I didn't test it and it has been quite some time, but I bonded two NICs for failover to the same switch. If one interface was down it put something in /var/log/messages. I monitored that entry.
Not saying this is a good, tested or even working approach, but that is what I did years ago.
Josh Luthman Office: 937-552-2340 Direct: 937-552-2343 1100 Wayne St Suite 1337 Troy, OH 45373
On Thu, Jan 5, 2012 at 4:32 PM, Henrik Størner <henrik at hswn.dk> wrote:
On 05-01-2012 16:45, Neil Simmonds wrote:
A large number of our systems have dual port network cards and I’m wondering if there is an easy way to monitor if a particular interface is down.
Each adapter has the following information in the client data,
General Statistics:
No mbuf Errors: 0 Adapter Reset Count: 1 Adapter Data Rate: 2000 Driver Flags:Up Broadcast Running
[snip]
I’m wondering if I could check the value highlighted and alert if it is not “up” without having to write an external script to do it?
Not out of the box, you'll have to do some scripting yourself.
Since it's a large number of systems and you already have the data available in the client data, it seems an obvious candidate for a server-side module.
Take a look at the "xymond_rootlogin.pl script in the Xymon sources (you'll find it in the xymond directory). Adapting this to fit your needs should be a rather simple job for someone who is familiar with Perl programming (sad to say, I'm not).
Regards, Henrik
Xymon mailing list Xymon at xymon.com http://lists.xymon.com/mailman/listinfo/xymon