Tuesday, 21 September 2010

Create A Bootable USB Flash Drive Larger Than 4 GB

With the advent of larger and larger OS installs that come on dual layer DVDs (8 GB+), we need to revise our previous post methodology for these new OS files on a bootable USB flash drive.

The methodology we have been using for USB flash drives with large OS install files has been the following (Assume [Enter] after each command line):

  1. Start –> Diskpart (type in search)
  2. Right click on Diskpart and Run As Administrator
  3. Click Continue if prompted by UAC.
  4. List Disk
  5. Select Disk 3 (use whatever number in place of “3”)
  6. Clean
  7. Create Partition Primary
  8. Select Partition 1
  9. Format FS=NTFS Quick
    • Make sure to have the Quick switch or it will take _a long_ time!
  10. Active
  11. Assign
  12. Exit

Now, we can run the above commands using their abbreviations:

  1. Lis Dis
  2. Sel Dis 3
  3. Clean
  4. Cre Par Pri
  5. Sel Par 1
  6. Format FS=NTFS Quick
  7. Act
  8. Ass
  9. Exit

The following is a screenshot of the above command set being used to format an 8 GB OCZ Rally2 Turbo USB flash drive:

image

Once done we can copy the OS files off of its ISO using Virtual-Clone Drive (SlySoft Site) to mount that ISO with a drive letter in Windows.

This method requires Windows Vista, Windows 7, or server OS with the same code base or newer.

Note that there is no requirement to edit the Boot Configuration Database at all (BCD). We have been using this method for a while now to install beta OSs in our lab here without an issue.

We sometimes drop a Drivers folder on the USB flash drive too. As a just-in-case we add the Everyone NFTS group and give it FULL permissions to the USB flash drive after running the above steps.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Create Cluster Wizard Error: An enabled computer object for ‘TD-Cluster-01"’ was found.

We are on our second run through creating a cluster on our own network here.

After selecting the correct subnet for the management network and setting an IP for the cluster administration point we received the following error:

image

Create Cluster Wizard

An enabled computer object for ‘TD-Cluster-01’ was found. This usually means that the name is in use by another computer or cluster network name. If this is not the case then please disable or delete the Active Directory Computer Object.

After opening ADUC on our SBS v7 box we saw the following in our SBSServers OU:

image

We deleted the Computer account out of the SBSComputers OU and also removed the TD-Cluster-01 DNS A record under our domain’s forward lookup zone.

We then clicked Next in the Create Cluster Wizard after they were deleted.

The wizard went on to create the cluster successfully.

image

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Failover Cluster Validation Report – Network Failed, Valid IP Configuration Failed

We just finished figuring out how to work with each Hyper-V Server 2008 R2 cluster node having a pair of Intel NICs teamed and then having a set of defined VLANs _on the teams_.

We will publish a post on the proper methodology for NIC teaming and team VLAN creation on the Intel Modular Server soon.

Now, we ran the Validate a Configuration wizard in Failover Cluster Manager (FCM) and ended up being surprised by the following:

image

Validate a Configuration Wizard

Testing has completed, but one or more tests indicate that the configuration is not suitable for clustering.

When we click the View Report button we see the following:

image

Network: Failed

Click on the Network hyperlink and we see:

image

Validate IP Configuration: Failed

When we scroll down to the Valid IP Configuration list for the nodes we see:

image

Found duplicate physical address 00-15-17-xx-xx-xx on node hv_node-01.xx.local adapter Local Area Connection 9 and node hv_node-01.xx.local adapter Local Area Connection 8.

Found duplicate physical address 00-15-17-xx-xx-xx on node hv_node-02xx.local adapter Local Area Connection 9 and node hv_node-02.xx.local adapter Local Area Connection 8.

Note that the MAC addresses are indeed the same for both Local Area Connections on each node. The MAC address indicated belong to the two VLANs running on Team 1 (the Management team).

So, we searched to see what would come up to work around the problem.

From the KB article:

A configuration which uses IEEE 802.1Q VLAN Tagging to connect to multiple VLANs for the public network interfaces, and has a seperate (sic) heartbeat network, will fail the cluster validation tests.

