belkin WAP resets by belkin USB pods....

Seeing a rather strange amount of behaviour with a Belkin F5D6130 WAP
talking to F5D6050 USB wireless pods - we've had these a while and
they've been working fine. However recently (for no good reason that I
can see) communication has become erratic to the point where even simple
accessing a remote drive over the network will time out with an error.

Since there's a Linux box on the wired side of the network, I've been
having a look at the traffic (ethereal & have also loaded ap-trapd which
logs SNMP traps from the access points). What seems to be occurring is
that even during a simple Explorer folder access, the WAP is
broadcasting a series of SNMP Cold Start traps (which I intepret as it
performing a reset) thus causing the remote connection to loose brief
connectivity, resume then trap again - typical system log looks like:

Jan 9 15:08:31 fridge ap-trapd[1242]: Agent:v1 192.168.0.9
(public@192.168.0.9:162) ColdStart. SysUptime 0:00:00.00
Jan 9 15:09:15 fridge ap-trapd[1242]: Agent:v1 192.168.0.9
(public@192.168.0.9:162) ColdStart. SysUptime 0:00:00.00
Jan 9 15:09:46 fridge ap-trapd[1242]: Agent:v1 192.168.0.9
(public@192.168.0.9:162) ColdStart. SysUptime 0:00:00.00
Jan 9 15:10:20 fridge ap-trapd[1242]: Agent:v1 192.168.0.9
(public@192.168.0.9:162) Disassociation 00:30:BD:631:0A. SysUptime
0:00:34.00
Jan 9 15:10:20 fridge ap-trapd[1242]: Agent:v1 192.168.0.9
(public@192.168.0.9:162) Reassociation 00:30:BD:631:0A. SysUptime
0:00:34.00
Jan 9 15:10:21 fridge ap-trapd[1242]: Agent:v1 192.168.0.9
(public@192.168.0.9:162) ColdStart. SysUptime 0:00:00.00

The packet sniffer also implies that the SNMP trap isn't correct either,
generating the following:

Simple Network Management Protocol
Version: 1
Community: public
PDU type: TRAP-V1
Enterprises: 1.3.6.1.4.1.410.1.1
(SNMPv2-SMI::enterprises.410.1.1)
Agent address: 192.168.0.9 (the WAP)
Trap type: COLD START
Specific trap type: 0 (0)
Timestamp: 0
[Malformed Packet: SNMP]

I am fairly sure this is OS related & is down to the drivers for the USB
pods because we have two of these, both behave very similar under
Win2k/XP but work fine under Win98. Interestingly under XP using the OS
support (rather than the Belkin drivers) file transfers to the WAP
succeed without a problem but trying to copy a file from the WAP side
always fails with this similar problem.

Anyone else seen something similar or can shed some light on what
generates these traps/resets?

thanks in advance.

--
== jon bird - software engineer
== <reply to address _may_ be invalid, real mail below>
== <reduce rsi, stop using the shift key>
== posted as: news 'at' onastick 'dot' clara.co.uk