[Samba] CTDB and NFS4

[Stephen Ward]

Hi everyone,

I'm sorry if this has already been covered somewhere, but I've had a look
and can't find it. Also if there is a better place for this question,
please let me know.

I know that the CTDB documentation states that only NFS2/3 are supported
with CTDB but what I'd really like to know is why NFS4 won't work and if
NFS4 support is planned for the future (timeline?). Also can anyone
suggest any work arounds/development versions/anything that may help with
this?

I have a cluster that works well with samba4 and NFS3 (on GPFS/RHEL) but
unsurprisingly, while NFS4 works ok normally, if I stop the server a client
is using any access attempts will freeze on the client.

Regards
Stephen
--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/options/samba

[Volker Lendecke]

On Fri, Mar 07, 2014 at 12:32:01PM +1300, Stephen Ward wrote:
> I'm sorry if this has already been covered somewhere, but I've had a look
> and can't find it. Also if there is a better place for this question,
> please let me know.
>
> I know that the CTDB documentation states that only NFS2/3 are supported
> with CTDB but what I'd really like to know is why NFS4 won't work and if
> NFS4 support is planned for the future (timeline?). Also can anyone
> suggest any work arounds/development versions/anything that may help with
> this?
>
> I have a cluster that works well with samba4 and NFS3 (on GPFS/RHEL) but
> unsurprisingly, while NFS4 works ok normally, if I stop the server a client
> is using any access attempts will freeze on the client.

Is that really a NFS4/CTDB question? With hard mounts that's
pretty much what I would expect.

Volker

--
SerNet GmbH, Bahnhofsallee 1b, 37081 Göttingen
phone: +49-551-370000-0, fax: +49-551-370000-9
AG Göttingen, HRB 2816, GF: Dr. Johannes Loxen
http://www.sernet.de, mailto:kon...@sernet.de

[Stephen Ward]

Hi Volker, thank you for your comments.

Using CTDB with an NFS server if I mount with the version 3 option then
stop ctdb on the server the client fails over to the new server fine. Same
client, same servers, same process but mounting with the version 4 option
and when CTDB fails over the client locks. After ten minutes or so I
killed the process on the client, it was still not responsive.

As I understand it this works because the NFS daemons share records and
'tickle' connections on failover (see
https://wiki.samba.org/index.php/CTDB_Setup#Setting_up_CTDB_for_clustered_NFS
).

NFS4 works fine except for failover and NFS3 works fine including failover
so I assume that it is specifically an issue with NFS4 with CTDB.

My understanding of specifically how this works is limited so if anyone has
any ideas about this at all I'd be really keen to hear!


On Tue, Mar 11, 2014 at 8:30 PM, Volker Lendecke
<Volker....@sernet.de>wrote:

[L.P.H. van Belle]

only thing in know for NFSv4
add in krb5.conf in
[libdefaults]
allow_weak_crypto = true

then nfs4v wil work. ( tested in a non CTDB setup )





>-----Oorspronkelijk bericht-----
>Van: sw_s...@nzgeek.com
>[mailto:samba-...@lists.samba.org] Namens Stephen Ward
>Verzonden: dinsdag 11 maart 2014 9:09
>Aan: Volker....@sernet.de
>CC: sa...@lists.samba.org
>Onderwerp: Re: [Samba] CTDB and NFS4

[Stephen Ward]

Hi L.P.H van Belle,

Thank you for your suggestion, as NFS4 is already working and it is just
the clustering setup that I am having issues with I'm not sure if this
would still help?

[steve]

On Tue, 2014-03-11 at 21:47 +1300, Stephen Ward wrote:
> Hi L.P.H van Belle,
>
> Thank you for your suggestion, as NFS4 is already working and it is just
> the clustering setup that I am having issues with I'm not sure if this
> would still help?

