want to donate to the
>"remailer network" I wouldn't even know where to send the money.
>
>Perhaps it's because I'm lazy and haven't tracked down any particular
>address, but I suspect there is no centralized address.
>
>Those operating remailers with good reliability deserve recompense.
>Donations could be requested "officially" and a single address
>or account could be specified for anonymous donations.
How about the EFF. Have they ever taken care of remailers? In
particular I think of the Thrasher case. Were there any contacts?
..
| |
| Twisty Creek Admin 2006-10-26, 7:12 pm |
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 22 Oct 2006 20:06:16 GMT, CL@this.group wrote:
>
>adejoode+usenet@sabotage.org (Alex de Joode) wrote in news:ehg1km$s0r$1
>@heroin.sabotage.org:
>
>
>Actually, it might be.
>
>I've donated to causes in the past (usually one time in amounts
>of $100.00 to $200.00), but if I were to want to donate to the
>"remailer network" I wouldn't even know where to send the money.
>
>Perhaps it's because I'm lazy and haven't tracked down any particular
>address, but I suspect there is no centralized address.
>
>Those operating remailers with good reliability deserve recompense.
>Donations could be requested "officially" and a single address
>or account could be specified for anonymous donations.
>
>It could work, perhaps, but it sure would complicate matters. How would
>you distribute the money, and to whom? Who would be in charge of the
>money? What rules would determine who was ineligible even if they
>operate a reliable remailer?
>
>CL
Speaking only for myself, I operate a re-mailer to repay those that did it
before me and those who's services i used. Payback so to speak. I wouldn't
accept a penny. Every once in a while I get a thanks. That's enough for me.
Twisty Creek Admin
-----BEGIN PGP SIGNATURE-----
iQA/ AwUBRUEr9f9qPDhkRaWyEQIfrQCeMOAd9n4mUETc
9oeWs/qTHi2xvgsAn3GA
lqJT68ypch+uhqHb660li8fQ
=8zoW
-----END PGP SIGNATURE-----
| |
|
| Twisty Creek Admin <admin^@^twistycreek^dot^com> wrote in
news:mma2k2hk54amplc1kt2hjlgpvo0fmij3jn@
4ax.com:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 22 Oct 2006 20:06:16 GMT, CL@this.group wrote:
>
>
> Speaking only for myself, I operate a re-mailer to repay those that
> did it before me and those who's services i used. Payback so to speak.
> I wouldn't accept a penny. Every once in a while I get a thanks.
> That's enough for me.
>
> Twisty Creek Admin
>
>
> -----BEGIN PGP SIGNATURE-----
>
> iQA/ AwUBRUEr9f9qPDhkRaWyEQIfrQCeMOAd9n4mUETc
9oeWs/qTHi2xvgsAn3GA
> lqJT68ypch+uhqHb660li8fQ
> =8zoW
> -----END PGP SIGNATURE-----
>
I respect that.
If, however, someone without sufficient assets loses a machine that
is a remailer, replacing it may not be an option - and closing
might be the result.
Having a place to donate would allow some of the users of remailers
a way to "payback" those who work hard to provide these services.
It would be best if it could potentially benefit the entire network
rather than one in particular. With my luck, I would have donated
money to "Widow" remailer or "Gretchen." And they would then
immediately disappear.
Operating a remailer? I tried that. And I am not up to the task.
Oh, and Thanks!
CL
|
|
|
|
|