packet monitoring in high availability cluster

Is there a reason you can't use ping? If you want to check a particular application or port (like if you've clustered SQL Server and want to see when it comes back up on port 1433), I'd recommend a tool called tcping:

http://www.elifulkerson.com/projects/tcping.php

If you've not used it before, it's an awesome tool that let's you perform ping functionality to any TCP port. Using this tool, you can repeatedly ping your application port, see it respond, see it unresponsive while you perform your failover, and then see it respond again once the service comes back up.

hello

thank you. I think I wasn't precise in my question.

We have a tcp packet sender that sends many tcp packets on the network , and we want to check that during the cluster switchover there are no packet loses.

So we need a tool that will monitor and give precise tcp packet traffic info on the cluster.

Thank you

Packets will be lost during the cluster failover - either because the service isn't currently online or during the new seconds it takes to switch ownership of the VIP from one server to the other. It should be brief, but it will definitely drop a few packets (unless you're trying to test the retry mechanics of your sending application).

However, you can use something like wireshark to capture the network traffic, if that's what you're after - you'll get a lot of noise as well from regular network chatter between the servers, but it will capture packets sent and their acknowledgement (or lack of) for you to analyze.

thank you
But if we want to count exactly how many packets were transferred\received- is there a way to do that?

thank you very much