Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 14 Jun 2004 00:50:46 +0200
From:      "Willem Jan Withagen" <wjw@withagen.nl>
To:        "Willem Jan Withagen" <wjw@withagen.nl>
Cc:        freebsd-ports@freebsd.org
Subject:   Re: anybody using net-snmp on AMD64 or other 64bit platforms??
Message-ID:  <033301c45198$ddd10b40$471b3dd4@dual>
References:  <023d01c45161$8f486360$471b3dd4@dual><40CCA3CB.6090808@anduin.net> <02b001c4517d$46da25c0$471b3dd4@dual>

next in thread | previous in thread | raw e-mail | index | archive | help
> > Willem Jan Withagen wrote:
> > > Reason I ask, is that it crashes with me at all kinds of places...
> > >
> > > I've started looking at the warnings from a regular make.
> > > And in the process of tryin to fix those there are a lot of places where
> > > typeconversions are really messed up.
> > >
> > > int, size_t, I32, function addresses and pointers are all intermixed.
> > > And this of course will create all kinds unwanted effects....
> > >
> > > Are there any suggestions as to enhance portability this way...
> > >
> > > eg. what kind of generic pointer type would one use to make shure
> > > that pointers are the correct size on both architectures....
> > >
> > > Or as a side questions:
> > >     Are there other snmp-daemons that run on 64bit platforms??

I've got it more or less running, and snmpd is no longer crashing on me "yet".
But I'm now getting wrong answers for the tcp/udp connections. But for
established connections as well as open server ports....

Both netstat and snmpd do this by retreiving data with sysctl from the kernel.
And where netstat has got its things right, snmpd seems to be really off.
The most likely reason is that snmpd defines some structures by itself, but now
the question is which part is that???

Does anybody know of tools to compare or analyse the raw data coming out of a
struct tcpcb.....

--WjW




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?033301c45198$ddd10b40$471b3dd4>