On 22 Dec 1998, Niels [ISO-8859-1] Möller wrote:
> Do you think that is good enough? If we really want to, we could
> always let lshd dlopen some pluggable authentication modules of its
> own ;-)
>
> The runtime detection is basically
>
> #ifdef HAVE_GETSPNAM
> if (passwd->pw_passwd && !strcmp(passwd->pw_passwd, "x"))
>
> (I don't know much about the shadow mechanism myself. Keresztfalvi
> Gabor Agoston helped me with the code above, i.e. he should have the
> credits if it works, and I should be blamed if it doesn't).
It works for me on my Linux box, but I realized, that for example on AIX this
wouldn't work, because in /etc/passwd there is '!' (and not 'x') instead of
the password... This needs some workaround. Possibly other systems use other
chars instead of the password...
As you see this runtime detection works only if there was shadow support
in your development environment, when you compiled the progam (and you run the
lshd as UID=0 ;)
BTW 1.: On some systems there is not any crypt.h. It should be checked for in
configure.
BTW 2.: I still can't get lsh to work because of that EPIPE. I tried it on
SuSE 5.3 Linux with kernel 2.0.35(SuSE patched) and 2.1.130, with libc5.4.46
on two different hardware (but I think this is not important (P133 and
AMDK5PR100, both with HX motherboard and 64MB RAM different networkcards)).
C compiler is: gcc version 2.7.2.1
Please, who has ever compiled and tried lsh, send me a mail with your HW/SW
config, lsh version and results of running, because I'd like to make a
summary, where lsh worked, and where didn't. I hope this will help us to solve
that EPIPE.
Keresztg
P.s.: Please forget Agoston from my name, because it's my second first name :)
and I don't use it. Indeed my name is still long enough without that. :)
Use my nickname: Keresztg
+ Keresztfalvi Gabor
+ Student of Technical University of Budapest
+ mailto: [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED]
+ http://www.piar.hu/~keresztg/ Pubkey: finger [EMAIL PROTECTED]
