We're in the process of migrating our DHCP servers, and we've always faced issues with DHCP before—mainly related to our DNS records going haywire. I'm curious if anyone here has experience with setting up a hot-standby failover? Did it work out for you? We're considering disabling DNS scraping before we proceed with the migration.
8 Answers
A lot of DNS issues stem from caching problems. Each DNS record has a TTL which can complicate things, especially if your MINIMUM TTL is set high. When negative results get cached, they stay around longer. If you control how clients connect to the DNS servers, you can manage these issues better. Knowing how to use tools like nslookup or dig is key!
Absolutely, learning how to back up your database and services is crucial to any successful migration!
I’ve implemented a hot standby before. It worked well in tests, but I haven’t deployed it in production yet. Just remember to adjust your IP helper settings on switches for smooth operation.
Can you explain why you went for hot standby instead of an active/active setup?
The best approach I found is to export your settings, import them to the new server, set your new DHCP scopes as active, and inactive for the old ones. Then just adjust your Helper IPs, and you should be good to go!
Just a heads up, if you make any changes to the scopes or options afterward, you might have to sync them manually.
This is definitely the way to go.
I haven't done hot standby, but I have tried active/active setups. It's been smooth for us—operating at a 50/50 split on each subnet. The primary responds immediately, while the secondary has about a 2-second delay. I wouldn't change this setup; it's really solid!
I loved the idea of active/active, but management opted for hot standby due to concerns about IP usage. Is that something you've noticed as well?
This is how I run my setup, super reliable!
Setting up load-balanced DHCP has been a breeze since Windows 2016! Most of the work is just ensuring your routers have the forwarders set up right. Just to be clear, this isn’t the same as Split Scope or Failover, it's Active/Active storage!
Proactively informing the other server that one is offline can be beneficial rather than just waiting for it to time out. It’s also crucial to test your strategy before going live; backups are essential!
Good to know! I’ve already confirmed that with the network team, thanks!