On Tue, 25 Aug 1998 04:16:20 Jeremy W. wrote:
>We use both NV/6000 4.1 and NV5.0a/NT. Our customer is very anxious to
>get us to agree that NetView status polling may be unreliable, and that
>interfaces may appear to be down when in fact, all that is happening is
>that the TCP/IP protocol on the machine being polled is TOO BUSY to
>answer the ICMP request!!!
>Is this in any way possible on TCP/IP for NT or AIX?
Yes. Netmon cannot listen forever for a
response, so he has a timeout and retry
value. If a device does not respond within
that configurable limit, he does generate
a down event.
>If the machine is TOO BUSY to answer a ping, is it not also TOO BUSY to
>answer request for resources (i.e. it is effectively down anyway)?
Absolutely. Another possible situation is
that the network itself is too overutilized
to respond to Netview status polling within
the allotted time.
>Is it possible that some WAN router between networks could be TOO BUSY
>to handle an ICMP packet? Would it not just keep trying?
What will usually happen is that the ICMP
request will get sent out. If it isn't
responded too in time, netmon will issue
an interface down event. If it responds
later, albeit late, it'll go ahead and issue
an interface up. If on the next polling
cycle the device is still down, nothing
happens, but if it is responding, an
interface up is generated.
>Any help from you network techies would be very useful!
Netview NT or AIX status polling will never
be 100 percent reliable, but you can alter
timeout variables and retry counts to make
it as close as possible. Remember that if
you have a large number of devices that are
showing down but are really busy, you
probably have too much traffic on your
network. Netview is still showing a problem,
though it may not be with your individual
devices, but the very core or your network.
Either way, your client has a problem that
NationsBank Network Management Systems Support
Free web-based email, Forever, From anywhere!