supportman Posted March 5, 2019 Posted March 5, 2019 Is your local storage shared or is there no requirement for fault tolerance? what you suggest would definitely be my prefered method though I worry about a server falling over with the all the locally attached storage with it and the vms not being able to migrate across. No fault torrerance, SSDs are pretty stable if you buy quality ones. We backup with veeam anyway, so restore is very quick if the worst happened.
pablo007 Posted March 5, 2019 Author Posted March 5, 2019 Ok thanks, that makes sense unless of course, the mobo dies and any VMS on that server are also dead until its fixed or restored from backup onto alternative hardware but this rarely occurs in my experience.
mavhc Posted March 5, 2019 Posted March 5, 2019 https://github.com/ewwhite/zfs-ha/wiki build your own, no single point of failure
mavhc Posted March 5, 2019 Posted March 5, 2019 Ok thanks, that makes sense unless of course, the mobo dies and any VMS on that server are also dead until its fixed or restored from backup onto alternative hardware but this rarely occurs in my experience. Just pull the drives and put them in another server. If the VMs were critical you'd be replicating them anyway
supportman Posted March 5, 2019 Posted March 5, 2019 Ok thanks, that makes sense unless of course, the mobo dies and any VMS on that server are also dead until its fixed or restored from backup onto alternative hardware but this rarely occurs in my experience. We run a spare Hyper-V host for an emergency. We use Intel NUC's here in the server cabinet with Windows Server 2019 on them. Virtualisation makes redundancy and disaster recovery so easy these days, it really is a non issue if you design it simply.
DrCheese Posted March 5, 2019 Posted March 5, 2019 (edited) I've seen others just do Replica's for critical stuff - If a host dies you can switch to the replica server until it's repaired, tho it's not an automatic process. (You can script it but risk split brain) As above, avoid storage spaces direct unless you buy a supported platform (Yes you can do it yourself but you shouldn't - You *will* run into issues) Edited March 5, 2019 by DrCheese
markwilfan Posted March 5, 2019 Posted March 5, 2019 +1 for shared nothing hyper-v config. We have 2 main hypervisors and a backup less powerful one. Super stable and really quick with local raid 10 storage . OS running on ssds in raid 1
Orangescott Posted March 12, 2019 Posted March 12, 2019 As someone who's been in the selling game (now on hiatus) I'd recommend speaking to a chap called Dan Douglas at Total Computers [email protected] One of the best professionals I've worked with he's honest, good technical knowledge and great vendor relationships. 1
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now