[syslog-ng] [patch] Fixes for 2.0.7 on not ipv6 aware systems

[Balazs Scheidler]

On Tue, 2008-01-08 at 10:33 +0100, Corinna Vinschen wrote:
> > On Mon, 2008-01-07 at 21:32 +0100, Corinna Vinschen wrote:
> > > - Two functions in src/afsocket.c use AF_INET6 even if ENABLED_IPV6
> > >   is not defined.
> > > 
> > > - The new loggen tool has a typo in usage(), printing -i instead of
> > >   -I for the interval arg.
> > > 
> > > - loggen has a bug in the option handling.  Due to a missing colon
> > >   in the option string, it crashes when using the -s option.
> > 
> > I have commited these patches as they are trivial changes, thus do not
> > need a CLA.
> 
> Thanks.  On second thought, I missed another trivial patch to loggen.
> The -s option takes a parameter, but the usage string does not reflect
> that fact:
> 
> 	* tests/loggen/loggen.c (usage): Clarify -s usage.
> 
> --- loggen.c.X	2008-01-08 10:03:03.477272000 +0100
> +++ loggen.c	2008-01-08 10:03:11.752376000 +0100
> @@ -146,7 +146,7 @@ usage()
>           "  --unix, or -x           Use UNIX domain socket transport\n"
>           "  --stream, or -S         Use stream socket (TCP and unix-stream)\n"
>           "  --dgram, or -D          Use datagram socket (UDP and unix-dgram)\n"
> -         "  --size, or -s           Specify the size of the syslog message\n"
> +         "  --size, or -s <size>    Specify the size of the syslog message\n"
>           "  --interval, or -I <sec> Number of seconds to run the test for\n");
>    exit(0);
>  }

Thanks, I've fixed this together with the getaddrinfo() changes.

> 
> 
> > > - loggen uses getaddrinfo/freeaddrinfo even on systems on which
> > >   HAVE_GETADDRINFO is undefined.  I created replacement code using
> > >   gethostbyname/gertservbyname on systems lacking getaddrinfo.
> > 
> > I did not commit this one as this would require a signed CLA. Either I
> > fix this on my own or you sign a Contributory License Agreement. Do you
> > think the latter is feasible for you?
> > 
> > http://www.balabit.com/network-security/syslog-ng/opensource-logging-system/contribution/
> 
> Ok with me, but I have to ask my employer first.  Given the simplicity
> of the code, it's probably faster if you just create your own version :)

I've just did. The results should be available in our git repository or
in the nightly snapshot.

Can I ask you to retest whether it compiles fine on cygwin?

Thanks for the patches.

> 
> > > Another problem I have is this:  Given that loggen is in the tests
> > > subdir, I'm wondering if it's correct that loggen is installed to
> > > ${bindir} when running `make install'.  Is that really supposed to
> > > happen or should `make install' skip loggen?
> > 
> > It was meant to be installed, though putting it to "tests" might not be
> > the best idea. "loggen" can be used to check the performance of one's
> > configuration.
> 
> Ok, thanks.  I didn't want to create a syslog-ng package for Cygwin
> with loggen if it's not clear that loggen has to be installed or not.
> However, "loggen" is easily confused with "logger".  Wouldn't something
> along the lines of "syslog-perf" be a better name for the tool?

The similarity was intended, I thought it was funny. So if it is not a
big problem I would leave it as it is.

-- 
Bazsi