[daceo]

I set up myself a sever using menu, system settings, server settings, NFS, and I set the directory to my home directory, and the host(s) to the IP address of the machine I want to share with (I am using static IP addresses). I then set up my suse 8.2 machine to be a client, specifying the appropriate paths and options (rw). This worked, and I can access the home on the fedora box.
But when i come to use them the next day, it doesn't work. The suse complains that “RPC: Remote system error-No route to host” Now I originally thought it must be some thing wrong with the suse, so I installed fur fedora core 1 on a spare partition on that machine (over the top of win98!!!!!!!), and low and behold it has the same problem. So I suspect the fedora. My exports says
“/home/freeda 192.168.0.30(rw,insecure,sync)”,
my 2001 inside linux book says try “ps ax | grep portmap” which come back with
” 4420 ? S 0:00 portmap
5046 pts/1 S 0:00 grep portmap”
it also says try rpcinfo -p
“[freeda@fedora freeda]$ rpcinfo -p
bash: rpcinfo: command not found”
this is about as far as I have got. I feel it shouldn't be that difficult.
Also while I am here I have been using gftp, and it corrupts the JPG I have uploaded to my website. (the Kbear works ok on the suse 8.2 (but the most up to date pics are on the fedora and the network is not working...)).

Any help would be appreciated

cheers.

[mhelios]

For your NFS setup, I would suggest following along with this doc to see if you've done things right:
http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/ref-guide/ch-nfs.html

For the "missing" rpcinfo command, whenever that happens, it means it's not in your path. To find out the program location you can use a nice little program called "whereis"
So `whereis rpcinfo` produces /usr/sbin/rpcinfo.

For your gftp image corruption, make sure you set the transfer mode to binary.

[daceo]

ok, Am confused, even if I go in to directory where rpcinfo lives and try and run it (as root or not) it still doesn't work (and I have noticed this with other things as well) .... for example:
"[root@fedora sbin]# whereis rpcinfo
rpcinfo: /usr/sbin/rpcinfo /usr/share/man/man8/rpcinfo.8.gz
[root@fedora sbin]# cd /usr/sbin
[root@fedora sbin]# rpcinfo -p
bash: rpcinfo: command not found
[root@fedora sbin]#"

the "setup" program in that directory works,

confused

[mhelios]

Ok, if you're sure you're running the program as root, then check your root's PATH contains /usr/sbin/ in it.
To do that issue a echo $PATH.
It should look something like:

/usr/kerberos/sbin:/usr/kerberos/bin:/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin:/usr/X11R6/bin:/root/bin


The command not found should only happen for non-root users because "/usr/sbin" is not in their PATH.
For non-root user's to run it and any program not in their path, you must type the full pathname to the program:

$/usr/sbin/rpcinfo -p

[daceo]

Ok slowly we get there, it is
"[root@fedora freeda]# echo $PATH
/usr/kerberos/sbin:/usr/kerberos/bin:/usr/local/bin:/usr/bin:/bin:/usr/X11R6/bin :/home/freeda/bin
[root@fedora freeda]#"
which probably explains something, and
"[root@fedora freeda]# /usr/sbin/rpcinfo -p
rpcinfo: can't contact portmapper: RPC: Remote system error - Connection refused"

which is what we knew already, but it is nice to see it like that.
Incidentaly, I have now got the two machines to be clients on my p166 slack9.1 experimental server...
and it seems that my gftp is set to binary.
thanks for the help so far, I will be interested on how to move forward on this one....
cheers

[mhelios]

Ok, first, I bet when you switch to the root user, you're doing it with:
`su root` right?

DON'T do it like this! This will change your id but not your environment. That would be the reason why your PATH doesn't contain /usr/sbin/, /sbin and the other root PATHs.

So, to switch to root's enviroment, ALWAYS change users to root like this:
`su -`
The hyphen does the magic. WHen you want to switch back to your regular user, do:
`su - user` where user is your username.

Now, moving on. The error you're getting:
rpcinfo: can't contact portmapper: RPC: Remote system error - Connection refused"
is because portmapper service is not running. As root, issue:
`/sbin/service portmap start` to start it up then run the rpcinfo command again.
In fact, I would bet that your NFS problem is due to portmapper not running.

HTH

[daceo]

spot on with the "-", it has even changed me to root as well!!!!!!! on me propt!!, so now we have,
"[root@fedora root]# /sbin/service portmap start
Starting portmapper: [ OK ]
[root@fedora root]# rpcinfo -p
rpcinfo: can't contact portmapper: RPC: Remote system error - Connection refused
[root@fedora root]#"

I looked into the"service configuration" box, and and portmap is ticked, as is netfs, network, nfslock, and a few other things, nfs is not ticked, that might have been me... I am not sure. having said that the nfs is working as we speak as a client onto my slack server. I just tried clicking on the nfs and then restart, but it (service configuration) seemed to crash... nevermind, I still get connection refused on the rpcinfo. I will now boldly retry clicking on the nfs, saving and restarting the machine.....
cheers

[mhelios]

Are you sure portmapper is running? This message should only occur when it's not.
Type this to confirm:
#service portmap status

You should get this:
portmap (pid 1234) is running...

Also, try adding localhost or your IP address to the rpcinfo command:
rpcinfo -p localhost
rpcinfo -p 192.168.x.x

You might want to add the IP address and your slackware's IP address to your /etc/hosts.allow file.

[daceo]

"[root@fedora root]# service portmap status
portmap (pid 4386) is running...
[root@fedora root]# rpcinfo -p 192.168.0.20
program vers proto port
100000 2 tcp 111 portmapper
100000 2 udp 111 portmapper
100024 1 udp 32768 status
100024 1 tcp 32770 status
391002 2 tcp 32771 sgi_fam
[root@fedora root]#"

this is why I am confused... it seemed to me that it was running. I shall now try the hosts allow.
cheers

[ywoly]

On your server try:
service nfs start

showmount -e

Showmount should show what you are exporting from your server.

On your client try:

showmount -e <servername>

You should see the exported filesystems.

I'm assuming that you don't have iptables running.

If this works, then on the server do:
chkconfig nfs on

Good Luck.

[daceo]

Hi, tried the show mounton the serverand client:

[root@fedora root]# service nfs start
Starting NFS services: [ OK ]
Starting NFS quotas: [ OK ]
Starting NFS daemon: [ OK ]
Starting NFS mountd: [ OK ]
[root@fedora root]# showmount -e
mount clntudp_create: RPC: Port mapper failure - RPC: Timed out
[root@fedora root]#

suse:~ # showmount -e 192.168.0.20
mount clntudp_create: RPC: Program not registered



and this is the result. I have no fire wall setup (on this machine).I feel sure that I may have missed some thing obvious. Please feel free to think I am absolutly clueless, even though I got it to work on the slackware quite easily. I am a fairly noob to the nuts and bolts of linux...
cheers

[daceo]

I have given up on this aproach now as I now have a proper server running, it is in a cubord to so less noise!!, and all my pcs can talk to it.. thanks for all the help however, I am more learned now..

cheers

[mhelios]

Great news daceo. :) Don't forget to keep learning..

Set as RESOLVED.