paparazzi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] Redundant System


From: Chris Gough
Subject: Re: [Paparazzi-devel] Redundant System
Date: Wed, 12 Aug 2009 08:45:56 +1000

I've seen a HA cluster go down when some networking gear failed.
Storage could also be single points of failure shared semaphore
fail-over design. I used to look after some HA servers with a
'watchdog' gadget that was also a potential single point of failure.

I'm hazy on the details, but I think I remember Samba (SMB file
server) having an 'election' protocol where if the primary domain
controller disappeared from the network the other samba servers would
quickly decide which one should take over. Basically each samba server
was also a watchdog for the whole cluster. With a zigbee mesh network,
might there be a pure software solution to multiple paparazzi
autopilot fail-over (assuming servos on rails)?

Chris Gough

On Wed, Aug 12, 2009 at 7:57 AM, David<address@hidden> wrote:
> Hello,
>
> I work daily with HA products used in business transactions (Oracle, Tibco).
> they get by with shared storage and state/heartbeat files etc...with two as
> a minimum.
>
> If both autopilots had a process that both could access the same shared file
> it's possible to reliably determine if the other is "alive"...the failover
> one could take over should the primary go down.
>
> So most apps I am familiar with require two and a shared file minimum so I
> am pretty sure it can be done with two.
>
> - David Conger
>
> On Aug 11, 2009, at 11:22 AM, antoine drouin <address@hidden> wrote:
>
>> Hello World
>>
>> If you're going to make a redundant system, you need at least 3
>> systems in order to be able to decide if one failed.
>> You'll also need a way to switch actuators and other peripherals to
>> the "good" instance of the autopilot.
>>
>> Regards
>>
>> Antoine
>>
>> On Tue, Aug 11, 2009 at 10:05 AM, Rui Costa<address@hidden> wrote:
>>>
>>> Hello all,
>>>
>>> It would be very nice to implement a redundant system control for the
>>> paparazzi project.
>>> A system with for example two TWOGs. In case of a twog failure the other
>>> one
>>> assume the control.
>>>
>>> What do you think about it? It's difficult to design?
>>>
>>> Best Regards
>>> Rui Costa
>>> www.azoreanuav.com
>>>
>>> _______________________________________________
>>> Paparazzi-devel mailing list
>>> address@hidden
>>> http://lists.nongnu.org/mailman/listinfo/paparazzi-devel
>>>
>>>
>>
>>
>> _______________________________________________
>> Paparazzi-devel mailing list
>> address@hidden
>> http://lists.nongnu.org/mailman/listinfo/paparazzi-devel
>
>
> _______________________________________________
> Paparazzi-devel mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/paparazzi-devel
>




reply via email to

[Prev in Thread] Current Thread [Next in Thread]