4K keys increasing size of messages?

Anonymous

unread,

Dec 30, 2014, 2:08:21 PM12/30/14

to

I just noticed that messages encrypted using a 4K key appear to
be larger that those using the 1K keys. In some cases double to
three times. Will this cause the messages to fail if they
encounter a remailer message size limit?

Bergman Admin

unread,

Dec 31, 2014, 7:35:30 AM12/31/14

to

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Mixmaster remailers make every outgoing message the same size (approx 28K).
If the original message is smaller than 28k it will be padded. If it is
larger than 28K it will be broken into multiple 28K pieces.

And so you need not worry about Mixmaster messages being rejected because they
are too large.

PGP-encrypted messages are a different story. They are not padded to 28k
like mix messages are.

- --

Bergman Remailer Admin 0x770251F2
Fingerprint: 8124 C17B 7FD9 1738 066F 4E29 4407 A967 7702 51F2

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)

iQEcBAEBCAAGBQJUo+2QAAoJEEQHqWd3AlHydigH/j1xoz4MSRAhUQKGDKJn8QWU
z/6ZvmjIDcl3ptB3UtRDChYJVitrxcBwtSHOyL2mvSwqpIr7B5pSA0e3bZFABJGZ
sItIxBqfPOs++kho8H7iW+BxndxZ3eBYS9osYLrvKbsqp0jf1PAASnVM0xq92HxI
iIJpJ+7PTHPoJMLGSjr5kajNZ9u1Dfy7reniHNVxtV9g7847RMZk6nlCTk8uTWBQ
rfwXCfJqaSsa6O3T/piauP0zERLef3K6zPRGTepbDf6uEyqKTir79fipiXqwtmH+
wvmkFjhZlK5p+IS3efjdGCQ4IzxvnKvPJvMHG8yNVOvjKcUzdJGShfbDDs50+qE=
=mFPN
-----END PGP SIGNATURE-----

Anonymous Remailer (austria)

unread,

Dec 31, 2014, 9:54:32 AM12/31/14

to

> I just noticed that messages encrypted using a 4K key appear to be larger
> that those using the 1K keys. In some cases double to three times. Will
> this cause the messages to fail if they encounter a remailer message size
> limit?

> Mixmaster remailers make every outgoing message the same size (approx 28K).
> If the original message is smaller than 28k it will be padded. If it is
> larger than 28K it will be broken into multiple 28K pieces.

> And so you need not worry about Mixmaster messages being rejected because they
> are too large.

> PGP-encrypted messages are a different story. They are not padded to 28k
> like mix messages are.

I think he is talking about klen.

Bergman Admin

unread,

Dec 31, 2014, 12:02:11 PM12/31/14

to

Yes, I assumed he was talking about klen.

So, for example Paranoia Remailer has a klen of 150k.
It would therefore accept a mix message of length 28k.
And it would silently drop a cpunk message of 151k.

Anonymous

unread,

Dec 31, 2014, 12:12:18 PM12/31/14

to

On Wed, 31 Dec 2014 12:02:05 -0500, Bergman Admin <nob...@bergman.delphy.org> wrote:

>Yes, I assumed he was talking about klen.
>
>So, for example Paranoia Remailer has a klen of 150k.
>It would therefore accept a mix message of length 28k.
>And it would silently drop a cpunk message of 151k.

My understanding is that a klen of 150k would also drop a multi-part mix message to the final recipient that exceeded 150k.

Bergman Admin

unread,

Dec 31, 2014, 12:43:09 PM12/31/14

to

You are likely correct, Stray Cat. And I'm mistaken. I'm going to send a few
multi-part test messages to confirm.

Message has been deleted

Anonymous Remailer (austria)

unread,

Dec 31, 2014, 2:53:19 PM12/31/14

to

> Yes, I assumed he was talking about klen.

> So, for example Paranoia Remailer has a klen of 150k.
> It would therefore accept a mix message of length 28k.
> And it would silently drop a cpunk message of 151k.

> My understanding is that a klen of 150k would also drop a multi-part mix message to the final recipient that exceeded 150k.

Sounds like Bergman is about to eat crow again.

John Kennerson

unread,

Dec 31, 2014, 4:30:43 PM12/31/14

to

In article
<dcb8aaltqiquql0h4...@news.stray.cat.edu>

Yes that is what I was asking about. Thank you all for
responding. I think I understand this better now.