Re: Announcement: FreeS/WAN 2.00-pre6 released

On Wed, 5 Feb 2003, Sam Sgro wrote:

> This is mostly a bugfix release. We encourage users to help us debug this 
> pre-release. We've completed our feature-set for 2.00; now it's just a matter 
> of polishing the rough edges.

First:

../../linux/include/freeswan.h:70:1: let op: "DEBUG_NO_STATIC" redefined
../../linux/include/freeswan.h:42:1: let op: this is the location of the previous definition

Second:

cc -o pluto  connections.o constants.o cookie.o crypto.o defs.o foodgroups.o log.o state.o plutomain.o server.o timer.o id.o ipsec_doi.o kernel.o rcv_whack.o demux.o packet.o lex.o keys.o dnskey.o rnd.o spdb.o sha1.o md5.o  ../../lib/libdes/libdes.a ../../lib/libfreeswan/libfreeswan.a -lgmp -lresolv 
crypto.o: In function `init_crypto':
/usr/src/freeswan-2.00-pre6/programs/pluto/crypto.c:50: undefined reference to `mpz_init_set_str'
/usr/src/freeswan-2.00-pre6/programs/pluto/crypto.c:50: undefined reference to `mpz_init_set_str'
/usr/src/freeswan-2.00-pre6/programs/pluto/crypto.c:50: undefined reference to `mpz_init_set_str'

That also looks vaguely familiar.
I didn't have this problem yesterday with freeswan-snap2003feb03h
Both compiles were done against a RedHat source tree (albeit minorly
patched in the abi stuff, but based on 2.4.18-14)

Also, with freeswan-snap2003feb03h and with the pre6 it seems to be the case
as well (by looking at the pluto code) there is still the "Can't 
Opportunistically initiate" message even with debug is disabled. I was
really hoping that this message would be gone in the 2.x series. It
causes massive logfiles on OE production servers. Can we make it a
special plutodebug=oeonly option or something? Or perhaps a
plutodebug=lessthenone?

Paul