Jump to content


 


Register a free account to unlock additional features at BleepingComputer.com
Welcome to BleepingComputer, a free community where people like yourself come together to discuss and learn how to use their computers. Using the site is easy and fun. As a guest, you can browse and view the various discussions in the forums, but can not create a new topic or reply to an existing one unless you are logged in. Other benefits of registering an account are subscribing to topics and forums, creating a blog, and having no ads shown anywhere on the site.


Click here to Register a free account now! or read our Welcome Guide to learn how to use this site.

Photo

Best way to go with 1 server?


  • Please log in to reply
10 replies to this topic

#1 sweedishfish

sweedishfish

  • Members
  • 63 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:12:02 AM

Posted 23 June 2016 - 10:23 PM

We have a Server 2012 r2 Datacenter at work along with decent hardware specs. We are thinking about creating a domain for about 75 users. The thing is, even though the server is decent, we have no redundancies and this server will be our ADDS, DNS, Print Server as well as our DHCP. So once we get everyone onto domain, and lets say 3 months later if something happens  to the server, no computer will have internet connection until we fix it.

 

So the question is, what would  be the  best way to tackle this issue, and how to prevent it from before it even happens?



BC AdBot (Login to Remove)

 


#2 sweedishfish

sweedishfish
  • Topic Starter

  • Members
  • 63 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:12:02 AM

Posted 24 June 2016 - 02:35 PM

I would greatly appreciate it if someone can reply to this and provide me some suggestions.



#3 Wand3r3r

Wand3r3r

  • Members
  • 2,027 posts
  • OFFLINE
  •  
  • Local time:12:02 AM

Posted 26 June 2016 - 12:06 PM

Everyone is a volunteer here.

 

Unclear why you would be running datacenter in a standalone environment.  Putting all your eggs in one server is never wise which is why Microsoft recommends a minimum of two servers at DCs.  This gives failover/preserves AD, DNS, and dhcp if you do split scope.



#4 sweedishfish

sweedishfish
  • Topic Starter

  • Members
  • 63 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:12:02 AM

Posted 27 June 2016 - 06:53 PM

I'm telling my boss the same thing but they don't want to pay money for an other server. They want to do this to make things easier on IT.

 

The real question is, lets say they didn't listen to me and they added everyone to domain, and then sometime later motherboard died and there you go, no one has internet - What is the best way to solve the internet issue other then replacing the motherboard? Get everyone (75+ computers) out of the domain and make them local machines?

 

The only reason why I'm asking this because I want to solve the issue as quickly as possible. 



#5 RolandJS

RolandJS

  • Members
  • 4,527 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Austin TX metro area
  • Local time:02:02 AM

Posted 27 June 2016 - 07:05 PM

I've never set up a network server nor workstations depending upon a server; however, I am aware of scuttlebutt here and elsewhere that mention buying necessary external-existing hardware, buying necessary software, setting up a routine to backup the entire server [just the server contents] at least tri-weekly if not every evening, suggested was 1-3am+-.  If one is going to backup the workstations' contents, it was recommended a separate backup system be employed, here too 1-3am+-.


"...Get everyone (75+ computers) out of the domain and make them local machines?"  How long would this process actually take for each computer?


"Take care of thy backups and thy restores shall take care of thee."  -- Ben Franklin revisited.

http://collegecafe.fr.yuku.com/forums/45/Computer-Technologies/

Backup, backup, backup! -- Lady Fitzgerald (w7forums)

Clone or Image often! Backup... -- RockE (WSL)


#6 Wand3r3r

Wand3r3r

  • Members
  • 2,027 posts
  • OFFLINE
  •  
  • Local time:12:02 AM

Posted 28 June 2016 - 08:18 AM

"They want to do this to make things easier on IT."

 

Ask the age old question of "how important is your data?"  Doesn't appear too important if management is only considering one piece of hardware.

 

Concerning your question about removing everyone from AD.  You would not do that.  Being in AD doesn't mean its in charge of internet access.  You would not be using the server as a router.