The validaton (sic) test fails because the test detects that the same MAC address is in use on the 802.1Q VLAN networks.   However, this configuration is valid and is supported if this is the only failure in the validation tests and the failure is due to the use of IEEE 802.1Q VLAN Tagging .  All other cluster requirements still apply.

The validation will fail with the error, “Found duplicate physical address ...", and will detail which address is duplicate.

The article then states that the error will be changed to a warning in future product versions.

So, we can safely ignore the message.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Monday, 20 September 2010

Our WWW Is Migrating To SharePoint Foundation 2010

Our Web site is in the process of being migrated to SharePoint Foundation 2010 on OWN’s servers as of this writing.

While the DNS A record change winds its way through the various DNS servers out there our original SharePoint v3 site or the new site may show up.

Please note that any Outlook connected SharePoint features from our site will be broken and need to be reconnected once we set up the new SharePoint Foundation 2010 site!

Our original site will still be available and anyone with a logon and password will still have access to our exclusive content:

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Intel PROSet Adapter Team – Creating a VLAN Setup for Clustering on IMS

Our previous post on setting up an adapter team:

Once our adapter teams are created the next step for us is to create a dedicated VLAN for our Heartbeat network.

In this case we will have two adapter teams created:

  1. Management
    • NIC 0 Switch 1
    • NIC 0 Switch 2
  2. vSwitch
    1. NIC 1 Switch 1
    2. NIC 1 Switch 2

The above team names are what we are working with on our Intel Modular Server at this point.

So, we need to know what the index number is for each team:

  1. PROSetCL Team_Enumerate
    • image

Now that we know our index numbers, we can go ahead and create the two needed VLANs:

  1. PROSetCL Team_CreateVLAN 2 untagged Management
  2. PROSetCL Team_CreateVLAN 2 99 Heartbeat
  3. PROSetCL Team_EnumerateVLANS 2
    1. image

The “untagged” VLAN will remain connected to our management network. VLAN 99 will be our exclusive heartbeat network.

Now, in our Hyper-V Server 2008 R2 “console” we can see that we do indeed have 3 NICs to work with:

image

We will then go ahead and set our heartbeat subnet and bind the Hyper-V virtual switch to the vSwitch team.

Now that our teaming has been configured our node is ready to become a part of our cluster!

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

64-bit Linux Kernel Exploit – Full Root Rights

There is a new Linux kernel exploit that apparently is being exploited big time:

image

What does this mean for us?

We need to make sure that _any_ of our client’s hosted services running on Linux have an up to date kernel. We do so by contacting the hosting provider to verify their systems.

Otherwise our client’s sites may end up serving some exploits.

Further to detecting whether a system has been exploited:

One of the linked to articles on Slashdot Linux: Hole in Linux Kernel Provides Root Rights has this comment which is just too funny not to point out:

 image

Good to see a sense of humour there! :)

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Software Installer Setup.exe Link On Companyweb

With Office 2010 being such a large install there are some difficulties around deploying Office 2010 via Group Policy.

An alternative to deploying by Group Policy is to provide users with a simple way to access the Setup.exe file.

In our case, what we have been doing is dropping a link to Setup.exe on our client’s Companyweb site.

image

Users will get a couple of security prompts which they can click through and then the Office setup routine will start.

The format of the link when we set it up on the site is as follows:

  • file:///\\MY-SBS\ClientApps\Office2010ProPlus\Setup.exe

Once the link is created we will send out a memo to users indicating that they can find a link to the Office 2010 install on their Companyweb site.

The above screenshot would be in that memo as all users have their Internet Explorer open to the Companyweb page.

Note that this method can only be used with an Open License version of Office.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Saturday, 18 September 2010

Intel DQ35JO On Board Graphics Problem – Really Poor Performance

One of our shop desktops has a Core 2 Duo E6850 running at 3 GHz installed on an Intel Desktop Board DQ35JO Executive Series.

The system had Windows 7 Enterprise x64 installed a while back with all of the latest driver and BIOS updates being applied along the way.

