Host machine not able to ping the server until a ping is initiated from the server

[govindyad1]

We are facing an issue in the network.

We are not able to ping the server from the host machine until we initiate a ping from server to the host.

We are having 1 Cisco L3 Switch {(C3560E-UNIVERSALK9-M), Version 12.2(55)SE5, RELEASE SOFTWARE (fc1)} which is directly connected to 6 Cisco L2 switch{(C2960-LANLITEK9-M), Version 12.2(55)SE7, RELEASE SOFTWARE (fc1)}

L3 Switch is connected with 9 Redhat servers and 6 L2 Switches.

In L3 switch we have Vlan100 Server(192.168.1.1) , Vlan110 Client(172.168.1.1), and Ip routing is enabled

Issue:

When i try to ping from host to server its not pinging until we first initiate a ping from the server to client. Once a ping is initiated from the server to host the connectivity between server and host is up.

We have to keep the ping on all the time from the host machine to have the connectivity up and running. If we stop the ping the connectivity stops and again we have to initiate a ping from server to get the connectivity up.

I checked all the arp entries they are find but am not able to find the exact problem.

Kindly help me out to resolve it.

 

[bill001g]

First be very sure to check the subnet masks on all the end devices and the vlan interfaces on the switch. If it is just 1 client machine I would be suspect of a duplicate ip address if it is multiple then it is more likely a switch issue.

It is pretty simplistic when this works correctly. The end devices know that the ip is outside their range so they do not issues a ARP but instead send the traffic to the switch. The switch does issues a ARP if it does not have the ip and then forwards the packet.

So it is a matter of finding out who is not following the rules. This is when it is nice to have commercial switches. What you need is a laptop that you can load wireshark on. Then use the port monitor function in the switch to mirror the port the client is on so you can see the data it is receiving and sending. I would start on the layer 2 switch and work your way to the layer 3 switch and eventually to the server port.

Either the machine is not receiving the ARP or it is not responding to it.

There are some other more complex things that can get messed up but they generally only occur when you are running things like HSRP when you have redundant layer 3 switches.

Yours is a extremely simplistic issue so I am still betting it is a misconfigured subnet mask on a device.

 

[govindyad1]

Thank you all for your support.

Finally i got the problem resolved.
Answer: That problem was occurring due to improper startup to network manager services in my rhel 6 server.
After properly configuring network manager in rhel 6 its working efficiently.