 

Properly configured dns settings for a workstation would have the local [the server] dns server listed first and isp or google dns second.  If the local server is down the first dns entry would become unresponsive and the local workstation would use the 2nd dns entry.

 

"replacing the motherboard"

You really need to think about this one.  Unless you buy two same mainboards at the same time, you will find that a replacement mainboard of the same make and version is not available by the time of failure.  You will end up reinstalling server from scratch.  Only with a same mainboard is it plug and play.



#7 androbourne

androbourne

  • Members
  • 25 posts
  • OFFLINE
  •  
  • Local time:02:02 AM

Posted 29 June 2016 - 09:38 AM

While for something like 50+ users I would recommend a secondary server. If this just isnt going to happen you better have a good backup plan.

 

You may want to look into Acornis or some other form of backups that will allow you to convert your backup into a .vmk format and you will also need some spare equipment standing by. (again requires secondary server or high powered workstation etc... however it could be speced lower then a full blown failover server as a temporary solution).

 

You could then setup VMware or Hyper-V and just kick off your server to run in a VM in case of hardware failure, etc...

 

Long story short, without a secondary server for redundancy. You are only as good as your backup are and no matter what if that primary server goes down, they will go down for a brief time depending on your DR method.

 

(either that be a spare vm server, or changing DHCP and DNS to the router and doing a data file restore to a share somewhere etc...) 


Edited by androbourne, 29 June 2016 - 09:41 AM.


#8 neuronic

neuronic

  • Members
  • 52 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:New Hampshire
  • Local time:03:02 AM

Posted 29 June 2016 - 10:19 AM

We have a Server 2012 r2 Datacenter at work along with decent hardware specs. We are thinking about creating a domain for about 75 users. The thing is, even though the server is decent, we have no redundancies and this server will be our ADDS, DNS, Print Server as well as our DHCP. So once we get everyone onto domain, and lets say 3 months later if something happens  to the server, no computer will have internet connection until we fix it.

 

So the question is, what would  be the  best way to tackle this issue, and how to prevent it from before it even happens?

I would first start with an suitable backup solution so that if things good south you are not totally in the woods. I would then get a smaller server to run as a backup hypervisor, which can function as a disaster point until you can get a new fully loaded server shipped. Using your backup solution setup a continuous recovery stream to the backup low end server just in case. 



#9 neuronic

neuronic

  • Members
  • 52 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:New Hampshire
  • Local time:03:02 AM

Posted 29 June 2016 - 10:25 AM

We have a Server 2012 r2 Datacenter at work along with decent hardware specs. We are thinking about creating a domain for about 75 users. The thing is, even though the server is decent, we have no redundancies and this server will be our ADDS, DNS, Print Server as well as our DHCP. So once we get everyone onto domain, and lets say 3 months later if something happens  to the server, no computer will have internet connection until we fix it.

 

So the question is, what would  be the  best way to tackle this issue, and how to prevent it from before it even happens?

 

And any device with a static IP will have an Internet connection. Maybe you should also be looking into some system management options. 

 

 

Hybrid onsite/cloud system and file based backups -> a small recovery point server just in case of failure -> System management to proactively avoid issues with hardware failures, security issues, and other things. -> On going verification of backups -> I.T. Budget for secondary primary server. 


Edited by neuronic, 29 June 2016 - 10:26 AM.


#10 sweedishfish

sweedishfish
  • Topic Starter

  • Members
  • 63 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:12:02 AM

Posted 29 June 2016 - 09:51 PM

Everyone, thank you so much for your responses. I definitely have a better understanding now.

 

By the way, that motherboard issue was just an example.



#11 JohnnyJammer

JohnnyJammer

  • Members
  • 1,117 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:QLD Australia
  • Local time:05:02 PM

Posted 30 June 2016 - 06:47 PM

Basically as others have said, with that many users. With in 5 hours the cost of a second server would have paid for its self witht he amount of down time you woudl ahve.

The cheapest and easiest way would be to use vmware convertor and create a virtual server from the physical, this would at least allow you to fire it up and start running again if you had a decent enough workstation but i must say this would be with a risk.






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users