Call for testers.

[Damien Miller]

On Sat, 15 Sep 2001, Wayne Davison wrote:

> Back in revision 1.133 of session.c, the conditional prototype for
> do_pre_login() was removed.  This either needs to go back into the file,
> or the do_pre_login() function needs to be moved up in session.c above
> do_exec_pty().  Here's a patch for the former option:

It is static in the lastest CVS.

> And, in revision 1.6 of sshlogin.h, the prototype for record_utmp_only()
> was removed.  This either needs to be restored, or an extern prototype
> for the function added to session.c inside the do_pre_login() function.
> Here's a patch for the former choice:

Do we even need this function? It appears identical to record_login()
except it takes one less argumentm which isn't used anyway.

> Also, the bugs I reported back on June 6 in the LOGIN_NEEDS_UTMPX
> handling are still there.  The following patch combines the two patches
> I sent back then (and resent together recently).  This only affects
> UseLogin=yes on LOGIN_NEEDS_UTMPX systems (currently only Solaris, but I
> think that the configure section for Unicos also needs to set this
> define -- Wendy?).  This patch doesn't affect the normal (non-
> LOGIN_NEEDS_UTMPX) processing at all, so it's quite safe.
> 
> Index: loginrec.c
> --- loginrec.c	2001/08/06 23:29:17	1.34
> +++ loginrec.c	2001/08/21 17:53:43
> @@ -448,6 +448,8 @@
>  login_utmp_only(struct logininfo *li)
>  {
>  	li->type = LTYPE_LOGIN;
> +	/* set the timestamp */
> +	login_set_current_time(li);
>  # ifdef USE_UTMP
>  	utmp_write_entry(li);
>  # endif
> @@ -456,9 +458,10 @@
>  # endif
>  # ifdef USE_UTMPX
>  	utmpx_write_entry(li);
> -# endif
> -# ifdef USE_WTMPX
> +# else
> +#  ifdef USE_WTMPX
>  	wtmpx_write_entry(li);
> +#  endif

This doesn't look right - this appears to make the wtmpx_write_entry()
call only if we are not using utmpx. Could you explain this a little 
more?

-d

-- 
| Damien Miller <djm at mindrot.org> \ ``E-mail attachments are the poor man's 
| http://www.mindrot.org          /   distributed filesystem'' - Dan Geer