I am getting the following NFS/TCP related errors in the message logs on my nfs (NFSv3) clients:
The clients mount directories from a central nfs server and are running NFS over tcp. All are running FreeBSD 6.2-RELEASE. The errors show up intermittently (i.e approximately once every two hours). The server is on a different network than the clients and connects to the clients via a fortigate firewall/router with the server connecting to the firewall's internal interface (192.168.1.99) and the clients connecting to the firewall's dmz interface (10.0.0.1). I have assigned a secondary ip address (10.0.0.100) to the internal interface and created a static route from the nfs server's nic (192.168.1.100) to the 10.0.0.100 ip address.
The files in the shared directories are web files that are served up on the internet via apache webservers.
Any pointers on how to track down the problem would be greatly appreciated
Code:
host kernel: nfs/tcp clnt: Error 60 reading socket, tearing down TCP connectionThe clients mount directories from a central nfs server and are running NFS over tcp. All are running FreeBSD 6.2-RELEASE. The errors show up intermittently (i.e approximately once every two hours). The server is on a different network than the clients and connects to the clients via a fortigate firewall/router with the server connecting to the firewall's internal interface (192.168.1.99) and the clients connecting to the firewall's dmz interface (10.0.0.1). I have assigned a secondary ip address (10.0.0.100) to the internal interface and created a static route from the nfs server's nic (192.168.1.100) to the 10.0.0.100 ip address.
The files in the shared directories are web files that are served up on the internet via apache webservers.
Any pointers on how to track down the problem would be greatly appreciated