em0 watchdog timeouts

[Rudy]

> What about system load, perhaps something is bogging the thing down so that
> it cannot adequately service the network interrupts??
>
Hardly anything is running on the box...
Only things on the box: zebra bgpd (3 peers...)  sshd snmpd


Here is the top of 'top':

load averages:  0.06,  0.08,  0.07       up 7+01:08:16  17:26:39
15 processes:  1 running, 14 sleeping
CPU:  0.0% user,  0.0% nice,  4.5% system,  0.0% interrupt, 95.5% idle
Mem: 193M Active, 42M Inact, 156M Wired, 196K Cache, 83M Buf, 1610M Free


> The specs of the motherboard are respectable, how much memory does it have?
>
> Another thought, are you using the out-of-band management features (like
> IPMI)?
> If you are not then go into the BIOS and disable that stuff.
>
No IPMI card added to that motherboard (you have to add a daughter card).
> Have you run netstat or some other resource monitor to see if you run out of
> anything that might coincide with the watchdogs...
What should I look for?

# netstat -s
4105/4610/8715 mbufs in use (current/cache/total)
4103/2303/6406/25600 mbuf clusters in use (current/cache/total/max)
4103/2297 mbuf+clusters out of packet secondary zone in use (current/cache)
0/44/44/12800 4k (page size) jumbo clusters in use (current/cache/total/max)
0/0/0/6400 9k jumbo clusters in use (current/cache/total/max)
0/0/0/3200 16k jumbo clusters in use (current/cache/total/max)
9232K/5934K/15166K bytes allocated to network (current/cache/total)
0/0/0 requests for mbufs denied (mbufs/clusters/mbuf+clusters)
0/0/0 requests for jumbo clusters denied (4k/9k/16k)
0/6/6656 sfbufs in use (current/peak/max)
0 requests for sfbufs denied
0 requests for sfbufs delayed
0 requests for I/O initiated by sendfile
0 calls to protocol drain routines



Are there specific router-only tunings that may help?

Here are my sysctl settings:

kern.ipc.somaxconn=256
kern.random.sys.harvest.interrupt=0
kern.random.sys.harvest.ethernet=0
kern.ipc.nmbcluster=32768

net.inet.icmp.icmplim=1000
net.inet.ip.fastforwarding=1
net.inet.ip.intr_queue_maxlen=92
net.inet.icmp.drop_redirect=1

dev.em.0.rx_processing_limit=200
dev.em.1.rx_processing_limit=200
dev.em.2.rx_processing_limit=200
#dev.em.4.rx_processing_limit=200
# test setting processing limit up to 300
dev.em.4.rx_processing_limit=300
dev.em.5.rx_processing_limit=200

Thanks,
Rudy