FreeBSD 8.0-STABLE & mpd -> system freeze

[Roar Pettersen]

Hi !



Upgraded some servers from 7.2-stabel to 8.0-stable early april and since 
then I have seen stability problems with 8.0 servers which use mpd (vpn).
I have tried several mpd version (5.5, 5.3 and 5.1), but the system freeze
within 6 hours or 3-5 days. Early in april we got typical watchdog timeout
error message just before the system freeze, but now we don't get any 
error message.

Sometimes we also see that the mpd process goes into a none killeable 
stauts, and then when I execute a "shutdown -r" the system hang with
this message :

stopping mpd5
Waiting for PIDS : 114830 second watchdog timeout expired. Shutdown terminated.
Apr 29 21:04:52 init : some process would not die; ps axl advised
Waiting (max 60 seconds) for system process 'vnlru' to stop...

We see this problem on both i386 and amd6 systems.

FreeBSD vpn-gw4.uib.no 8.0-STABLE FreeBSD 8.0-STABLE #0: Sun May  2 
01:04:22 CEST 2010     root at vpn-gw.uib.no:/usr/obj/usr/src/sys/VPN-GW  amd64

# Netgraph
options         NETGRAPH        #netgraph(4) system
options         NETGRAPH_BPF
options         NETGRAPH_IFACE
options         NETGRAPH_KSOCKET
options         NETGRAPH_L2TP
options         NETGRAPH_MPPC_ENCRYPTION
options         NETGRAPH_PPP
options         NETGRAPH_PPTPGRE
options         NETGRAPH_SOCKET
options         NETGRAPH_VJC
options         NETGRAPH_NETFLOW

#
kern.ipc.shm_use_phys=1
kern.ipc.maxsockbuf=4194304
kern.ipc.somaxconn=4096
#
net.inet.tcp.sendspace=131072
net.inet.tcp.recvspace=262144
net.inet.tcp.syncookies=0
net.inet.tcp.inflight.enable=0
net.inet.tcp.reass.maxqlen=128
#
net.inet.ip.intr_queue_maxlen=512
net.inet.ip.redirect=0
net.inet.ip.process_options=0
#
net.inet.udp.recvspace=131072
#
net.inet.icmp.icmplim=2000
net.inet.icmp.drop_redirect=1
#
net.local.stream.recvspace=131072
net.local.stream.sendspace=131072
#
net.graph.recvspace=262144
net.graph.maxdgram=262144


Systems with low network traffic doesn't have this problem, only observed 
on systems with ca +50 mpd tunnels or more. We also have other 8.0 servers 
with high network load, but they doen't use Netgraph.

I don't have any core files, no crash files, etc so I can't give any more 
info.

Downgrade to FreeBSD 7.3 solves the problem, but I would like to stay on 
version 8.


--

   Roar Pettersen
   Universitetet i Bergen -  The University of Bergen