headers
Ken Hornstein | 17 Mar 2004 20:22
X-Face
Picon
Picon
Favicon

Re: RC3

(Sigh, work travel sucks).

Alright, so I did some more research on the SASL issues.  In short, we
have the following problems:

- CPPFLAGS isn't actually used in any of the compilation lines.  Okay, that's
  unfortunate, but one can always use CFLAGS in a pinch.

- Unfortunately the location of the SASL2 sasl.h include file is not exactly
  standardized, it seems.  Some systems (like RedHat) put it in 
  include/sasl2; other systems (like MacOS X) put it in include.  In the
  future we should probably do an autoconf test for it, but I'd rather
  not hold up the release for it.

So, I think I'm going regen the tar file with the right directory name,
and then call it a release.  We can fix these other problems in a future
release (hopefully it won't take too long for the next release).

--Ken

_______________________________________________
Nmh-workers mailing list
Nmh-workers <at> nongnu.org
http://mail.nongnu.org/mailman/listinfo/nmh-workers
Permalink | Reply | 0 comments |
headers
Jeffrey P. Morgenthaler | 12 Mar 2004 23:57
Picon

VFAT support and gdbm woes

Hi,

I am a long time user of MH (since 1986) and mh-e (since 1992) + am
working on a dual-boot toy (... I mean computer) at home that has a
completely uniform interface no matter what side you are on.  Cygwin is
essential to this + I thank Earl Hood for porting nmh to that side of
things.  Now I am in the embarrassing situation of having MH work better
in Windows on the shared vfat partition than LINUX side of things
(Debian testing distribution).  It seems as though MH is very
sophisticated when it moves files around + uses hard links (essentially
mucking directly with the filesystem).  Earl (or the cygwin primitives
he relied on) must have found a way to turn this off.  I was wondering
if there was any plan to implement this in the UNIX version if things.

I would be happy to play with the code myself, but I am having trouble
getting it to compile (nmh-1.1-RC3.tar.gz) because of the gdbm problems.
I used Debian's dselect to download the gdbm developement package
(header files + libraries) + they get put directly in /usr/include as
gdbm-ndbm.h and gdbm.h.  I can get a little further on the compile if I
make a directory and (soft) link ndbm.h -> /usr/include/gdbm-ndbm.h but
it still fails since the version of libgdbm.a Debian provides has all
the function names prepended with g (e.g. gdbm_open).  Fixing this is
more than I have time for right now, so I thought I'd bring it to your
attention + see what the plans might be for VFAT support.

Thanks!

jpm

------------------------
(Continue reading)

Permalink | Reply | 1 comment |
headers
Tracy Di Marco White | 17 Mar 2004 11:54
Favicon

Digital Unix (4.0f) doesn't have snprintf


and configure is able to tell that it doesn't, yet things still seem
to require it.  This is with nmh-1.1-RC3.tar.gz I just downlaoded.  I
tend to use the built in cc, but I have the same results with gcc 3.2.

in configure:
checking for snprintf... no
yet while building:
....stuff....
cc: Info: ../h/prototypes.h, line 135: In this statement, the declaration for intrinsic function
"snprintf" referenced at line number 258 in file aliasbr.c, parameter number 2 is of type "unsigned
long", which is not consistent with the expected type "long".  It will be treated as an ordinary external
function. (intrinsicdecl)
int snprintf (char *, size_t, const char *, ...);
----------------------^
cc -s -o ali ali.o aliasbr.o ../config/version.o ../config/config.o ../mts/libmts.a ../sbr/libmh.a    -lm
ld:
Unresolved:
snprintf
vsnprintf
*** Exit 1
Stop.
*** Exit 1
Stop.

Am I missing something obvious?

Tracy J. Di Marco White
Academic Information Technologies
Iowa State University
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jay Nietling | 10 Mar 2004 20:42

nmh-1.1, cyrus-sasl-2.1.17, cyrus-imapd-2.2.3, mit krb-1.3.2


does anyone have this combo working with gssapi authentication?

inc -sasl works with tiny messages.  larger ones fail:

inc: Error during SASL decoding: SASL(-1): generic failure: Unable to
find a callback: 32775

-jay

_______________________________________________
Nmh-workers mailing list
Nmh-workers <at> nongnu.org
http://mail.nongnu.org/mailman/listinfo/nmh-workers
Permalink | Reply | 0 comments |
headers
Ken Hornstein | 17 Mar 2004 23:10
X-Face
Picon
Picon
Favicon

Re: nmh-1.1, cyrus-sasl-2.1.17, cyrus-imapd-2.2.3, mit krb-1.3.2

(Note: if you want your messages to actually make it to the list without
moderation, you should subscribe).

>does anyone have this combo working with gssapi authentication?

That exact combo?  No.  I'm using it fine with a SASL-ified qpopper
(which I think is using cyrus-sasl 2.1.17) with GSSAPI.  But I tested
it with 20-30MB messages without a problem.

>inc -sasl works with tiny messages.  larger ones fail:
>
>inc: Error during SASL decoding: SASL(-1): generic failure: Unable to
>find a callback: 32775

Hm.  I looked over the library a bit.  I suspect that the real error is
being hidden, because the "Unable to find a callback" really shouldn't
apply here.

Unfortunately, I don't have a good solution here.  I don't suppose you
could find the exact message size at which it breaks?  I think the only
reasonable solution is to perhaps run it under a debugger and find the
root cause of the problem (Unfortunately, we don't use Cyrus, so I'm
not sure what's going wrong).

--Ken

_______________________________________________
Nmh-workers mailing list
Nmh-workers <at> nongnu.org
http://mail.nongnu.org/mailman/listinfo/nmh-workers
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bill Wohler | 18 Mar 2004 03:57
Picon
Picon
Gravatar

Re: VFAT support and gdbm woes

Jeffrey P. Morgenthaler <jpmorgen <at> midnight.gsfc.nasa.gov> wrote:

>                                                               Cygwin is
> essential to this + I thank Earl Hood for porting nmh to that side of
> things.

Sorry, Jeff, I don't have any information about turning of the linking
(other than not using refile -link), but if you, or someone else who
appears to be using nmh on Windows can write a short paragraph for the
FAQ on how one gets nmh working on Windows, I would be much obliged.

--

-- 
Bill Wohler <wohler <at> newt.com>  http://www.newt.com/wohler/  GnuPG ID:610BD9AD
Maintainer of comp.mail.mh FAQ and MH-E. Vote Libertarian!
If you're passed on the right, you're in the wrong lane.

_______________________________________________
Nmh-workers mailing list
Nmh-workers <at> nongnu.org
http://mail.nongnu.org/mailman/listinfo/nmh-workers
Permalink | Reply | 0 comments |

Gmane