The system has two older BenQ FP931 19” 4x3 LCD monitors connected to the on board video ports.

This system has had a strange behaviour since Windows 7 64 bit has been installed on it. When dragging a window between the two monitors or even moving it from one point on the desktop to another on the same monitor the drag process would be ridiculously slow.

It was almost like the window was being pulled through the water in the way it moved across the screen or screens.

We have a few in-house projects on the go today so while our client systems were doing their thing we pulled the workbench system apart.

There are two 320GB Seagate 7200.10 drives set in a RAID 0 stripe that had no connection issues. The CPU and cooler looked to be as expected as did the optical drive.

When we got to the memory we found:

  • Channel 1 slot 1: 1 GB KVR667D2N5/1G
  • Channel 2 slot 1: 512 MB KVR533D2N5/512
  • Channel 1 slot 2: 1 GB KVR667D2N5/1G
  • Channel 2 slot 2: 512 MB KVR533D2N5/512

When checking in the BIOS we saw that the memory sticks were in the correct place as memory status showed “Dual Channel Mode”.

So, we had a look at Kingston’s memory support site for the DQ35JO.

image

Note the distinct lack of any part number beginning with KVR533. We must have had the two 533 MHz parts in the parts bin when we put this box together and tossed them in to get a full 3 GB of RAM in the box.

We dug around in the parts bin and came up with four KVR800D2N6/1G sticks of RAM. We installed them, checked in the BIOS to make sure that all four sticks were being seen and then booted up Windows 7.

image

Once into Windows 7 we saw the above as well. So, we knew that our now 4 GB of RAM was being seen by the operating system.

But what of the window dragging and overall graphics problem?

It was gone!

Windows would now drag between monitors or around on a monitor with no lag time or any “fluid” like motions.

Since the DQ35JO’s on board graphics uses system RAM we can see why there would be a performance hit of significant proportions when the RAM in the system was the wrong spec.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Western Digital RE3 Series SATA Drives In Intel Hot Swap Expander Backplanes May Error Or Go Offline

It seems that there is an issue between the Western Digital RE3 SATA drives running at 3 Gb/s in Intel’s hot swap expander backplanes.

Root Cause
The SAS expander chip on Intel® Server Drive Enclosure AXX6DRV3GEXP and AXX4DRV3GEXP experience an undesired flow-control behavior when connected to a SATA disk drive running at 3.0 Gb/s. Drives operating at 3.0 Gb/s may not receive proper data or SATA flow control and may result in a timeout error. There is no timeout at 1.5 Gb/s. This interoperability limitation of the SAS expander with SATA disk drive is currently seen on Western Digital’s RE3 series SATA drives.

Products affected:

  • Intel Server Drive Enclosure AXX6DRV3GEXP
    • 6 drive expander backplane.
  • Intel Server Drive Enclosure AXX4DRV3GEXP
    • 4 drive expander backplane.

To mitigate the issue the drive needs to be set to run at 1.5 Gbp/s by setting the jumper at position 5/6.

image

It is also important to make sure that the hot swap backplane has the most current firmware.

The hot swap backplane firmware download link can be found on the server chassis support page.

SATA vs. SAS

As a rule we no longer use SATA drives in any of our server configurations.

With the cost of SAS drives having come down quite substantially over the last few years along with the huge performance benefits we get with using SAS drives there is really no reason to use SATA drives in a server setting anymore.

The exception to this rule is for Intel Solid-State Drives that run with a SATA interface.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Friday, 17 September 2010

16 GB Super Talent USB 3.0 Flash Drive Performance

We finally had a bit of time today to put together a system that we could install the Vantec 2-Port SuperSpeed USB 3.0 PCI-e Host Card.

One of the first things we did was to test the read and write speed of some USB 3.0 storage devices that we have here.

One device is the 16 GB Super Talent Express Drive USB 3.0 flash drive.

When the drive gets plugged into a USB 3.0 or USB 2.0 port we see the following:

image

The Super Talent has a burned-in “CD-ROM” partition with a few help files and this as shown on an XP Pro workstation:

image

There are some instructions out there on how to remove that partition but they are not too clear on the process as it is burned right into the stick.

