In my company we have to do a lot of end to end testing to make sure that the packets can get from point A to point B -- and many times there is a firewall blocking. I wrote a tool to check the connectivity in Go that will ping the endpoint on a given port, and respond back if the port is open, closed or if it times out. Usually if it times out I'll consider it to be blocked by a firewall. Then we have to traceroute to find out where the packet gets blocked. I am not sure if there is a way to tell from a packet what device that packet timed out on. It would be nice for my tool to return that the packet was blocked, and on what device that packet was blocked by.

Thanks!!

Richard M.

traceroute should offer the possibilty to set the first hop that should be used. This is done by setting the ttl to a given value before starting the loop.

Repro: run ping to any host,

dbillor@netdev2-westus2:~$ ping bn6-0100-0002-01sw
PING bn6-0100-0002-01sw.aznet.gbl (25.184.112.234) 56(84) bytes of data.
64 bytes from BN6-0100-0002-01SW.aznet.gbl (25.184.112.234): icmp_seq=1 ttl=48 time=62.2 ms
64 bytes from BN6-0100-0002-01SW.aznet.gbl (25.184.112.234): icmp_seq=2 ttl=48 time=62.2 ms
64 bytes from BN6-0100-0002-01SW.aznet.gbl (25.184.112.234): icmp_seq=3 ttl=48 time=62.2 ms
64 bytes from BN6-0100-0002-01SW.aznet.gbl (25.184.112.234): icmp_seq=4 ttl=48 time=62.2 ms

while ping is going run traceroute to any other host you will see entries of icmp requests from the ping to the initial host.

traceroute to ibr02.mwh01 (25.184.240.242), 65 hops max, 52 byte packets
1   MWH03-0101-0304-04T0-lo.aznet.gbl. (25.101.225.21)  550.236µs
2   MWH03-0101-0304-15T1-lo.aznet.gbl. (25.101.224.224)  562.237µs
3   MWH03-0101-0100-01T2-lo.aznet.gbl. (25.101.224.0)  551.736µs
4   MWH01-0100-0200-08RHW-lo.aznet.gbl. (10.244.23.41)  498.233µs
5   MWH01-0100-0100-02RA-lo.aznet.gbl. (10.244.23.1)  464.931µs
6   25.75.128.68 (25.75.128.68)  770.45µs
7   be-142-0.ibr03.mwh01.ntwk.msn.net. (104.44.21.163)  1.565803ms
8   ae141-0.icr01.mwh01.ntwk.msn.net. (104.44.21.156)  6.031596ms
9   MWH01-0100-0100-02MA.aznet.gbl. (25.75.131.15)  773.851µs
10  MWH01-0100-0103-01M1.aznet.gbl. (25.75.131.7)  1.212879ms
11  25.184.240.207 (25.184.240.207)  1.016467ms
12  BN6-0100-0002-01SW.aznet.gbl. (25.184.112.234)  308.874275ms
13  25.184.240.207 (25.184.240.207)  1.021867ms
14  25.184.240.207 (25.184.240.207)  1.163577ms
15  BN6-0100-0002-01SW.aznet.gbl. (25.184.112.234)  35.935159ms

This is due to just blindly looking at icmp requests in the receive socket.

$ go get github.com/aeden/traceroute/cmd
# github.com/aeden/traceroute
..\..\..\github.com\aeden\traceroute\traceroute.go:185: undefined: syscall.IPPROTO_ICMP
..\..\..\github.com\aeden\traceroute\traceroute.go:198: undefined: syscall.SO_RCVTIMEO