I've seen something similar with my setup.
eth0: 3c509? private network 192.168.x.x
eth1: 3c509-TPO Connected to cable modem (Road Runner)
No routing, forwarding, or address translation between networks.
May be related: I noticed that about the time I start having these problems, I
seem to get fifo errors on the interface. Looking at /proc/net/dev (reformated
so it doesn't have to wrap in the mail):
Inter-| Receive |
face |bytes packets errs drop fifo frame compressed multicast|
lo:82511981 421410 0 0 0 0 0 0
eth0:14333318 109942 0 0 0 0 0 0
eth1:526333628 534627 101 0 103 101 0 0
Inter-|Transmit
face |bytes packets errs drop fifo colls carrier compressed
lo:82511981 421410 0 0 0 0 0 0
eth0:12342428 88136 0 0 0 3 0 0
eth1:27267828 300709 0 0 0 2402 0 0
I also noticed that the ping times (for me) seem to be a multiple of 1000 + a
small number of milliseconds.
Looking through dev_do_watchdog (in linux/net/sched/sch_generic.c) I noticed
that the watchdog expires in 5*HZ. If for some reason either the card stopped
interrupting, or the driver got confused, could you have something like this? :
software sends ping
500 ms later watchdog sends packet
500 ms later watchdog receives reply
ping responds with 1000+ time.
If this is the case, does anyone know if there is a way for the busy flags to
get out of sync, or a way to reset the card without downing & uping the
interface to debug this? Or any other ideas (or other possible problems?)
Thanks for any help you can give me.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/