@dyasny said in New Infrastructure to Replace Scale Cluster:
@Dashrender said in New Infrastructure to Replace Scale Cluster:
In general "buying" HA is generally though of as just by VMWare with the HA feature and tada you have HA. which of course is wrong. If you don't have redundant power and redundant switches, and storage, and HVAC, etc, etc - then you don't have real HA - you have one tiny piece that's HA, but you don't have HA for the the likely end goal.
Everything you mentioned is something you can buy, and is usually specified as a prerequisite for an HA solution.
At least this is what I take from Scott's comments.
VMWare won't sell you the solution that includes all the switches and internet and power and HVAC, etc, etc.. they will only sell you the things they sell. But HA requires so much more than they can provide.
I frankly don't remember the last time I actually bought anything from a vendor. People usually go to an integrator, and pay for building a solution for them. That integrator should provide all the prerequisites, and you pay for a solution not a single standalone product.
Everyone wants to sell solutions, not products, to the point where these solutions in turn become productized. And an HA solution can be bought.
Now, if we stop the pissing contest (yes Scott, you are the god of IT, you are always right and everyone else is always wrong, even when you don't drop your 2c in every conversation) and think rationally for a second here, we are talking about a small setup with just a few hosts. We already established we have brand name hardware, which means IPMI can be used as SBA. Out of the box with oVirt, this provides us with the ability to make VMs highly available, just configure the hosts, and mark the VMs as HA, nothing else. Assuming networking was done properly, and we have switch redundancy, and hoping the building has a generator and the storage is also reliable is nice, but out of scope for this particular scenario. All the OP wanted was VM HA, where if a VM dies or the host carrying it stops being able to host it, it gets safely started elsewhere. It isn't hard to grasp the concept, really.
Started? That's not HA. At least at this very moment I wouldn't consider it HA, I'd consider it SA (Standard Availability).
As for your reasoning that people buy solutions - oh, if only that were true more often than not. But one look at Spice--- oh you know that place, you can see that people buy a SAN and think they have HA. period.
Also to more points you made - You simply can't have HA without having ALL of those other parts. It's great that you have HA at the server level, but what are the chances that's where your issue is going to be and not at the electrical power level? It's great that the servers have HA, but if your internet doesn't is the solution really HA?
No - it's not.
It's sudo HA. or more likely - simply SA or in the worst setup - LA.