nfs: port mapper failure

S

sehrguey

Dear all,

There are 2 FreeBSD 7.4 boxes linked via ssh, pinging each other ok. All my tries to establish an nfs connection between them fell through. I faithfully followed the Handbook in their configuration:

Serverside /etc/rc.conf:
Code: 
rpcbind_enable=”YES”
nfs_server_enable=”YES”
nfs_server_flags=” –u  -t -n  4 –h xxx.xxx.xxx.14”
mountd_flags=”-r”
/etc/exports:
Code: 
/usr/ports –maproot=root alldirs, quiet testbox
Clientside /etc/rc.conf:
Code: 
nfs_client_enable=”YES”
nfs_client_flags=”-n 4”
But when I try the command`
mount lab:/usr/ports /mnt
the result is:
Code: 
[udp] lab:/usr/ports:RPCPROG_NFS:RPC:Port mapper failure:Timed out
Serverside, cat /var/log/messages reveals:
Code: 
can’t bind udp addr xxx.xxx.xxx.14  Can’t assign requested address

Help if you can, please.

sehrguey
 
Dear DutchDaemon,

Thank you dearly for editing my first post and pointing out the forum usage rules. Still I don’t know what linkdumping or profile pages means nor do I care about the things, my only wish is to solve the nfs connection problem I'm scrambling with. Anyway and once again I do appreciate your kind cooperation.

Yours
sehrguey
 
Dear SirDice,
Thank you for the attempted help.
Your tip was promptly used and now at rebooting the server box does not complain about impossibility to assign the requested address to the client box.
All the rest remains unchanged

# mount lab:/usr/ports /mnt
Code: 
[udp] lab:/usr/ports:RPCPROG_NFS:RPC:Port mapper failure:Timed out

Yours,
Sehrguey
 
Just ignore it. NFS will try to use TCP or UDP. The message just indicates it can't use UDP for some reason.
 
SirDice said:
Just ignore it. NFS will try to use TCP or UDP. The message just indicates it can't use UDP for some reason.
Click to expand...

Dear SirDice,

Could you, please, be more specific about how long it is reasonable to ignore not getting the goodies promised by the holy Handbook?

The message
Code: 
[udp] lab:/usr/ports:RPCPROG_NFS:RPC:Port mapper failure:Timed out
keeps repeating itself for hours.

Besides, trying to figure out what is what, I typed the following on the client box:
# rpcinfo
and got the message:
Code: 
rpcbind: Can’t contact rpcbind: RPC: Port mapper failure – RPC: Success

The same command on the server box results in a long, nice-looking table.

It makes me dizzy and I just do not know what to do – to start another thread on rpcbind's “failure-Success”, or to just ignore it too?

yours,
sehrguey
 
Here's what I have on my server:
Code: 
rpcbind_enable="YES"
rpcbind_flags="-h 192.168.1.190 -h 2001:888:1c5b::190"
mountd_enable="YES"
mountd_flags="-h 192.168.1.190 -h 2001:888:1c5b::190"
nfs_server_enable="YES"
nfs_server_flags="-u -t -n 4 -h 192.168.1.190 -h 2001:888:1c5b::190"

Note the -h options are only there to bind the services to specific IP addresses.

On my client:
Code: 
rpcbind_enable="YES"
 
Dear SirDice,

Thank you for your prompt reply.

Before I start making changes to my box's configuration would you kindly elucidate if any of the specific IP addresses in your server box's rpcbind_flags configuration is that of the server box itself?

Yours,
sehrguey
 
sehrguey said:
Before I start making changes to my boxes configuration would you kindly illucidate
if any of the specific IP addresses in your server box rpcbind_flags configuration is that of the server box itself?
Click to expand...
Obviously.

Code: 
ifconfig_re0="192.168.1.190  netmask 255.255.255.0"
ipv6_ifconfig_re0="2001:888:1c5b::190 prefixlen 64"
 
Dear SirDice,

Enabling rpcbind in the client's rc.conf made its rpcinfo work ok, (though "Chapter 29.3 NFS" in the Handbook never mentioned it is needed on the client side). Then I copied your (working) server configuration into my server box's rc.conf, (substituting, of course, the IP addresses for the local ones), rebooted both boxes and had the same message:

Code: 
[udp] lab:/usr/ports:RPCPROG_NFS:RPC:Port mapper failure:Timed out
going on and on and on.

Stumbling in an elementary stage is a discouraging experience, on the other hand it's not your job to put aright another guy's boxes, still and yet: what am I to do, sir SirDice?

I can't even imagine in what direction to dig.

Yours,
sehrguey
 
Network details please. Are both in same LAN, if not, what's in between, VPN/firewalls. Start digging there.
 
It's probably a firewall that's blocking the UDP requests.
 
Dear Zare and SirDice,

As I mentioned in the first post the boxes have ssh connection.
They are not on the same LAN – server box is at the ISP’s headquarters and the client box at my home.

There is no firewall whatsoever between them just patchcord, switch, modem, dish, another dish, another modem, another switch, another patchcord.

Thanks for your attention.
sehrguey
 
You must log in or register to reply here.
Back
Top