Burned by Eschewing Best Practices
-
This is my personal favorite - https://community.spiceworks.com/topic/232181-disaster-recovery-on-the-fly-is-a-disaster-waiting-to-happen?page=1. I'm sure @scottalanmiller remembers this one.
-
@NetworkNerd I try to forget.
-
@scottalanmiller said in Burned by Eschewing Best Practices:
Yet another... didn't virtualize but it would have protected you. This is another one of those "unknowns" that sure, you would never guess that you would need virtualization for this reason, but it would have been all that was needed to make this a non-issue.
This topic here doesn't seem so bad, unless he started running production on it and then realized that the system would begin to have activation issues.
Simple solution, reinstall the proper OS to the hardware, but even better, virtualize the system, and then create VMs onto it.
-
@DustinB3403 said in Burned by Eschewing Best Practices:
@scottalanmiller said in Burned by Eschewing Best Practices:
Yet another... didn't virtualize but it would have protected you. This is another one of those "unknowns" that sure, you would never guess that you would need virtualization for this reason, but it would have been all that was needed to make this a non-issue.
This topic here doesn't seem so bad, unless he started running production on it and then realized that the system would begin to have activation issues.
Simple solution, reinstall the proper OS to the hardware, but even better, virtualize the system, and then create VMs onto it.
It's only so bad... but it shouldn't have been bad at all. It was built locally and shipped out. So now it needs to be shipped back. Expensive and time consuming. All because he didn't virtualize the first time.
-
Oh I don't disagree, but this was purely time expensive. I didn't see that he purchased the wrong license or anything.
Just that he made a mistake with the setup of the system. Virtualizing it, he would still have to dump the VM and rebuild.
Which is a quite a bit easier to do rather than having to reinstall from scratch on the hardware.
-
@DustinB3403 said in Burned by Eschewing Best Practices:
Oh I don't disagree, but this was purely time expensive. I didn't see that he purchased the wrong license or anything.
No, he was simply burned by eschewing best practices. He skipped the best practice, and now it is costing time and money for no reason. That's the point.
-
@scottalanmiller said in Burned by Eschewing Best Practices:
@DustinB3403 said in Burned by Eschewing Best Practices:
Oh I don't disagree, but this was purely time expensive. I didn't see that he purchased the wrong license or anything.
No, he was simply burned by eschewing best practices. He skipped the best practice, and now it is costing time and money for no reason. That's the point.
haha good point... I was thinking to far into it.
-
@DustinB3403 said in Burned by Eschewing Best Practices:
Just that he made a mistake with the setup of the system. Virtualizing it, he would still have to dump the VM and rebuild.
No, the BIG mistake was not virtualizing and protecting against other mistakes. The build time is minutes and is trivial. I did ten manual builds while watching a television show two nights ago. That's nothing. It's having to have a remote site derack a server, box it, ship it, receive it, rebuild it, box it, ship it, rack it, etc.
Installing the wrong OS is trivial. People make mistakes. The failure here was not protecting against mistakes by skipping the best practice that makes sure that those little mistakes don't turn into big problems.
Shipping and racking servers and in person physical installs are all very time consuming and expensive items.
-
That's the thing about virtualization... it protects against the unknowns. The changes, the mistakes the... whatever. People always say "I can't see how it would help me". ANd that's the point, that they can't see why it would help is why they need it.
-
Oh wow, when you mentioned deracking it I was lost for a second. I didn't realize this system was 1500 miles away.
OK yeah... I totally agree.
-
@scottalanmiller said in Burned by Eschewing Best Practices:
That's the thing about virtualization... it protects against the unknowns. The changes, the mistakes the... whatever. People always say "I can't see how it would help me". ANd that's the point, that they can't see why it would help is why they need it.
Before I really got into virtualization, I just didn't know about it. Now I love, love, love it.
-
I guess the simple question is, what cost more an Upgraded license(to install hyper-v to the system), or the time and expense of deracking it, and shipping it back just to turn it back around after installing the proper OS.
-
@DustinB3403 said in Burned by Eschewing Best Practices:
I guess the simple question is, what cost more an Upgraded license(to install hyper-v to the system), or the time and expense of deracking it, and shipping it back just to turn it back around after installing the proper OS.
Yeah, I can't believe that that doesn't justify the license upgrade. $800 max, probably a lot less. Shipping a server is not cheap.
-
I would think having to pay someone to pull it out of the rack, package and pay for return shipping, and then to re-return ship it would be way more expensive.
-
@DustinB3403 said in Burned by Eschewing Best Practices:
I would think having to pay someone to pull it out of the rack, package and pay for return shipping, and then to re-return ship it would be way more expensive.
Not likely more, but it's all a waste whereas the update is at least a fresh license.
-
@scottalanmiller but it's also a hard cost (measured in a PO to outright purchase the license) where as someone's time might just be a "Salary tasks for today".
When I worked for a local sheet metal facility we had our own site to site truck driver which we regularly shipped material (daily between facilities) to and from this facility. So it could be near free for the business.
But in terms of the system not being accessible for however long it takes to address this issue. That I see as being a big cost. He also probably doesn't want to approach the management team and claim he made a mistake.
-
@DustinB3403 said in Burned by Eschewing Best Practices:
But in terms of the system not being accessible for however long it takes to address this issue. That I see as being a big cost. He also probably doesn't want to approach the management team and claim he made a mistake.
Another violation of best practice, if that is the case. Trying to hide this will just cost more money.
-
@scottalanmiller said in Burned by Eschewing Best Practices:
@DustinB3403 said in Burned by Eschewing Best Practices:
But in terms of the system not being accessible for however long it takes to address this issue. That I see as being a big cost. He also probably doesn't want to approach the management team and claim he made a mistake.
Another violation of best practice, if that is the case. Trying to hide this will just cost more money.
Of course, best practice being "I made a mistake, we can do x or z to fix it."
Rather than it coming back around in 180 days (or when someone notices)
-
Just poking fun at @jospoortvliet for not using a VM for this.
-
The thing @scottalanmiller and @DustinB3403 are missing on this last one is that the OP on the SW thread has a 100% valid hypervisor in place now.
There is zero lack of virtualization if the OP so chooses.
The OP could very simply reallocate one of their Server 2012 R2 DC licenses to that box and be perfectly licensed for anything he needs.
Now that is quite expensive if that remote office will not be running the 8+ Server 2012 VMs needed to make it cost effective.
So shipping it back and redoing it can easily be the most cost effective.
At that point yes, the best thing would be to use Hyper-V Server 2012 R2 on the bare metal, but do not miss the forest for the trees. Server 2012 R2 (DC or Standard) is a perfectly viable hypervisor itself once the role is installed.
