[Linux-cluster] GFS on 3-node cluster corrupted after full network outage

[Klaas van der Lugt]

Hi David;

With "no fencing" of course I mean that we configured
fencing, but fencing is simply not initiated by the three remaining
nodes since they all become inquorate.

So, my question is still valid: Am I right that in the
situation of a 3-node cluster using GFS, upon full network
outage, will the inquorate nodes keep on running GFS?

What I'm actually referring to is the explanation in
the final answer on the following thread:
http://www.webservertalk.com/archive391-2006-3-1430956.html

Regards,
Klaas





> On 12/7/06, Klaas <klaas at klaas.nl> wrote:
>> Hi;
>>
>> We have Cluster software running because we need GFS.
>>
>> Lately, we had a 100% network outage and all three
>> nodes kept on running (of course inquorate)
>> and the GFS stayed available too.
>>
>> No fencing (I understand this is documented behaviour) but
>> this way our GFS will get corrupted if the rest of the
>> software on the nodes keeps accessing the GFS...
>>
>> Please advise,
>> Klaas
>>
>
> http://sources.redhat.com/cluster/faq.html#fence_what
>
>>From the FAQ:
>
> #  Can't I just use my own watchdog or manual fencing?
>
> No. Fencing is absolutely required in all production environments.
> That's right. We do not support people using only watchdog timers
> anymore.
>
> Manual fencing is absolutely not supported in any production
> environment, ever, under any circumstances.
>
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> https://www.redhat.com/mailman/listinfo/linux-cluster
>