on Centos 6.5 i have a bridge network connected to eth4 and my virtual machine (Centos6.5) running on KVM is also connected to that interface(vnet0)
bridge name bridge id STP enabled interfaces
network 8000.588d090ee15e no eth4
vnet0
VM is having an IP address 10.99.9.10 and host an ip address of 10.99.9.2 on eth2
VM can communicate outside,and its reachable from other Hosts,but not from the Host where its running.
pinged a different Host (10.99.9.3) from VM and it works fine tcpdump shows
ping 10.99.9.3 -c 3
tcpdump -n icmp
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth0, link-type EN10MB (Ethernet), capture size 65535 bytes
02:06:52.701448 IP 10.99.9.10 > 10.99.9.3: ICMP echo request, id 44807, seq 1, length 64
02:06:52.701896 IP 10.99.9.3 > 10.99.9.10: ICMP echo reply, id 44807, seq 1, length 64
02:06:53.701434 IP 10.99.9.10 > 10.99.9.3: ICMP echo request, id 44807, seq 2, length 64
02:06:53.701904 IP 10.99.9.3 > 10.99.9.10: ICMP echo reply, id 44807, seq 2, length 64
02:06:54.701117 IP 10.99.9.10 > 10.99.9.3: ICMP echo request, id 44807, seq 3, length 64
02:06:54.701489 IP 10.99.9.3 > 10.99.9.10: ICMP echo reply, id 44807, seq 3, length 64
But if i ping the same host 10.99.9.2
ping 10.99.9.2 -c 3
PING 10.99.9.2 (10.99.9.2) 56(84) bytes of data.
From 10.99.9.10 icmp_seq=1 Destination Host Unreachable
From 10.99.9.10 icmp_seq=2 Destination Host Unreachable
From 10.99.9.10 icmp_seq=3 Destination Host Unreachable
--- 10.99.9.2 ping statistics --- 3 packets transmitted, 0 received, +3 errors, 100% packet loss, time 3001ms pipe 3
For me its strange that tcpdump -n icmp shows nothing!!
What could be the reason for this behavior,even if something blocks why isnt tcpdump not showing even ICMP request in tcpdump.
I only have one interface inside VM and tcpdump is performed on only available interface eth0