[ClusterLabs] Antw: [EXT] Re: DRBD split‑brain investigations, automatic fixes and manual intervention...

[Ulrich Windl]

 >>> Ian Diddams via Users <users at clusterlabs.org> schrieb am 21.10.2021 um 09:28 in
Nachricht <1152166797.7562713.1634801283402 at mail.yahoo.com>:

> 
>     On Wednesday, 20 October 2021, 18:08:50 BST, Andrei Borzenkov 
> <arvidjaar at gmail.com> wrote: 
> It depends on what hardware you have. For physical systems IPMI may be
> available or managed power outlets; both allow cutting power to another
> node over LAN.
> 
> For virtual machines you may use fencing agent that contacts hypervisor
> or management instance (like vCenter).
> 
> There is also SBD, it requires shared storage. Third node with iSCSI
> target may do it.
> 
> Qdevice with watchdog may be an option.
> 
> Personally I prefer SBD which is the most hardware-agnostic solution.
> ========
> 
> Thanks again.  As I feared these are whole new areas that Ive never even 
> heard of so this is going to be a long road.

Well, my personal opinion is this: If I could decide whether to spend my time reading _before_ implementing a DRBD based database, or spending my time recovering after data corruption or data loss, I'd perfer the first one. Osually the time pressure for the second one is much harder.

Don't get me wrong: Even when reading before and implementing after that, there will be several iterations of reading, reconfiguring and testing to get a stable configuration.

"HA" is not a product to install, but a way to do things (starting at hardware selection, cabling, etc.)...

Regards,
Ulrich


> 
> In the emantime while I try and find out what does what and how and how ti 
> implement it if anybody (based on previous info preovided) has a list of "do 
> tyhis
> 
> * run A
> * run B
> * run C"
> 
> to get this implementation covered Id be greateful
> cheers
> ian
>