I am having trouble with nfs mounting I have machines A, which is
supposed to be an nfs server, and B a client.
On A I have
/etc/exports
/local 142.103.xxx.0/26(no_subtree_check,rw,no_root_squash)
where B's address and A's address are both in that set of IP addresses.

On B I have
/etc/fstab
A:/local/ /local1/ nfs4 rw,rsize=8192,wsize=8192 0 0

The directory /local1 exists on B
B:[root]>ls -la /local1
total 8
drwx------ 2 root root 4096 Jan 11 2023 .
drwxr-xr-x 32 root root 4096 Jan 11 2023 ..

I now run
mount /local1
on B and it times out. I cannot see anything in the /var/log/messages or
/var/log/syslog or dmesg on A as to why the mount failed, and similarly
on B.

I note that nfs is now version 4.2 Could that have changed something? Is
this a bug, or more likely a bug sitting in front of the keyboard.

On Thu, 19 Oct 2023 20:52:51 -0000 (UTC), William Unruh wrote:

> I am having trouble with nfs mounting I have machines A, which is
> supposed to be an nfs server, and B a client.
> On A I have
> /etc/exports
> /local 142.103.xxx.0/26(no_subtree_check,rw,no_root_squash)
> where B's address and A's address are both in that set of IP addresses.
>
> On B I have
> /etc/fstab
> A:/local/ /local1/ nfs4 rw,rsize=8192,wsize=8192 0 0
>
> The directory /local1 exists on B
> B:[root]>ls -la /local1
> total 8
> drwx------ 2 root root 4096 Jan 11 2023 .
> drwxr-xr-x 32 root root 4096 Jan 11 2023 ..
>
> I now run
> mount /local1
> on B and it times out. I cannot see anything in the /var/log/messages or
> /var/log/syslog or dmesg on A as to why the mount failed, and similarly
> on B.
>
> I note that nfs is now version 4.2 Could that have changed something? Is
> this a bug, or more likely a bug sitting in front of the keyboard.

Have you tried using mcc to go to Network Sharing, then to Configure
NFS Shares, and then to Access NFS shared drives and directories or
Share drives and directories using NFS, and there clicked on
Search for new servers (on my machine this takes a long
time), and then looked at each machine that is listed ?

Looking at things this way is a PITA, especially if the settings
have to be revised or created, but it eventually works, for my
systems at least.

Cheers!

jim b.

--
UNIX is not user-unfriendly, it merely
expects users to be computer friendly.

On 2023-10-20, Jim <jim.beard@verizon.net> wrote:
> On Thu, 19 Oct 2023 20:52:51 -0000 (UTC), William Unruh wrote:
>
>> I am having trouble with nfs mounting I have machines A, which is
>> supposed to be an nfs server, and B a client.
>> On A I have
>> /etc/exports
>> /local 142.103.xxx.0/26(no_subtree_check,rw,no_root_squash)
>> where B's address and A's address are both in that set of IP addresses.
>>
>> On B I have
>> /etc/fstab
>> A:/local/ /local1/ nfs4 rw,rsize=8192,wsize=8192 0 0
>>
>> The directory /local1 exists on B
>> B:[root]>ls -la /local1
>> total 8
>> drwx------ 2 root root 4096 Jan 11 2023 .
>> drwxr-xr-x 32 root root 4096 Jan 11 2023 ..
>>
>> I now run
>> mount /local1
>> on B and it times out. I cannot see anything in the /var/log/messages or
>> /var/log/syslog or dmesg on A as to why the mount failed, and similarly
>> on B.
>>
>> I note that nfs is now version 4.2 Could that have changed something? Is
>> this a bug, or more likely a bug sitting in front of the keyboard.
>
> Have you tried using mcc to go to Network Sharing, then to Configure
> NFS Shares, and then to Access NFS shared drives and directories or
> Share drives and directories using NFS, and there clicked on
> Search for new servers (on my machine this takes a long
> time), and then looked at each machine that is listed ?
>
> Looking at things this way is a PITA, especially if the settings
> have to be revised or created, but it eventually works, for my
> systems at least.

