Hey @stefano have you ever built a clustered environment with FreeBSD?
=> More informations about this toot | More toots from blake@bsd.cafe
@blake hi! Which kind of cluster do you mean? Generally speaking, I perform HA at app level
=> More informations about this toot | More toots from stefano@bsd.cafe
@stefano HA at the app level - I know it won't be anywhere near what Proxmox does with Ceph Storage or it's HA clustering (or if it is I need to find some articles to read on)
Building those two new 1U systems it would be nice to be able to shutdown one system and keep most of my applications running on the other.
=> More informations about this toot | More toots from blake@bsd.cafe
@blake Ceph should also be working on FreeBSD. I tried it years ago, and it was running fine. As for the 'Proxmox-style' HA, we're not there yet. However, I found that the trade-offs and the increase in complexity often lead to more problems than the ones they solve.
=> More informations about this toot | More toots from stefano@bsd.cafe
@blake you can still achieve it with ceph or ZFS send/receive. I don't know if the live migration is now stable/considered production ready, but keeping workloads simple and separated means that a reboot is just a few seconds of outrage (stopping the VM in one node and starting into the other)
=> More informations about this toot | More toots from stefano@bsd.cafe
@stefano Very true, it'll just be a lot of tinkering to figure out what I want to be running 24/7/365.
For example I know I'd want my Wazuh/SIEM setup running 24/7/365 so I'd have to figure out how I'd want it setup. But that is way down the line as I have just the one R720 for the time being.
=> More informations about this toot | More toots from blake@bsd.cafe
@Blake :tux: :freebsd:
Check out HAST. https://docs.freebsd.org/en/books/handbook/disks/#disks-hast
=> More informations about this toot | More toots from knt@streams.knthost.com This content has been proxied by September (3851b).Proxy Information
text/gemini