0

I'm using a utility to create an ICMP packet flood/attack called Hyanea since windows doesn't have native support like linux does.

I want to be able to test a new AP to see if any packet loss occurs. My target is a laptop. Only issue is, I think before the AP has trouble handling the packets, I believe the device itself will not be able to handle it and misreport how many packets really got delivered. The target machine I am running is using Wireshark. Again, my goal is to measure at which point I start seeing packet loss on the AP, not the device.

I don't have many resources which is why I'm using another laptop machine as a target. Is there any quick way to test this without setting up a server to calculate the packet loss?

Faahmed
  • 113
  • 4
  • Are you looking for packet loss on the wired side of a WAP? – Ron Maupin Jan 03 '16 at 02:20
  • the wireless side. Both are connected to the same AP wirelessly. No wired connections. Although now I feel as I should probably connect at least one, either the generator of the flood or the target device to ethernet. – Faahmed Jan 03 '16 at 02:28
  • Well, you should have dropped packets on the wireless side since that is the nature of wireless. It is a half-duplex technology, and, although it uses collision avoidance, you cannot prevent collisions (which can't be detected by the sender because radios can't send and receive at the same time) and data loss. – Ron Maupin Jan 03 '16 at 02:39
  • How can I measure how many packets were dropped? – Faahmed Jan 03 '16 at 02:45
  • Count the packets sent on one workstation, and count the packets received on the other; the WAP has no idea if any sent packets are actually received. You need to figure out some things. For instance, when you send packets which need a reply packet, you are increasing the likelihood of packet loss since the replies will be vying for the same medium you are using to send. I'm not sure of your motives for this – Ron Maupin Jan 03 '16 at 02:51
  • My motive is simply to test two different APs for how much they can handle before packet loss begins to happen. I'm evaluating from two different vendors. I also have the ability to send a ICMP unreachable (TCP) packets rather than pings using the tool. My worry is, the packets on the receiving end may get delivered, but the NIC or CPU of the targeted machine may not handle it. I'm thinking of changing my strategy and using multiple devices instead so I can be sure the devices are able to correctly count the number of packets delivered. – Faahmed Jan 03 '16 at 02:56
  • Let us [continue this discussion in chat](http://chat.stackexchange.com/rooms/33761/discussion-between-an-alien-and-ron-maupin). – Faahmed Jan 03 '16 at 02:58

0 Answers0