| CPUG | |
| The Check Point User Group | |
| A Resource For The Check Point Community. Fast. Useful. Independent. | |
|
| |||||||
![]() |
| | LinkBack | Thread Tools | Display Modes |
| |||
| We have the following problem at work : We tried to bring in the secondary firewall into the cluster. The issue that we are seeing is that when the secondary node joins the cluster then outgoing NAT starts to fail randomly. When I try and telnet on port 80 to an external website from the firewall command line it is successful from one and not the other firewall. I don’t believe this is normal behaviour Also I am not seeing any “permanent published (proxy only)” when I do an arp –a. I should be seeing them as we are doing auto NAT When running under following command stylecpf2[admin]# fw ctl arp No proxy ARP entries Which appears incorrect Below is the fw ver command stylecpf2[admin]# fw ver This is Check Point VPN-1(TM) & FireWall-1(R) NG with Application Intelligence (R55) HFA_06 for IPSO 3.8, Hotfix 624 - Build 004 We have also tried to reboot to solve this issue but without success. it's Nokia IP380 box. If you have any idea? Thank you |
| |||
| I have suspicions that you need to use the Proxy ARP section in IPSO and add ARP entries using the CLUSTER MAC (as opposed to the previously unclustered single firewall before). You do say you are using auto-NAT but R55 is so old, I can't remember the caviats to Auto-NAT on an IPSO cluster. I always used auto-nat just so I wouldn't have to mess with manual rules and then added Proxy ARP entries for all of my public IPs into IPSO so that my upstream routers would always be able to forward traffic to my gateways even if the firewall rejected it (so I could see drops in logs for unused IPs). There are MANY ways to do ARP... maybe this info will help you. __________________ There's no place like 127.0.0.1 Last edited by lammbo; 2008-02-18 at 13:12. |
| |||
| Also check the "external gateway" for the firewall it is possible that the arp time out on the routers is to high. When you have access to the gateway recreate the problem and try to manual delete the arp entry. Also make sure that both the node in the cluster use the same magic mac addressing. Eduard |
![]() |
| Thread Tools | |
| Display Modes | |
| |