I have a laptop working as an NFS client and a desktop as the NFS server, both
When I have my laptop connected to my Lynksys cable/dsl router using a cable I
am able to mount the filesystem with no problem. The issue that I have is
when I use my wireless connection in my laptop I am not able to mount the
filesystem. I am able to "rpcinfo -p nfs_server" from my laptop, but when I
try mount it I get an "mount: RPC: Timed out" .
Again, everything works fine when using a cable connection. Using my wireless
connection I am able to ping the nfs server, ssh, rpcinfo etc., except mount
The command I used is "mount alvaro_linux:/home/public /home/alvaro/linux"
mount: RPC: Timed out
Please, I would really appreciate any suggestions regarding this issue.
What seems to be the difference in how the connections are handling NFS?
Does the client still fall into the allowed IP range when on wireless connection? Check the logs on both the client and server, and monitor the network with ethereal or your favorite network monitor.
Thank you very much for your suggestions. I still have some issues and here they are:
1. Yes, the IP range falls into allowed IP addresses.
2. The logs from both (nfs server and client) show no errors with regards to the connection - I check the /var/log/messages and found no relevant information.
3. I was not able to run ethereal, keep having problems with "libsnmp.so.0", so I finally gave that up, but do you have any other suggestiongs for the network monitor tool?
Question? do you think this issue is somehow related to the fact that my wireless is set up as ETH1, not eth0 (which is the cable connection?) -- this is just a wild suggestion, since I am running out of ideas.
If you could shed some light I would really, really appreciate it.
NFS - Cont....
This is an abstract from /var/log/messages on the NFS server after I tried to mount the filesystem:
Dec 23 17:46:29 alvaro_linux rpc.mountd: authenticated mount request from Linux:848 for /home/public (/home/public)
Dec 23 17:46:47 alvaro_linux last message repeated 6 times
there is nothing on the client side...
No, I don't think it has to do with it being eth1 instead of eth0. That shouldn't matter.
It almost seems that the client machine doesn't receive the replies, for some reason. Have you checked so that the client isn't running a firewall of some sort?
I don't know of a better network monitor than ethereal, since it performs packet analysis and stuff, not to mention the UI, which is really good. I'd try and solve the libsnmp issue if I were you.
I just wanted to thank you for suggesting ethereal, after carefully reviewing the log -- which by the way, it shows everything! -- I was able to see that there was a conflict after the server received the request from the client. With that information I decided to completely shutdown IPTABLES and wham!! that was it...it seems sort of silly that I did not check that before, but in reality what i did on a session was to use chkconfig to stop the service, instead of using the /etc/rc.d/init.d/iptables stop -- duh!
Now, the issue that I need to address is the fact that I shutdown the IPTABLES...but that's another story....thanks for your suggestion. :D