Hi,
	I've been getting "Network timeout on input" lately
	everytime I try to sup allsrc from sup.netbsd.org, after
	that sup quits with Transfer aborted.

	It seems that we have problems with our networks in
	this part of germany and are thus being routed over
	dfn (a datex-p network) which is causing ping-losses
	between 25 and 75%. A a certain point I can't ping
	a machine with more than one packet before "ping"
	stops. If I "kill" it it doesn't go and I have to
	"kill -9" the process.

	Is a buffer overflowing or is there anything I can
	do other than hope that our network gets repaired soon.

	I'm running a NetBSD compiled with 15.Feb's sources.
	Is more data needed, which data is needed, does it make sense.

Cheerio
PS (The bad network connection has been verified from other
    machines, but the ping-hanging is only happening on NetBSD,
    it could also be trouble with another card though. What happens
    when sb0 uses the same irq as lpt0 but lpt0 is not used/nor compiled
    instead lpa0 is compiled into the kernel. (irq 5 was not viable
    since it didn't recognise, but sb0 irq 7 seems to work fine.
    I'll try removing the card tommorow but the Network timeout on input
    existed before adding the card too.)
Cheerio^2
PS2:
I should add that the timeouts only happen with
"allsrc", both "doc" and "othersrc" (one before and the other after)
sup fine.

Also the ping-problem doesn't happen as often as the Network 
timeout on input (which happens 6 out of 7 tries.)
Cheerio^3

-- 
The cost of living hasn't affected it's popularity. (unknown)
current	release=doc host=wipux2.wifo.uni-mannheim.de \ "NetBSD-current mirror"
hostbase=/mit/ftp/pub/NetBSD base=/usr prefix=/usr backup delete use-rel-suffix