Virtualization Redemption?
-
So I want to backup a moment - @scottalanmiller
You think it would generally be more important to have two local servers with something like Starwind than having a replicated off site system? If you can only pick on or the other, which would you pick? -
@Dashrender said:
You think it would generally be more important to have two local servers with something like Starwind than having a replicated off site system? If you can only pick on or the other, which would you pick?
90% of the time I'd say two on site. But as Hub points out, the situations are not always uniform. His primary cause of failure is loss of site, not loss of server. A server dies, roughly, once every eight years (very rough.) His concern is loss of site every two to three years, I think. But that is very rare overall as a risk case.
-
We have tornado risk here too, but I think his is a lot higher than mine.
-
@Dashrender said:
We have tornado risk here too, but I think his is a lot higher than mine.
Everyone has tornado risk in the US. But even if you lose as site completely, let's say meteor strike, not everyone has a need for a DR site.
Example: you are a spa, you only have one physical site. Is a DR site for your computers valuable if the business itself cannot keep running? Probably not.
-
I only underbought when trying to run all workloads on a single server. The role of the new DR server wasn't always DR. Our other servers are perfectly happy with extra space
Not going the starwind route, doesn't make sense for me as the DR site will in essence be providing me with the same (technically moreish) protection from failure. both site, and vm, or physical host failure on main site. if i wanted to buy 6 more drives, i could starwind it out and have uber ultra super safe mega secure situation.... just dont think they want to spend that right now. maybe a next year project expansion
-
However, he'd still get the desired functionality out of his DR site if SERVER2 just dies / slings a rod / explodes or dies some othre violent death...
Spin up the replicated SERVER2 at the DR site and be happy.
If he were using StarWind, it could potentially be set up to automatically migrate to the DR site?
-
@dafyre said:
If he were using StarWind, it could potentially be set up to automatically migrate to the DR site?
Not sure if SW would automate the DR site. It would automate the local failover for sure. Actually SW does nothing there, it just keeps the storage humming. HyperV HA is what handles the failover piece. StarWind just keeps the storage from having failed at all.
-
Is not doing SW. Does not matter.
-
@hubtechagain said:
Is not doing SW. Does not matter.
No longer needed, forum now has topic forking and the information that was requested was forked to a new topic. The above was in reference to posts that are no longer here.
-
StarWind isn't possible in this setup because of the lack of enough local storage on the two servers in the same location.
-
@Dashrender wut
Could you please explain that? I mean, maybe Im missing something, but I dont see why the usable capacity is the stopper here, even if we can reconfigure RAID. -
@Dashrender said:
@KOOLER said:
@hubtechagain said:
WTF is starwind?! ha
This is who we are
https://www.starwindsoftware.com/starwind-virtual-san-free
We give away free version to use on a bare metal servers (so you take a pair of them and turn into HA NFS or SMB3 NAS). This one has no restrictions at all (capacity is unlimited, production use is OK and you can be anybody to get it).
HA iSCSI and hyper converged versions are available to different set of people like MVPs, SpiceHeads, VCPs, some restricted ones to MCTs & bloggers. Technically we can bring same program to MangoLassi community as well. I just need some sort of a low watermark (points, rank or whatever) to make the program look a bit of private so my VP of Sales would not burn me with a blow torch
Cheers and let me know if you'd have any questions
That is pretty cool. Most of us here are Spiceheads as well, so we're probably covered, though getting ML on the list would be awesome!!!
It's still a smaller community here as with a set of drawbacks it definitely has own benefits: much easier to have "special" handling
P.S. You know bigger you become more bureaucratic processes start to happen to slow the things down and complicate everything...
-
@Dashrender said:
StarWind isn't possible in this setup because of the lack of enough local storage on the two servers in the same location.
You can go virtual on top of an exiting hypervisor nodes. There's a way to obtain free license for hyper converged setup if you plan to support and maintain everything on your own. FYI.
-
@original_anvil said:
@Dashrender wut
Could you please explain that? I mean, maybe Im missing something, but I dont see why the usable capacity is the stopper here, even if we can reconfigure RAID.Why wouldn't usable capacity be the stopper for using StarWinds? Sure, he could reconfigure to RAID 6 from the RAID 10 he has now, but these are his production servers, and I'm assuming that he has a set number of IOPs that he wants and gets from the RAID 10.
He's previously stated that he's willing to take the performance hit for the DR server that will be offsite, but the local ones need to remain at the current level or above. He's designed a system that currently allows him to restart all of the VMs on a host that fails at the DR site (even more, he has the ability to spin up all servers at the main location at the DR site, though at a reduced performance rate).
-
@Dashrender well, that applies only if the production is 24/7, which hasn't been mentioned yet. Thus, if that is not so, he can do the reconfiguration afterhours. I'm pretty sure that there will be enough of time for RAID rebuild and StarWind implementation
-
@original_anvil said:
@Dashrender well, that applies only if the production is 24/7, which hasn't been mentioned yet. Thus, if that is not so, he can do the reconfiguration afterhours. I'm pretty sure that there will be enough of time for RAID rebuild and StarWind implementation
Eh? what does production hours have to do with IOPs? If the RAID 6 doesn't provide the desired normal production IOPs, then rebuilding to RAID 6 won't be viable.
-
He (me) doesn't want starwind.
-
-
How would starwind help here? if it's useful, i'm down. i just dont know enough about it to on a whim whip it up in a production environment.
-
@hubtechagain said:
How would starwind help here? if it's useful, i'm down. i just dont know enough about it to on a whim whip it up in a production environment.
Well, to that, I don't think it can. Unless you're willing to move to RAID 6 on both the local servers, or purchase more drives, you won't have enough storage to allow full storage failover between the hosts (we know that because you had to go to RAID 6 to get enough storage at the DR site).
What it could gain you - full server failure recovery on site, for free (well for more drive space or RAID 6 vs RAID 10 penalties).
