Re: [exim-dev] exim 4.50 hp-ux 11.23 (ipf) startup problems

Αρχική Σελίδα
Delete this message
Reply to this message
Συντάκτης: Philip Hazel
Ημερομηνία:  
Προς: V. T. Mueller
Υ/ο: exim-dev
Αντικείμενο: Re: [exim-dev] exim 4.50 hp-ux 11.23 (ipf) startup problems
On Mon, 25 Apr 2005, V. T. Mueller wrote:

> The compile run produced some warnings like:
>
>         cc -c -DNOPOSIX -Ae +O3 +Onolimit  -I. pcretest.c
> "printint.c", line 205: warning #2111-D: statement is unreachable
>       break;
>       ^


That is harmless.


>         cc -c -Ae +O3 +Onolimit  -DCOMPILE_UTILITY -o util-spool_in.o
> spool_in.c
> "/usr/include/net/if6.h", line 67: warning #2047-D: incompatible redefinition
> of macro "IFF_PRIORITY" (declared at line 140 of "/usr/include/net/if.h")
>   #define IFF_PRIORITY  0x4000000000000000      /* VLAN IP precendence
> enablement */

>
> [/usr/include/net/if.h defines it this way:
> define IFF_PRIORITY         0x80000000      /* VLAN IP precendence enablement
> */ ]


That is nothing to do with Exim; it's all in your system's header files.
It may well be causing trouble, but I have no idea what it is about.
Exim does not use IFF_PRIORITY anywhere.



> "daemon.c", line 192: warning #2167-D: argument of type "socklen_t *" is
> incompatible with parameter of type "int *"
>        &ifsize) < 0)
>        ^


Sigh. This one has caused trouble before. It may be fixable by putting
an appropriate definition for SOCKLEN_T in the system configuration
file. It's possible this may cause trouble.


> "receive.c", line 188: warning #2068-D: integer conversion resulted in a
> change of sign
>   *inodeptr = (statbuf.F_FILES > 0)? statbuf.F_FAVAIL : -1;
>                                                         ^


Don't know if this matters or not.


> "rfc2047.c", line 262: warning #2167-D: argument of type "const char **" is
> incompatible with parameter of type "char **"
>         (void)iconv(icd, (ICONV_ARG2_TYPE)(&dptr), &dlen, CSS &outptr,
> &outleft);
>                          ^


That should be fixable by an appropriate definition of ICONV_ARG2_TYPE
in the system-dependent file. Clearly this has caused trouble on other
systems, leading to the creation of a system-dependent setting.

> I have no idea if this behaviour is platform-specific or not. Full logs or ssh
> access to a build environment can be provided if needed, though.



I will look at these further when I get some time (I've been away and
there is a lot to catch up). I can't see why any of them should prevent
a daemon start-up, though.

-- 
Philip Hazel            University of Cambridge Computing Service,
ph10@???      Cambridge, England. Phone: +44 1223 334714.
Get the Exim 4 book:    http://www.uit.co.uk/exim-book