Hi
We have nfs4 working against the Samba4 KDC with arcfour keys. They're
not weak I don't think. Is there anything you need with nfs4 that nfs3
doesn't provide? We prefer the latter due to the ease with which you can
set up acls. Does the user with the hang have any unusual acls set?
Cheers

[Volker Lendecke]

On Tue, Mar 11, 2014 at 09:09:15PM +1300, Stephen Ward wrote:
> Hi Volker, thank you for your comments.
>
> Using CTDB with an NFS server if I mount with the version 3 option then
> stop ctdb on the server the client fails over to the new server fine. Same
> client, same servers, same process but mounting with the version 4 option
> and when CTDB fails over the client locks. After ten minutes or so I
> killed the process on the client, it was still not responsive.
>
> As I understand it this works because the NFS daemons share records and
> 'tickle' connections on failover (see
> https://wiki.samba.org/index.php/CTDB_Setup#Setting_up_CTDB_for_clustered_NFS
> ).
>
> NFS4 works fine except for failover and NFS3 works fine including failover
> so I assume that it is specifically an issue with NFS4 with CTDB.
>
> My understanding of specifically how this works is limited so if anyone has
> any ideas about this at all I'd be really keen to hear!

Ok. Understood. The problem here is that we (Samba/ctdb
team) have no real connection to the NFS4 people except for
some random chats with the Ganesha team. But I guess you're
not using Ganesha anyway. NFS4 is probably a lot more
difficult to correctly fail over in a cluster case, because
it is stateful and does have a notion of open files. So I'm
not sure how the clients in your setup are supposed to work,
and how Samba/ctdb could improve things.

Do you have an idea how this is supposed to work in a
Pacemaker failover scenario, no ctdb and Samba involved,
just NFS4?

Volker

--
SerNet GmbH, Bahnhofsallee 1b, 37081 Göttingen
phone: +49-551-370000-0, fax: +49-551-370000-9
AG Göttingen, HRB 2816, GF: Dr. Johannes Loxen
http://www.sernet.de, mailto:kon...@sernet.de

[J. Echter]

Am 11.03.2014 09:47, schrieb Stephen Ward:
> Hi L.P.H van Belle,
>
> Thank you for your suggestion, as NFS4 is already working and it is just
> the clustering setup that I am having issues with I'm not sure if this
> would still help?
>
>

> On Tue, Mar 11, 2014 at 9:30 PM, L.P.H. van Belle <be...@bazuin.nl> wrote:
>
>> only thing in know for NFSv4
>> add in krb5.conf in
>> [libdefaults]
>> allow_weak_crypto = true
>>
>> then nfs4v wil work. ( tested in a non CTDB setup )
>>
>>
>>
>>
>>
>>> -----Oorspronkelijk bericht-----
>>> Van: sw_s...@nzgeek.com
>>> [mailto:samba-...@lists.samba.org] Namens Stephen Ward
>>> Verzonden: dinsdag 11 maart 2014 9:09
>>> Aan: Volker....@sernet.de
>>> CC: sa...@lists.samba.org
>>> Onderwerp: Re: [Samba] CTDB and NFS4
>>>

>>> Hi Volker, thank you for your comments.
>>>
>>> Using CTDB with an NFS server if I mount with the version 3 option then
>>> stop ctdb on the server the client fails over to the new
>>> server fine. Same
>>> client, same servers, same process but mounting with the
>>> version 4 option
>>> and when CTDB fails over the client locks. After ten minutes or so I
>>> killed the process on the client, it was still not responsive.
>>>
>>> As I understand it this works because the NFS daemons share records and
>>> 'tickle' connections on failover (see
>>> https://wiki.samba.org/index.php/CTDB_Setup#Setting_up_CTDB_for
>>> _clustered_NFS
>>> ).
>>>
>>> NFS4 works fine except for failover and NFS3 works fine
>>> including failover
>>> so I assume that it is specifically an issue with NFS4 with CTDB.
>>>
>>> My understanding of specifically how this works is limited so
>>> if anyone has
>>> any ideas about this at all I'd be really keen to hear!
>>>
>>>

>>>> Volker
>>>>
>>>> --
>>>> SerNet GmbH, Bahnhofsallee 1b, 37081 Göttingen
>>>> phone: +49-551-370000-0, fax: +49-551-370000-9
>>>> AG Göttingen, HRB 2816, GF: Dr. Johannes Loxen
>>>> http://www.sernet.de, mailto:kon...@sernet.de
>>>>
>>> --
>>> To unsubscribe from this list go to the following URL and read the
>>> instructions: https://lists.samba.org/mailman/options/samba
>>>
>>>
>>

Hi,

we use pacemaker / corosync for this tasks.

If one node fails fire up the services on the failover node.

No samba clustering involved.

Cheers

Juergen