Forums

Resolved
0 votes
Just installed ClearOS on a VM hosted on esxi5.5. The VM has 3 nics. Here is my setup

nic0 WAN-----internet
nic1 lan-------10.0.0.0/24 network, clearOS as gateway on 10.0.0.254 - Network1
nic2 lan-------10.10.0.0/24 network, clearOS as gateway on 10.10.0.254 - Network2

Both networks have DHCP working and can communicate with internet just fine

Any system on Network1 can ping and system/ip on Network2. (Can RDP from Network1 to Network2)
Any system on Network2 cannot ping or connect to anything on Network1

I would like the clearOS box to be able to route traffic across these 2 networks and it seems to be working from Network1 as excepted.

Both networks are set to LAN, so from the network guide it should be able to pass traffic.
Title

I tried setting a static route, but all I managed to do was isolate the networks. That could have been myself not knowing what I am doing.
Title


I have seen a bunch of old articles about setting the iptables to allow this. I have no issue with this but I am confused on why it is working from Network1 to Network2 but not the other way around. Seems to me if they are both set to LAN then it should work both ways. So at this point I don't know if it is a firewall issue or a routing issue.


Is there a simple place for me to check or something simple to try before I go down the iptables route.

Thanks you
Tuesday, June 27 2017, 02:25 PM
Share this post:

Accepted Answer

Wednesday, June 28 2017, 12:41 PM - #Permalink
Resolved
2 votes
Looks like it was some sort of arp issue with the physical network that Network1 is connected to. I isolated the network with just a VM on it and it was routing correctly. I have not moved this system over from testing, so my setup is a little odd. Think i had a loop somewhere in there. At least I know it isn't with the clearOS box now.

Thanks for all the help!
The reply is currently minimized Show
Responses (3)
  • Accepted Answer

    Tuesday, June 27 2017, 04:23 PM - #Permalink
    Resolved
    0 votes
    One tool in your arsenal here is tcpdump. If you have it on 3 machines then you can determine where the packet is getting dropped:

    1) ClearOS server
    2) BoxA on network A
    3) BoxB on network B

    Then run tcpdump on these interfaces:

    1) ClearOS server NIC on network A
    2) ClearOS server NIC on network A
    3) NIC on BoxA
    4) NIC on BoxB

    Something similar to this:

    tcpdump -i eth2 icmp[icmptype]=icmp-echoreply or icmp[icmptype]=icmp-echo

    Then run a persistent ping test from the working host to set a baseline for what it is supposed to look like. Then stop that and run the same from the other side. This will show you where the failure is occurring.
    The reply is currently minimized Show
  • Accepted Answer

    Tuesday, June 27 2017, 03:36 PM - #Permalink
    Resolved
    0 votes
    Yes, both Network1 and Network2 are using the same clearOS server as the default gateway.
    The reply is currently minimized Show
  • Accepted Answer

    Tuesday, June 27 2017, 03:07 PM - #Permalink
    Resolved
    0 votes
    You are correct, if set to WAN they both should communication both ways. Quick question though, do both networks use the ClearOS server as their default gateway. If there is another gateway involved in either of the other networks it can cause a violation of stateful packet inspection.
    The reply is currently minimized Show
Your Reply