Host B <--> Router <--> Host A
- Host A: IP = 192.168.1.10, Net = 192.168.1.0/24, VLAN = 1, Default GW = 192.168.1.1 (Router)
- Host B: IP = 192.168.2.10, Net = 192.168.2.0/24, VLAN = 20, Default GW = 192.168.2.1 (Router)
- Router: IP = 192.168.1.1, 192.168.2.1, VLAN = 1, 20
All devices are connected to a switch with these VLANs configured.
Now, if I try to ping Host A from Host B, the following occurs: Host B makes an ARP request to find out the MAC-address of the router and sends the Ping request to the router. The router makes also an ARP request to find out the MAC-address of the destination Host A and forwards the Ping request to Host A. That's ok and that works.
ARP requests for another subnet??
Now the strange part: Host A, of course, tries to answer the Ping, but(!) it doesn't make an ARP request to find out the MAC-address of the router to send it the Ping-Reply to forward it to Host B. Instead of that it sends an ARP request asking for the MAC-address of Host B directly. Of course, that doesn't work, there will be no answer on the local subnet, because the broadcast domain is restricted to the VLAN 1.
ARP cache on Host A (192.168.1.10) looks like this:
# arp -an ? (192.168.1.1) at 16:bc:aa:f2:bc:44 [ether] on eth0 ? (192.168.2.10) at <incomplete> on eth0
When I try to delete the weird ARP resolution attempt, I get this message and the failed ARP attempt is still in cache:
# arp -d 192.168.2.10 SIOCDARP(dontpub): Network is unreachable
ICMP-Redirects from router
So, no (bidirectional) communication between Host A and B is possible. And instead of Ping-Replies, Host B, gets an ICMP-Redirect-Request from the router: Host B should send packages direclty to Host A.
- What makes Host B trying to send an answer by ARP resolving a host of another subnet? Why is it the Ping-Reply not sent to the router?
- Any idea what role the ICMP-Redirect plays?
# route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 0.0.0.0 192.168.1.1 0.0.0.0 UG 0 0 0 eth0 192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 # ip a s 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000 link/ether ab:cd:a9:9a:cc:dc brd ff:ff:ff:ff:ff:ff inet 192.168.1.10/24 brd 192.168.1.255 scope global eth0 valid_lft forever preferred_lft forever 3: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000 link/ether ab:cd:a9:9a:cc:dd brd ff:ff:ff:ff:ff:ff # ip r s default via 192.168.1.1 dev eth0 192.168.1.0/24 dev eth0 proto kernel scope link src 192.168.1.10
# route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 0.0.0.0 192.168.2.1 0.0.0.0 UG 0 0 0 eth0 192.168.2.0 0.0.0.0 255.255.255.0 U 1 0 0 eth0 169.254.0.0 0.0.0.0 255.255.0.0 U 1000 0 0 eth0 # ip a s 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 40:7d:7a:a3:f5:dd brd ff:ff:ff:ff:ff:ff inet 192.168.2.10/24 brd 192.168.2.255 scope global eth0 3: wlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc mq state DOWN qlen 1000 link/ether 47:5e:33:a6:31:5e brd ff:ff:ff:ff:ff:ff
Destination-IP Subnet mask Default gateway Hop count Interface <public-net> 255.255.255.224 * 0 eth2 <public-net> 255.255.255.224 * 0 eth1 192.168.1.0 255.255.255.0 * 0 eth0 192.168.2.0 255.255.255.0 * 0 eth0 default 0.0.0.0 <public-router> 15 eth1 default 0.0.0.0 <public-router> 40 eth2 default 0.0.0.0 <public-router> 40 eth1
public-net ...... Address of public subnet (internet-uplink)
public-router ... Address of uplink-router
Router is a Cisco RV320 with web interface only, that's all I can get. PS: It's a load balancing dual uplink setup, but that shouldn't make a difference for the ARP problem.