[Nut-upsdev] minor building issues (svn HEAD revision 427)

[Charles Lepple]

On 5/19/06, Mike Patterson <mpatters at cs.uwaterloo.ca> wrote:
> ld: fatal: symbol `setenv' is multiply-defined:
>         (file ../common/setenv.o type=FUNC; file
> /software/netsnmp-5.3/lib/libnetsnmp.a(system.o) type=FUNC);
> ld: fatal: File processing errors. No output written to snmp-ups

Can you check and see if your version of netsnmp documents their setenv call?

> I can fix this by changing the setenv() definition in common/setenv.c to
> nut_setenv, and changing the calls to setenv() in client/upsmon.c
> appropriately.  I'm wondering if this is a known issue, or if perhaps
> it's a problem related to our environment (which is weird and wonderful
> and largely unknown to me)?  Should I submit a patch?

Probably just that people aren't testing with both Solaris (where
setenv is evidently not part of the C library) and netsnmp.

Depending on whether the netsnmp setenv is the same or not, we might
be able to just make a quick adjustment to the configure script. If it
is the same, then we just won't use common/setenv.o when netsnmp is
present.

-- 
- Charles Lepple