Unfortunately I can only access A and B remotely, and mcc--> NetworkSharing
does not have a text version. (ie diskdrake --nfs only has a gtk
version and claims it cannot run in a console-- even though I have X
sharing (Display=10.0)

On Fri, 20 Oct 2023 18:41:56 -0000 (UTC) William Unruh wrote:

> On 2023-10-20, Jim <jim.beard@verizon.net> wrote:
>> On Thu, 19 Oct 2023 20:52:51 -0000 (UTC), William Unruh wrote:
>>
>>> I am having trouble with nfs mounting I have machines A, which is
>>> supposed to be an nfs server, and B a client.
>>> On A I have /etc/exports /local
>>> 142.103.xxx.0/26(no_subtree_check,rw,no_root_squash) where B's address
>>> and A's address are both in that set of IP addresses.
>>>
>>> On B I have /etc/fstab A:/local/ /local1/ nfs4
>>> rw,rsize=8192,wsize=8192 0 0
>>>
>>> The directory /local1 exists on B B:[root]>ls -la /local1 total 8
>>> drwx------ 2 root root 4096 Jan 11 2023 .
>>> drwxr-xr-x 32 root root 4096 Jan 11 2023 ..
>>>
>>> I now run mount /local1 on B and it times out. I cannot see anything
>>> in the /var/log/messages or /var/log/syslog or dmesg on A as to why
>>> the mount failed, and similarly on B.
>>>
>>> I note that nfs is now version 4.2 Could that have changed something?
>>> Is this a bug, or more likely a bug sitting in front of the keyboard.
>>
>> Have you tried using mcc to go to Network Sharing, then to Configure
>> NFS Shares, and then to Access NFS shared drives and directories or
>> Share drives and directories using NFS, and there clicked on Search for
>> new servers (on my machine this takes a long time), and then looked at
>> each machine that is listed ?
>>
>> Looking at things this way is a PITA, especially if the settings have
>> to be revised or created, but it eventually works, for my systems at
>> least.
>
> Unfortunately I can only access A and B remotely, and mcc-->
> NetworkSharing does not have a text version. (ie diskdrake --nfs only
> has a gtk version and claims it cannot run in a console-- even though I
> have X sharing (Display=10.0)

Looks like a firewall issue -- are you sure that the needed ports are
open?

BR,
Markus

On 2023-11-01, Markus Robert Kessler <no_reply@dipl-ing-kessler.de> wrote:
> On Fri, 20 Oct 2023 18:41:56 -0000 (UTC) William Unruh wrote:
>
>> On 2023-10-20, Jim <jim.beard@verizon.net> wrote:
>>> On Thu, 19 Oct 2023 20:52:51 -0000 (UTC), William Unruh wrote:
>>>
>>>> I am having trouble with nfs mounting I have machines A, which is
>>>> supposed to be an nfs server, and B a client.
>>>> On A I have /etc/exports /local
>>>> 142.103.xxx.0/26(no_subtree_check,rw,no_root_squash) where B's address
>>>> and A's address are both in that set of IP addresses.
>>>>
>>>> On B I have /etc/fstab A:/local/ /local1/ nfs4
>>>> rw,rsize=8192,wsize=8192 0 0
>>>>
>>>> The directory /local1 exists on B B:[root]>ls -la /local1 total 8
>>>> drwx------ 2 root root 4096 Jan 11 2023 .
>>>> drwxr-xr-x 32 root root 4096 Jan 11 2023 ..
>>>>
>>>> I now run mount /local1 on B and it times out. I cannot see anything
>>>> in the /var/log/messages or /var/log/syslog or dmesg on A as to why
>>>> the mount failed, and similarly on B.
>>>>
>>>> I note that nfs is now version 4.2 Could that have changed something?
>>>> Is this a bug, or more likely a bug sitting in front of the keyboard.
>>>
>>> Have you tried using mcc to go to Network Sharing, then to Configure
>>> NFS Shares, and then to Access NFS shared drives and directories or
>>> Share drives and directories using NFS, and there clicked on Search for
>>> new servers (on my machine this takes a long time), and then looked at
>>> each machine that is listed ?
>>>
>>> Looking at things this way is a PITA, especially if the settings have
>>> to be revised or created, but it eventually works, for my systems at
>>> least.
>>
>> Unfortunately I can only access A and B remotely, and mcc-->
>> NetworkSharing does not have a text version. (ie diskdrake --nfs only
>> has a gtk version and claims it cannot run in a console-- even though I
>> have X sharing (Display=10.0)
>
> Looks like a firewall issue -- are you sure that the needed ports are
> open?

Interesting possibility. But doing a tcpdump shows that the server is
receiving nfs requests from the client and responding

Here is the tcpdump on client with host server. And after this there is
no further packets seen from client or server regarding any of those
ports. I do not know what the server response is, but it seems to
something like "get lost" because of the absense of any other packets.
As I said there seems to be noting in /var/log/{messages,syslog} or
dmesg regarding this attempt.
07:09:45.339496 IP client.778 > server.nfs: Flags [.], ack 2868036885, win 501, options [nop,nop,TS val 1057430033 ecr 610219865], length 0
07:09:45.339747 IP server.nfs > client.778: Flags [.], ack 1, win 501, options [nop,nop,TS val 610281305 ecr 95893137], length 0