We ran a number of read and write tests to the Super Talent to see what we would get for performance.

Read Performance

This is our average read speed:

image

  • 5 GB file copy Read: Average 58.0 MB/Second

Given that our 4 GB and 8 GB OCZ ATV Turbo USB 2.0 flash drives read and write at 35 MB/Second the above is a decent step up in overall read performance.

Write Performance

We were disappointed in the following though:

image

  • 5 GB file copy Write: 34.9 MB/Second

Ouch. That write speed is about the same as our OCZ ATV Turbo USB 2.0 drives.

The cost on the Super Talent was about the same as the price on the 4 GB version of the OCZ ATV Turbo drive too.

Conclusion

Between the CD-ROM partition and the not-so-great write performance we will be taking a pass on any further Super Talent Express Drive USB 3.0 drives.

With native USB 3.0 support a ways off on server and desktops with a few exceptions, we have some time to look out for products that will fully utilize the speeds that USB 3.0 has to offer.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Thursday, 16 September 2010

Some Intel Modular Server Reading

The Intel Modular Server site has a set of documents that explore a variety of subjects from solutions built on the IMS to guides and best practices for setting up the box.

image

On the New Documents page, as of this writing, is a document on how to work with the built-in IMS storage.

The above guide contains a very in-depth explanation of how to work with the IMS SAN.

A good overview of the IMS SAN storage is here:

We are finding that how we configure storage in the built-in SAN as well as any Promise VTrak E-Class and VTrak JBODs connected to the IMS is absolutely critical to the overall performance of a Hyper-V cluster.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Wednesday, 15 September 2010

TechDays 2010 – LFE240 – Clustering with Hyper-V Server 2008 R2 Presentation

I will be doing a Local Flavours track presentation on Tuesday October 5th at the Edmonton TechDays 2010 event!

My presentation is at 1540Hrs to 1645Hrs (3:40PM to 4:45PM).

An invitation code can be found in the above calendar link for an event fee reduction to $349.00 too.

Some of the topics to be discussed will be:

  • Intel Modular Server
    • Compute module, storage, and network configuration
  • Cluster Integration
    • How to create if new or existing domain.
  • Cluster Configuration
    • Hyper-V OS
    • Networking structures
  • Cluster Storage
    • Cluster Shared Storage
    • Promise VTrak integration
  • Cluster Maintenance
    • Live Migration
    • Troubleshooting

There are all of five or six slides for this presentation with most content being demonstrations based on a live cluster setup.

This will be a highly technical dive into Hyper-V clustering on the Intel Modular Server with a little business perspective tossed into the mix.

I will make sure to address some of the key technical points to successfully standing up a Hyper-V Cluster as there are a quite a few of them.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Create An iTunes Account Without A Credit Card

When it comes to starting up iTunes and wanting to go and purchase a song or other item new users need to create an iTunes account.

However, when we go through and do that we get to this point in the account set up process:

image

There is no room in the above screen to _not_ save a credit card and move on!

When setting up a new account we do not want to store a credit card number in iTunes anywhere!

So, how do we go about getting a new iTunes account without being asked for a credit card?

The answer is to run through the process of obtaining a Free App and use the account creation process that way.

When we get to the above page we see the following:

image

Note the None radio button that shows up.

Once we have finished setting up the account we can then go back in and register an iTunes gift card which is what we suggested that our client do for their iTunes account.

We also suggested that they keep a regular eye on their account just in case.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Promise VTrak E-Class – Storage Configuration

Much like the built-in SAN on the Intel Modular Server the Promise VTrak E310sD and E610sD Dual Controller RAID Systems are quite flexible when it comes to configuring storage.

If no storage configuration has happened on the VTrak yet, the WebPAM PROe console will land on the storage configuration wizard once we log on.

image

Since we need a few different disk configurations to be made available to our cluster based VMs, we will use the Advanced Configuration option.

