Achieving high availability in MAAS

Unlocking high-availability superpowers in MAAS

Errors or typos? Topics missing? Hard to read? Let us know!

Tap into MAAS's built-in high availability with a straightforward setup: just install extra rack controllers. The minute you do, you unleash not just highly-available BMC control but also highly-available DHCP. MAAS works like a maestro, orchestrating three critical metrics:

1. The number of racks each region supervises.
2. The connection tally of individual rack controllers.
3. The running count of regions and their worker processes.

With only a single rack in play, the system can't flex its full muscles. But add multiple racks and regions, and you witness MAAS juggling tasks seamlessly.

MAAS's load-balancing act kicks in each time a rack controller communicates with a region controller. It's always asking: is everything balanced? If the load starts to tip, MAAS adjusts. Not just between the big players—your region controllers—but also down to the micro-level, ensuring no single process gets snowed under.

And the balancing act doesn't stop there. Flip a network switch or adjust a VLAN, and MAAS re-calibrates its load-balancing strategy. What's more, you get a dynamic duo of primary and secondary rack designations, allowing for even more finely-tuned adjustments.

Power Cycle Like a Pro: BMC in High Availability

Scaling your BMC control to a highly-available setup is a no-brainer. Toss in a second rack controller, and watch MAAS spring into action. It identifies which rack controller takes charge of each BMC, keeping everything ticking along like clockwork.

DHCP High Availability Made Easy

Enable turbo-charged, highly-available DHCP by simply using MAAS-managed DHCP and adding more rack controllers to the mix. This is not just about robustness; it has a direct impact on how MAAS manages nodes through enlistment, commissioning, and deployment. You'll get a fast lane for your large networks, as all DHCP lease information gets replicated across the board.

failover peer "failover-partner" {
     primary;
     address dhcp-primary.example.com;
     peer address dhcp-secondary.example.com;
     max-response-delay 60;
     max-unacked-updates 10;
     mclt 3600;
     split 255;
     load balance max seconds 3;
}
failover peer "failover-partner" {
     secondary;
     address dhcp-secondary.example.com;
     peer address dhcp-primary.example.com;
     max-response-delay 60;
     max-unacked-updates 10;
     load balance max seconds 3;
}

Pay attention to the fact that unlike the usual 50/50 split (split 128), the primary DHCP server steps up to handle as many requests as possible (split 255). This high-octane setup only hits the pause button when it's an absolute last resort.

If you're rolling out DHCP for the first time with that extra rack controller, dig into the enabling DHCP guide. But if DHCP's already up and running, a quick reconfiguration will supercharge your existing setup.

Mastering High Availability in MAAS

Installing multiple rack controllers isn't just a set-it-and-forget-it action; it's your key to unlocking all the HA features MAAS has to offer. Keep your eye on the core metrics: rack and region counts, connection loads, and active worker processes. The system constantly refines its balancing act, making adjustments not just across controllers but also at the process level. So, whether you're dealing with network changes or routine operations, MAAS's smart algorithms keep things running smoothly.

By understanding and leveraging these features, you're not just using MAAS; you're mastering it.