Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
From: "Robinson, Eric" <eric.robinson at psmnv.com> >> Okay, I see we're thinking along the same lines... But why >> would the quorum server need to be in a third data center? >> The stranded server in the secondary data center would >> release its resources, while the two servers that can still >> talk to each other in the primary data center should be fine, right? What if center 1 (with 1 machine and the quorum server) loses its connection to the outside world? Then, the machine in center 1 can't see its peer but can see the quorum server, so it takes the resources even when it shouldn't. The machine in center 2 doesn't take the resources, because it can't see the quorum server or its peer. Since center 1 has the resources, and is unreachable from the outside world, you get Unscheduled Downtime. And then everyone yells at you. A config that fragile really shouldn't be used if you can avoid it. > But if that happens and I don't mind having to intervene and > manually promote the secondary server to primary, then I don't see > the need for the third data center. Am I missing something? The idea is to allow equipment failure to happen without causing extensive downtime or requiring someone to wake up at 2am and do stuff. If you're OK with waking up at 2am and doing semi-complex things when something goes wrong, that's fine. -- Matt G / Dances With Crows The Crow202 Blog: http://crow202.org/wordpress/ There is no Darkness in Eternity/But only Light too dim for us to see