The Textbook Things Gone Wrong in IT Thread
-
@DustinB3403 said:
Granted I think a second unit is reasonable for the cost of $10G its my boss who has to sell it.
but if a second unit is worth it, why would they even talk about less available solutions as the only other alternative? This makes no sense.
-
I've never talked about a less available solution, a minimum solution sure.
I've always wanted HA in this environment (because I'm tired of getting that 1AM call).
And I think it's a reasonable cost, doesn't mean that upper management does. They might say "eh they can wait". In my mind though it is not something that should simply thrown off, especially since we're growing our overseas clients are asking us for more and more.
-
I'm not saying that HA is bad, I'm saying that the shop - both the existing management and the MSP - have decided that HA is the last thing that they need and is doing the exact opposite.
Introducing this approach improves the reliability, improves the performance, lowers the cost, lowers the maintenance.
The last thing we should do is jump to "oh, now that we've improved things over what is required, maybe we need to make an order of magnitude leap farther." There is no reason for that.
-
@DustinB3403 said:
I've never talked about a less available solution, a minimum solution sure.
The SAN approach is less. The NAS approach is less. All less. This is proposed a a guaranteed improvement because it improves EVERY facet over anything currently suggested.
-
Ah well yes.
That is true that it would be a huge step to jump to HA, but it should would be nice
-
@DustinB3403 said:
I've always wanted HA in this environment (because I'm tired of getting that 1AM call).
Every IT person wants HA, that's a given. But the business is only looking at LA currently and is happy with that level of reliability.
-
That or the business (uppers) don't understand what HA offers at a "reasonable cost" even if not immediately purchased.
And Scott I did say:
@DustinB3403 said:
And down the road (6 months maybe) purchase a second matching unit for HA which we really want.
-
@DustinB3403 This is the way I think as well. If having a failover server seems feasible to the budget (and/or the bean counters), and the boss is sold on the idea as well, then it should be done.
Is it always cost effecient? No, not always.
Does it help the IT Department keep things running in the event of a major server outage? Yes.
Is it a replacement for good backups? Definitely not!
-
Again its a want, there isn't a solid need for HA.
And for the cost of it I'd rather have it as a raise to my check
But if it helps the company be more productive and healthy then go for it.
-
This entire conversation needs split into a new thread..
@DustinB3403 your answer here is to build the $4000 server as specified with the SSD.
Then Install Hyper-V or XenServer on it
Then take your 3 physical hosts and P2V them into three virtual machines.
No remapping, no changing of anything except the old hardware goes poof.Then Backup the three virtual machines to your Buffalo TeraStation.
End of f'n project.
-
@dafyre said:
@DustinB3403 This is the way I think as well. If having a failover server seems feasible to the budget (and/or the bean counters), and the boss is sold on the idea as well, then it should be done.
It's not from what he has described, they are doing exactly the opposite. The idea that HA would be entertained at all seems very far fetched. In one case we are taking what they have today and improving it in every way compared to their current situation and any proposed situation. In the other you are trying to double their costs and do something the very opposite of what their actions suggest that they value.
-
@scottalanmiller I was referring to his desire to build a backup / redundant / failover server, just in case of a fire, earthquake, tornado, flood, and/or other acts of God... and also only if it is reasonably done money wise.
But as he said... It is a want, and he'd rather have that money put on his paycheck... to this, I would agree.
But like you'v said in other threads... It might be worth doing a cost analysis to figure out whether or not it would be better for them to be down for a day, or have another server to just fail over to in the event of a dead server.
-
@dafyre said:
@scottalanmiller I was referring to his desire to build a backup / redundant / failover server, just in case of a fire, earthquake, tornado, flood, and/or other acts of God... and also only if it is reasonably done money wise.
Yes, but it isn't a reasonable thing given the context for the business in any way. Nor does it protect against those things, that would be a DR site, not a second box.
-
It's amazing where the prices for something like this is today. I built a similar setup just 20 months ago, but had a stupid HDD requirement to fullfil and each server was $50K, you're totally kicking my ass and it's only costing you $4k.
Nice!
-
Yeah well apparently the MSP is even kicking my butt at the price each server is $2000 (granted I have no idea what is inside of them)
-
Oh I know what it doesn't include the cost of the SAN...
haha there's the other $20,000
-
If the $2k doesn't include drives, who cares? that's what this whole thing is about.
-
@DustinB3403 said:
Yeah well apparently the MSP is even kicking my butt at the price each server is $2000 (granted I have no idea what is inside of them)
We could do it cheap if we didn't care about quality or the company at all. Getting the price down is easy.
-
Without local drives we could probably come in way lower.
-
Yeah I know....
Pennywise pound foolish..