Promise VTrak Storage Configuration

Our first step is to create a Disk Array (IMS equivalent is a Storage Group):

image

We can add as many or as few disks to the Disk Array as we need.

image

Now, once we have set our Disk Array configuration by choosing the required drives we need to configure the Logical Drives that will reside on the Disk Array.

The Promise process differs from the IMS storage configuration in that on the IMS we would be running each step as its own process. On the Promise VTrak we create the entire storage structure for a group of drives in one process.

image

  • Alias: DA01LD01
    • Disk Array 1 Logical Disk 1
  • Capacity: 275.31
    • .31 represents Disk Array (Storage Group) 3 Logical Disk 1

Once we click the Update button beneath the Logical Drive Creation section we will see:

image

Take note of the Maximum size noted to the right of the Capacity field when creating the logical drives.

Once we have finished creating the required logical drives we see:

image

When we click Next we are given a summary of the logical drives to be created. Click Submit on the summary page and eventually we will have our newly created logical drives.

image

In our case we had two 300GB 15K.7 SAS drives in the VTrak to start with. We received six more drives and installed them today.

image

So, after deleting the existing array and logical drive that was created on the original two disks any new disk arrays and/or logical disks that incorporate existing disks may require synchronization:

image

Managing SAN and DAS Storage

When it comes to managing storage from within the node’s Disk Management that .31 or .xx number really helps to figure out which disk belongs to which VM.

This is important because we need to stage storage based on I/O requirements where higher I/O storage will reside on Storage Groups/Disk Arrays with more spindles.

  • IMS Storage
    • Storage Group 1 (2-4 spindles/SSDs) (XX.0x)
      • Hyper-V OS
      • Quorum disk if required
      • Hyper-V Configuration and memory files
    • Storage Group 2 (generally 4 spindles) (XX.1x)
      • VM data storage partitions
    • Storage Group 3 (6-8 spindles) (XX.2x)
      • VM OS partitions
      • SQL database partitions
  • Promise VTrak
    • Disk Array 1 (XX.3x)
      • File shares for large file stores
    • Disk Array 2 (XX.4x)
      • Archival folder shares
      • Low I/O Desktop OS VMs

Since the dual quad port SAS connections represent an I/O limitation due to the pipe size we are careful to place lower I/O requirement storage on the VTrak.

Once we have our storage configured on the VTrak the next step is to get that storage available to the Hyper-V Server 2008 R2 cluster nodes on the Intel Modular Server.

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer

Tuesday, 14 September 2010

Intel PROSet Adapter Teaming In Server Core and Hyper-V Server 2008 R2

It turns out that creating an Intel adapter team in Server Core, or in this case Hyper-V Server 2008 R2 is actually quite simple.

Reading through the above page plus the PROSetCL.txt file made things seem a bit intimidating.

In reality, the process is actually quite simple. The PROSetCL.exe command is found in: C:\Program Files\Intel\DMIX\CL.

Here is the process assuming that we are in the above directory:

  1. PROSetCL Adapter_Enumerate
    • The NIC’s index number is at the front of each NIC listed.
    • image
  2. IPConfig /all
    • NIC MAC addresses are needed to map which NICs we need to team.
    • On the Intel Modular Server, the NIC’s MAC addresses will be close together for the on board NIC and again close together for the Mezzanine based NICs.
  3. PROSetCL Team_Create 4,2 TEAM1 VMLB
    • Where the numbers “4” and “2” are the index number for each NIC we want in the team.
    • We team one on board NIC and one mezzanine based NIC to allow for switch redundancy.
    • The management and heartbeat team will be ALB configured.
    • The team that we will bind the Hyper-V vSwitch to will be configured for Virtual Machine Load Balancing (VMLB).

Once we ran the above commands we saw:

image

Note the adapter team shown as TEAM1 will be our management and heartbeat NIC.

Further reading:

UPDATE 2010-09-20: A post on configuring the needed VLANs:

Philip Elder
MPECS Inc.
Microsoft Small Business Specialists
Co-Author: SBS 2008 Blueprint Book

*Our original iMac was stolen (previous blog post). We now have a new MacBook Pro courtesy of Vlad Mazek, owner of OWN.

Windows Live Writer