Jump to content

First use questions from first Windows Server 2008 test deployment


Recommended Posts

Guest Andrew Hodgson
Posted

Hi,

 

Today we upgraded a test domain in our environment from Windows 2000

server to Windows Server 2008, and came up against a few issues.

These were by no means show stoppers, but I want to get this right for

when we do this in a couple of weeks.

 

Firstly, I wanted to put the boot partition as 20GB, but I am reading

that we need 30GB partitions or larger for a Windows 2008 environment?

Is this the case, or will we be ok with a 20GB partition?

 

We had some errors when using the dcpromo tool to install AD onto the

server, mainly because it thought we had an adapter on the network

with a dynamic IP address. We are using a teamed system from Broadcom

designed for Windows Server 2008 on our PowerEdge 1950, however, there

is still an adapter with an unused IP address which is the Tunnel

Adapter. Should this be disabled, or is it fine to leave things like

this?

 

Finally, is it best to put all the AD files onto a separate partition,

or should we think about creating 2 extra partitions, one for the

database/sysvol share, and one for the AD logs? These will be on the

same RAID-6 volume.

 

Thanks.

Andrew.

  • Replies 2
  • Created
  • Last Reply

Popular Days

Guest Meinolf Weber
Posted

Re: First use questions from first Windows Server 2008 test deployment

 

Hello Andrew,

 

see inline

 

Best regards

 

Meinolf Weber

Disclaimer: This posting is provided "AS IS" with no warranties, and confers

no rights.

** Please do NOT email, only reply to Newsgroups

** HELP us help YOU!!! http://www.blakjak.demon.co.uk/mul_crss.htm

> Hi,

>

> Today we upgraded a test domain in our environment from Windows 2000

> server to Windows Server 2008, and came up against a few issues.

 

Did you prepare the schema with adprep before? I think it was an additional

server in the domain?

> These were by no means show stoppers, but I want to get this right for

> when we do this in a couple of weeks.

> Firstly, I wanted to put the boot partition as 20GB, but I am reading

> that we need 30GB partitions or larger for a Windows 2008 environment?

> Is this the case, or will we be ok with a 20GB partition?

 

Minimum harddisk space 10GB, recommended by MS is 40GB. A first install from

scratch with no roles and features is roundabout 7GB.

> We had some errors when using the dcpromo tool to install AD onto the

> server, mainly because it thought we had an adapter on the network

> with a dynamic IP address.

 

What about the error? For servers you should always use fixed ip's, will

solve some problems before they occur.

> We are using a teamed system from Broadcom

> designed for Windows Server 2008 on our PowerEdge 1950, however, there

> is still an adapter with an unused IP address which is the Tunnel

> Adapter. Should this be disabled, or is it fine to leave things like

> this?

 

For the first install i would use only ONE physical NIC, disable all others

and if the server is running, then switch to the teaming, so you know the

server itself is fine and if problems occur you know it's because of the

teaming. All other not used NIC's should be disabled on DC's.

> Finally, is it best to put all the AD files onto a separate partition,

> or should we think about creating 2 extra partitions, one for the

> database/sysvol share, and one for the AD logs? These will be on the

> same RAID-6 volume.

 

Think this is more a rhetoric question. The location depends on your space

requirements and how fast the company/AD is growing up. But even if you need

more space in the future you can relocate it also later to a different bigger

partition.

> Thanks.

> Andrew.

Guest Andrew Hodgson
Posted

Re: First use questions from first Windows Server 2008 test deployment

 

On Tue, 27 May 2008 20:57:15 +0000 (UTC), Meinolf Weber wrote:

> Andrew Hodgson wrote:

>>

>> Today we upgraded a test domain in our environment from Windows 2000

>> server to Windows Server 2008, and came up against a few issues.

>

>Did you prepare the schema with adprep before? I think it was an additional

>server in the domain?

 

Yep we prepared the schema and this all went well.

>

>> These were by no means show stoppers, but I want to get this right for

>> when we do this in a couple of weeks.

>> Firstly, I wanted to put the boot partition as 20GB, but I am reading

>> that we need 30GB partitions or larger for a Windows 2008 environment?

>> Is this the case, or will we be ok with a 20GB partition?

>

>Minimum harddisk space 10GB, recommended by MS is 40GB. A first install from

>scratch with no roles and features is roundabout 7GB.

 

That is interesting - I used the standard image, and I created a 20GB

partition, it wouldn't install onto this, telling me the minimum

partition size for installation was just over this, so I just

increased it to 30GB, but was just a bit worried about growth of the

OS partition if I made it 30GB.

>

>> We had some errors when using the dcpromo tool to install AD onto the

>> server, mainly because it thought we had an adapter on the network

>> with a dynamic IP address.

>

>What about the error? For servers you should always use fixed ip's, will

>solve some problems before they occur.

 

The machine was using fixed IP addresses, but the Tunnel adapter which

was installed as part of the default system was still on DHCP. We

couldn't see this anywhere in the networking properties, only our

standard NICs were listed, and the teaming was working successfully.

>

>> We are using a teamed system from Broadcom

>> designed for Windows Server 2008 on our PowerEdge 1950, however, there

>> is still an adapter with an unused IP address which is the Tunnel

>> Adapter. Should this be disabled, or is it fine to leave things like

>> this?

>

>For the first install i would use only ONE physical NIC, disable all others

>and if the server is running, then switch to the teaming, so you know the

>server itself is fine and if problems occur you know it's because of the

>teaming. All other not used NIC's should be disabled on DC's.

 

Actually looking at this after the event we don't have any spurious IP

addresses showing up in the DNS, so it appeared to work ok.

>

>> Finally, is it best to put all the AD files onto a separate partition,

>> or should we think about creating 2 extra partitions, one for the

>> database/sysvol share, and one for the AD logs? These will be on the

>> same RAID-6 volume.

>

>Think this is more a rhetoric question. The location depends on your space

>requirements and how fast the company/AD is growing up. But even if you need

>more space in the future you can relocate it also later to a different bigger

>partition.

 

Thanks. I will probably put everything onto a d:\ partition.

 

Thanks.

Andrew.


×
×
  • Create New...