3
0
I have a 32-bit centos 6 machine and a windows machine on the same switch: 192.168.92.26 and 27.
Ping from the Centos box to Windows shows ping times less than 1ms and vice versa.
tracert from windows to Centos shows ok:
However, trying the opposite, from Centos to Windows shows only stars to the max hop value... Any idea why?
The two machines otherwise communicate without problem.
Danny
Posted 2017-09-06T08:05:57.463
Reputation: 151
4
Traceroute works by sending pings with increasing Time To Lives, and watching the address the responses come from.
By default the windows firewall drops inbound ping requests, so traceroute to the windows box won't work either, as it uses them to trace the route to the remote machine.
By default, Windows Firewall does not allow incoming ICMP Echo messages
If you follow the steps in the source you can allow them to be received, and traceroute should work, as the machine will send responses to the ping requests instead of just dropping the requests.
jrtapsell
Posted 2017-09-06T08:05:57.463
Reputation: 408
1
Traceroute sends out packets (usually UDP, but might be ICMP depending on the client implementation) with a TTL that increases over the number of hops. The destination host will see that the TTL of the packet reaches zero, and will send back an ICMP Time Exceeded message back to the original host. This is what traceroute measures and displays.
If the remote host (or the local host) has a firewall (or other settings) that block ICMP Time Exceeded messages, then traceroute wouldn't actually show anything because no ICMP messages are returned.
You might try disabling Windows firewall and see if that makes a difference.
mtak
Posted 2017-09-06T08:05:57.463
Reputation: 11 805
Disabling firewall make no difference for me. – John Tribe – 2018-11-19T03:42:04.243
The firewall part is the key here. – Kitet – 2017-09-06T10:05:43.240
Updated my answer to emphasize the firewall aspect – jrtapsell – 2017-09-06T10:21:31.433