RH9 NFS Help requested

[DPA]

I am in need of some help, besides this forum, my last option is to call in a consultant--it seems that no matter what I do I cannot get the system to work. Below is a trouble sheet with information you may find helpful to keep from looking in places that have been looked at. Any and all help is appreciated--the ultimate goal for this NFS server is to implement beowulf and run any program we can get our hands on to test it.

Thanks in advance

Kip



-----------------------------------------------------------------------------
NFS Failures:

Static IP addresses in range of 192.168.0.1 (server) to 192.168.0.18 using subnet of 255.255.255.0

Realtek Ethernet card recognized and drivers working—
Set to device “eth0”
All dhcpd files have been edited and verified for consistence
DHCP does not work—network fails hence static network
Network communication successful, inter PC ping successful

Operating RedHat 9 fresh install, no updates—cannot seem to get to communicate with windows systems in order to gain access to the network for internet connectivity.
Upgraded from RedHat 7.3 which did work in text mode with DHCP
Changed to 7.3 Graphical and DHCP failed—hence RH9

In redhat-config-network
Hosts set to 127.0.0.1 (router)
DNS is set to localhost.localdomain 192.168.0.1

NFS server is up—but questionable settings implemented:
/root/Desktop/Share 192.168.0.1 “read only”
/root/Desktop/Share 192.168.0.255 –b “read only”
/root/Desktop/Share 192.168.0.7 “read only”

Share folder is located on server’s desktop with the intended use to test the NFS.

Final objective is to implement Beowulf or other distributed computing program.

Network is operating through 2 routers at 10Mbits/s
Hardware = 2 x 3Com linkbuilder 10BT running on CAT5 wire no longer than 15 feet

All nodes running on RH7.3 text mode in root usermode.
Is RH7.3 compatible with RH9 NFS?

Nodes running on PII 400MHz w/ 128MB RAM w/ 6.4GB HDD’s

What files are necessary on each of the nodes to access the NFS Server on server? What commands? Suggested walkthrough url?

[krohnjw]

I don't see a problem listed......

Where are you encountering a problem? Is the NFS share exported?

You will need portmap running and all of your share specified in /etc/exports.
After that run exportfs -ra

That will set up the exports.

After that it is necessary to know where the problem lies (ie where the mount is failing)
mount 192.168.1.X:/path/to/share /mount/point
to figure out where the problem lies.

[DPA]

within etc/exports files is listed:
/root/Desktop/Share 192.168.0.1 “read only”
/root/Desktop/Share 192.168.0.255 –b “read only”
/root/Desktop/Share 192.168.0.7 “read only”



Note that the server is 192.168.0.1 and we want it to be able to broadcast it between the range of .5 and .18

there was a change in our dhcpd.conf file where we had forgotten to remove the quotations from around the declaration parameter.

DHCP now active...

Static addresses have been lifting in preference to dynamic addresses.

so for right now, the main "problem" is: How do we get the nodes to gain access to the server's shared folder? i.e. NFS command on a node? We have it set to where root users on a node can gain access to the server with identification (we don't need security) to make it easier. Also, what should be put in the exports file. Is what's in there right? (I doubt it)


Thanks

[krohnjw]

I haven't seen the exports file formatted as so
Look at the nfs how to
http://nfs.sourceforge.net/nfs-howto/server.html

What does exportfs -ra return?

[DPA]

Thanks for all of your help, I appreciate every bit of it. I followed the link that you gave to me and did what it said. Upon restarting all of the services I typed in exportfs -ra and got the following return:

exportfs /etc/exports:1: syntax error : bad option list
192.168.0.7 /root/Desktop/Share : invalid argument
192.168.0.30 /root/Desktop/Share : invalid argument

Im not exactly sure what it means by invalid argument, according to the walkthrough, there wasn't any argument to be given to the hosts.allow file--Regardless, do you have any suggestions for that.

secondly:

each of the nodes reports this after entering

Mount 192.168.0.31:/root/Desktop/Share /etc/sample

Mount: RPC : Portmapper failure - RPC: Unable to receive

I typed in the ps aux | grep portmap command and got a return, and then checked it using the rpcinfo -p command, I am not sure what is going wrong, I feel that it can be any number of things. Thanks!

[pbharris]

hello,
here is what my exports file looks like:

/oggs                    sparches(ro,sync)
/home/pics/              sparches(ro,sync)
/tmp                     sparches(rw,sync)

is nfs running?

[krohnjw]

exportfs /etc/exports:1: syntax error : bad option list
192.168.0.7 /root/Desktop/Share : invalid argument
192.168.0.30 /root/Desktop/Share : invalid argument

It appears you have a problem with your exports file. Post it up here.

[DPA]

exports file:

/root/Desktop/Share 192.168.1.0(ro)/255.255.255.0(ro)


Due to unknown complications, DHCP has again failed us and we have resorted to using static addresses across the board. Interestingly enough this proves to be more useful as we only have to enter in the IP addresses once in the server for the distribution of CPU power.

We cannot seem to get NFS running--that is, the service says that it is in fact running, but does not seem to supply the nodes with the files inside of the Share folder.

when running in the terminal and checking on the status of NFS, this is what pops up:

/sbin/service nfs status
rpc.mountd (pid 1725) is running...

yet none of the nodes can access the server.

Can you point me to a url or show me how i should be accessing the server through the network? I fear that I may be doing something wrong.

Also. The server is running RH9 whereas each node is running RH7.3, is this a problem?

Thanks again.

Kip

[krohnjw]

/root/Desktop/Share 192.168.1.0(ro)/255.255.255.0(ro)

Try this, your line is malformed.

/root/Desktop/Share 192.168.1.0/255.255.255.0(ro)

Look at the nfs how-to I posted above if you want more information for setting up an NFS server. This should be a simple thing and results in nothing more than exporting the nfs mount, then mounting it via mount

Make the above change and exportfs -ra and see if there is an error. There should be no problems with the client machines running 7.3