Solved NFS mount time out

Hi,


I installed FreeBSD in January for the first time.
Code:
FreeBSD freebsd 11.1-RELEASE FreeBSD 11.1-RELEASE #0 r321309: Fri Jul 21 02:08:28 UTC 2017     [EMAIL]root@releng2.nyi.freebsd.org[/EMAIL]:/usr/obj/usr/src/sys/GENERIC  amd64
With the base system there was no problem mounting directories from Ubuntu 16.04-Clients.

Now I'm configuring some services like DHCP and (unbound) DNS server. So I think there is a correlation to the new installed software. But after deactivation of this packages in rc.conf the problem with NFS continues.

So I need your help.

on the FreeBSD NFS server

Code:
[root@freebsd /etc]# showmount -e
Exports list on localhost:
/DATA_test                         192.168.150.51 192.168.150.85
[root@freebsd /etc]#

I set access-authority to public with

chmod -R 777 /DATA_test

So I believe that everything (on the server) is correct configured.

On my Ubuntu system I can see the exported path bit the mount timed out

I have build a small script to catch the exact time.

Code:
root@pc000085:/home/superusr# ./mnt_test.sh
Mi 14. Feb 10:53:37 CET 2018
Export list for 192.168.150.26:
/DATA_test 192.168.150.85,192.168.150.51
Mi 14. Feb 10:53:42 CET 2018
mount.nfs: Connection timed out
Mi 14. Feb 10:55:53 CET 2018
root@pc000085:/home/superusr#

I hope you can help me.

Thx Thanks a lot.

OlliP
 
Last edited:
Never set 777 permissions on anything. That's a really bad habit to get in to.

Note that your exports are only allowed for two IP addresses. So the most obvious reason why it's timing out is because the client has a different IP address.

Not related to your issue but you should regularly update the system too, you should have 11.1-RELEASE-p6.
https://www.freebsd.org/security/advisories.html
 
Hi SirDice,

I agree with you with setting permissions...
The server is offline(!) and only one Workstation is connected.
Code:
root@pc000085:~$ ifconfig
enp2s0    Link encap:Ethernet  Hardware Adresse 08:60:6e:6f:a0:99  
          inet Adresse:192.168.150.85
As you see the ip-adress is correct.
So it must be another problem...
 
Few other obvious ones, firewall on the server and/or firewall on the client.

From the client, what does rpcinfo -p 192.168.150.26 output? (That's assuming the NFS server's IP address is 192.168.150.26)
 
Yes, 192.168.150.26 is the FreeBSD NFS server and 192.168.150.85 the Ubuntu workstation.

From my Ubuntu I received this output:
Code:
root@pc000085:/home/superusr# rpcinfo -p 192.168.150.26
   program vers proto   port  service
    100000    4   tcp    111  portmapper
    100000    3   tcp    111  portmapper
    100000    2   tcp    111  portmapper
    100000    4   udp    111  portmapper
    100000    3   udp    111  portmapper
    100000    2   udp    111  portmapper
    100000    4     7    111  portmapper
    100000    3     7    111  portmapper
    100000    2     7    111  portmapper
    100005    1   udp    893  mountd
    100005    3   udp    893  mountd
    100005    1   tcp    893  mountd
    100005    3   tcp    893  mountd
    100003    2   udp   2049  nfs
    100003    3   udp   2049  nfs
    100003    2   tcp   2049  nfs
    100003    3   tcp   2049  nfs
root@pc000085:/home/superusr#
 
Ok, those look to be in order. If you restart NFS processes you need to do this in a specific order or things won't be registered properly. That would also result in time-outs even when all relevant services appear to be running.
 
I have restarted the freebsd-server several time (while trying to deactivate DHCP and DNS).
But this doesn't solve the problem.

Now I restart in this order
Code:
# killall -9 mountd
# /etc/rc.d/nfsd stop
# /etc/rc.d/rpcbind stop
# /etc/rc.d/rpcbind start
# /etc/rc.d/nfsd start
But the test returns the same result - timeout.
 
/etc/rc.d/nfsd start triggers the start mountd - so it is not nessecery to do it manualy.

i see it also from the start-comand output.
 
/etc/rc.d/nfsd start triggers the start mountd - so it is not necessary to do it manually.
You are correct. This wasn't the case many years ago. Seems I'm still somewhat stuck on old habits.

Is there a firewall active on the FreeBSD machine or on Ubuntu?
 
Hi SirDice,

sorry for the late reply.

There is no firewall active.

Yesterday i found an influence in my actual free wiring.

Server and Workstation has an connection to a hub (HUB not a switch).
From the hub there is a connection to the first router. this router has
normaly a connection to the internet router.

This seems (logical) ok. Everything is in two rooms which are side by side.
The cable from the hub to the router was disturbing (and internet this time
not nessecary). So i unplugged the hub-router connection.

After this disconnection the NFS-Server was avalilable and connectable.
This was surprising me. So i reconnected hub and router and the NFS-server
wasn't connectable.

Normaly i see no influence from the router... but the situation teaches me something else.
Maybe the router is broken?

Today i have not so much time... so i will send information later the day

Thank you first of all

OllIP
 
Server and Workstation has an connection to a hub (HUB not a switch).
I haven't seen a hub in ages. Time to throw that thing in the bin. A cheap 1Gbps switch only costs a few euro and will work much better.
 
Hi,

now i've changed the hub with an switch and controlled the wire.

When the cable connetion between the switch and router exists the nfs mount timed out.
When i unplugged the router the mount will be executed in seconds.

So i think there is a config problem with my gateway/dhcp/unbound. It looks that
the ip-packages will be forwarded to the router and then they died.
When the rooter is not connected the ip-package will be forwarded - coincidentally - to the nfs-server.
Or does i have a wrong opinion?
 
Can we see the network configuration for these devices -- ip address, subnet, gateway/default router, routing table. ( ifconfig and netstat -nr on FreeBSD, same commands work on CentOS, Ubuntu I'm clueless)
It almost sounds like the router and nfs-server have the same IP address.
 
Hi together,

i've just finished my new configuration...

While there was no sufficient progress finding the error AND while my server and workstation was without any influence to my work or data i
decided to reinstall all (delete of my freebsd-server and delete of my ubuntu-worktstation).

Then i installed from scratch the freebsd server, then the bind991 (installation) and then dhcp server.
Now everything (that what i've tested) woks fine... There is no cabling problem and when i start an workstation
it receives the ip-address from dhcp and is able to 'find' with my dns-server.

Thanks alot of hour help.

Greetings

OlliP
 
Back
Top