[Pacemaker] pingd

[Andrew Beekhof]

On Fri, Sep 3, 2010 at 12:12 PM, Bernd Schubert
<bs_lists at aakef.fastmail.fm> wrote:
> On Friday, September 03, 2010, Lars Ellenberg wrote:
>> > > how about an fping RA ?
>> > > active=$(fping -a -i 5 -t 250 -B1 -r1 $host_list 2>/dev/null | wc -l)
>> > >
>> > > terminates in about 3 seconds for a hostlist of 100 (on the LAN, 29 of
>> > > which are alive).
>> >
>> > Happy to add if someone writes it :-)
>>
>> I thought so ;-)
>> Additional note to whomever is going to:
>>
>> With fping you can get fancy about "better connectivity",
>> you are not limited to the measure "number of nodes responding".
>
> I think for the beginning, just the basic feature should be sufficient.
> Actually I thought about to add an option to the existing ping RA to let the
> user choose between ping and fping, it would default to ping. I will do that
> mid of next week.

Could you make fping the default binary if its installed and send us
the patch when you\'re done? :-)

>
>
>> You could also use the statistics on packet loss and rtt provided on
>> stderr for -c or -C mode (example output below, chose what you think is
>> easier to parse), then do some scoring scheme on average or max packet
>> loss, rtt, or whatever else makes sense to you.
>> (If a switch starts dying, it may produce increasing packet loss first...)
>
> That will require quite parsing, which I'm not comfortable with in a shell
> script. I have no objections to later on add fping RA written in perl or
> python.
>
> [...]
>
>>
>> > >> PS: (*) As you insist ;) on quorum with n/2 + 1 nodes, we use ping as
>> > >> replacement. We simply cannot fulfill n/2 + 1, as controller failure
>> > >> takes down 50% of the systems (virtual machines) and the systems
>> > >> (VMs) of the 2nd controller are then supposed to take over failed
>> > >> services. I see that n/2 + 1 is optimal and also required for a few
>> > >> nodes. But if you have a larger set of system (e.g. minimum 6 with
>> > >> the VM systems I have in my mind) n/2 + 1 is sufficient, IMHO.
>> > >
>> > > You meant to say you consider == n/2 sufficient, instead of > n/2 ?
>>
>> So you have a two node virtualization stuff, each hosting n/2 VMs,
>> and do the pacemaker clustering between those VMs?
>
> Yes.
>
>>
>> I'm sure you could easily add "somewhere else" a very bare bone VM
>> (or real) server, that is dedicated member of your cluster, but
>> never takes any resources? Just serves as arbitrator? as your "+1"?
>
> No, I'm afraid it is not that easy. There is simply is nothing that can be
> used. If there is anything, it is always available on both hosts/controllers.
> Imagine you would sell a standalone DRBD system (black box), that provides for
> example NFS to clients. You would want to have each and every additional
> service mirrored again. And you could not rely on additional customer NFS
> clients.
>
>>
>> May be easier, safer, and more transparent than
>> no-quorum=ignore plus some ping attribute based auto-shutdown.
>
> I agree on safer and transparent, but unfortunately, it not easier in our
> case.
>
> --
> Bernd Schubert
> DataDirect Networks
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
>