There are probably good reasons why what I'm suggesting here is,
as stated, a bad idea; but since the packet storage format is up
for discussion I thought I'd throw this out to see if it peaks
anyones interest.
What about adding the concept of arbitrary meta-packets that can
sit anywhere in the c
-BEGIN PGP SIGNED MESSAGE-
{My appologies for the Reply-to: nonsense. Trying to fix it}
> "Guy" == Guy Harris <[EMAIL PROTECTED]> writes:
>> Itojun changed print-esp.c to look up crypto routines using
>> EVP_get_byname() instead of having a table.
>>
>> The problem w
-BEGIN PGP SIGNED MESSAGE-
Loris had previously written up an ID on a proposed pcap format.
It is similar, but not identical to what I had proposed.
It is in xml2rfc (rfc2629) format.
I can't say if the IETF would or should ever consider publishing it.
I think it is likely out of scope f
On Sat, Apr 03, 2004 at 05:44:55PM -0500, Michael Richardson wrote:
> It appears that we don't really do the right with:
>
>./configure --with-crypto=/path
>./configure --with-openssl=/path
>./configure --with-ssleay=/path
>
> (I'm uncertain which of these is right)
"--with-openssl"
On Sun, Apr 04, 2004 at 11:53:18PM -0400, Michael Richardson wrote:
>
> Itojun changed print-esp.c to look up crypto routines using
> EVP_get_byname() instead of having a table.
>
> The problem with that is that the ivlen differs for different
> algorithms. This is easily solved by calling EVP_CI
CVS log entries from 04.04.2004 (Sun) 09:05:46 - 05.04.2004 (Mon) 09:05:57 GMT
=
Summary by authors
=
Author: mcr
File: tcpdump/tests/esp5.gdbinit; Revisions: 1.1
File: tcpdump/te
