On Fri, Jul 8, 2011 at 1:00 PM, Anton Shterenlikht <[email protected]> wrote: >> I've had some trouble netbooting / jumpstaring recently with a similar >> pattern (using RARP/BOOTP/TFTP/NFS). It turned out to be a dying >> port on the switch whose errors were masked by TCP in day to day >> use, but alas were too frequent for UDP. > > I just have a direct ethernet connection > between my FreeBSD laptop with bootpd/tftpd > servers (one ethernet port only) and a node > which I want to boot (also a single ethernet > port). Is there way for me to check whether > either of these ports are "dying"? > Any further diagnostics I can do?
Just push a lot of data (/dev/zero, /dev/urandom, ...) over this connection, via ssh or something like that, and monitor the error rates in "netstat -in" on both ends. Look at Ierrs, Idrop and Oerrs columns. If one of those ports are dying, or if your cable isn't properly shielded, you'll notice immediately. > Many thanks > Anton -cpghost. -- Cordula's Web. http://www.cordula.ws/ _______________________________________________ [email protected] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[email protected]"
