Re: [cks-devl] Re: Sync thoughts (fwd)

[M. Drew Streib]

On Tue, Sep 04, 2001 at 02:21:55PM -0400, V Alex Brennen wrote:
> I'm much more interested in cleaning up all the segfaults
> in the code and getting the code to a point where it can
> be a stable replacement for pksd, than the advanced syncing
> designed right now. I think the subkey support issue and
> scalability issues with pksd are critical.  I need to get
> cks_import and cks_backup written and get support for
> keyrings.

I agree that these are critical issues.

I must also make the statement that without the ability to fit in
with the sync scheme for pgp.net I won't be able to use this as a full
replacement. :/

That _doesn't_ mean that I won't be running it in parallel for testing
purposes. I just won't be able to put it in 'production' on pgp.net
until the syncs work.

> Nevertheless, I want to mention that I talked with Werner
> Koch and he suggested that there's was support for a binary
> replacement for hkp - at the very least in gnupg. Syncing
> mechanisms could be built in to that protocol.

A new sync protocol that scales well might be nice, although the old
one isn't so bad that it is hindering anything now, nor does it appear
that it won't scale to keyrings/activity at least 10x current size/traffic.

On a pgp.net server now syncing with about 8-10 major keyservers
worldwide, I get about 6000-10000 emails/day from the pksd protocol, with
about 400-800 actual updated keys daily. This doesn't even begin to
generate any kind of load. I estimate that this is probably among
the busiest syncing keyservers around...

> For night now, I'm going to get the point to multipoint
> code working (the current stuff)...  Then I agree the
> mutlipoint to multipoint is the way to go - like the
> Usenet and pksd records of servers transversed.  I
> think we can most easily do that with comment lines
> in the ASCII encoded keys.

As far as pksd goes, following the current convention is necessary, which
is to add RFC822 header lines to the email.

I know that making something totally new is very tempting, but I'd suggest
that you'll get a _lot_ more testing and general use if you _first_ work
to support the current sync system, and then branch out to more advanced
methods.

-drew

-- 
M. Drew Streib <address@hidden> | http://dtype.org/
FSG <address@hidden>    | Linux International <address@hidden>
freedb <address@hidden>        | SourceForge <address@hidden>

pgp5juReyAW2I.pgp
Description: PGP signature