Tales from the Datacenter

WordPress Upgrade

As anyone remotely familiar with WordPress 2.7 will know, 2.7 introduced a very cool feature: the automatic upgrade.  This is cool for a few reasons.  First and foremost, it prevents the blog owner from having to manually copy the new version over to their web server.

In any case, it’s quite annoying when you administer four blogs and all but one detect the updated version.  You see, WordPress is on a 12 hour update-check.  So, you could conceivably be forced to wait for up to 12 hours for your WordPress installation to detect that it needs to be upgraded.  Not anymore!

If you have access to your raw database, you can browse to the wp_options table.  In that table will be a row called update_core.  Delete this row.  Now, I will give this disclaimer (as any good technical advice does): BACKUP! before you do anything to your database.  After you’ve deleted said row, refresh your dashboard (you don’t even have to log out and back in) and look at the top for your upgrade notification.

I’ve seen some message boards that suggest altering this row.  This is bad for a number of reasons, not the least of which is the fact that the update_core data itself is serialized and you can break your installation if you screw up.

Now, I say all of this to say this: WordPress 2.8 is pretty cool.  Check here for a complete list of the new features.  Now, an excerpt:

On June 10th, 2009, WordPress Version 2.8, named for noted trumpeter and vocalist Chet Baker, was released to the public. For more information on this enhancement and bug-fix release, read the Development Blog Announcement and see the Changelog for 2.8.

DVD Region

In my playings-around with Windows 7, I’ve really loved the fact that it plays DVDs and AVIs nativly with no extra codecs or applications.  I have to tell you, it’s really nice to be able to just pop a DVD in and have it play automagically.  Until this happens:

Now, if you don’t get it at first, read it again.  Anyway, I was forced to reset my region to 1 (from 1), taking up one of the five changes that I’m allowed.  Not that I’m going to be changing my region to anything else, but still, I find this dialog box humorous.

Windows 7 Windows Management

Around the office, most of us are used to using UltraMon to manage our windows on the beautiful 23″ monitors we have.  Unfortunately, for the time being, UltraMon doesn’t provide support for 7, so some of us are left manually wrangling windows around.

In a rare moment of helpfulness, I managed to find some keyboard shortcuts to help.  One of the plus sides here, is that when using the shortcuts below, a window can be docked to the inside of your dual monitor setup (since you can’t do it with the mouse).

Oh, and for those of you that I see using “Ctrl+Alt+Del” and pressing “Enter” to lock your computer, try Win + L.

Remote Server Administration Tools for Windows 7 Release Candidate

For future reference, here are the remote administration tools for Windows 7:

Remote Server Administration Tools for Windows 7 Release Candidate (RC)

Remote Server Administration Tools for Windows® 7 RC enables IT administrators to manage roles and features that are installed on computers that are running Windows Server® 2008 R2, Windows Server® 2008, or Windows Server® 2003, from a remote computer that is running Windows 7 RC.

Unsupported Cluster Configuration

Maybe Michael should go away more often.  I seem to get a lot more done when he’s out.

Today, I added the third node to the Hyper-V R2 cluster we’re currently testing.  When I did, VMM freaked out and all of the guests on the other two nodes went into a warning state.  The error they displayed:

Warning (13921)

Highly available virtual machine <INSERT NAME> is not supported by VMM because one or more of its network adapters is not configured correctly.

Recommended Action

Ensure that all of the virtual network adapters are either disconnected or connected to highly available virtual networks.

So, how do we fix this issue?  You use Google, of course.  A quick Google-ing gave the following information (edited for grammar):

For a VN [Virtual Network] to be marked as HA, the location and tag of the VN in all nodes must be the same.  For each VN in the cluster[, be sure that]:

• The NICs to which the VN is attached in each node have the same location.
• The Tag in the VN in each node is the same.
• The VN Name is the same.
• After you commit changes, refresh the cluster so ensure that the VN is detected as HA.

Our problem, specifically, was related to the network tagging.  To fix the tag, do the following:

1. Open VMM, and navigate to the “Hosts” view.
2. Right click on one of the hosts and select “Properties.”
3. Navigate to the “Networking” tab.
   
4. Once, there, you’ll need to click each network (in this case, we have two) and add text to the “Network tag:” field.  Keep in mind that these tags will need to be the same across all your hosts in order to make them highly available.
   
5. Rinse and repeat for the rest of your hosts.
6. Once you finish tagging, you’ll need to refresh the cluster node in VMM.
   

That fixed our problem.  Maybe it will help you.

More information here in the “Troubleshoot ‘Unsupported Cluster Configuration’ Status for a Highly Available Virtual Machine” section.

The subsystem needed…

So, I’m bee-bopping along setting up a new Server 2008 R2 virtual host, when I get to our SAN software’s install.  When I tried to run the executable, the following error was thrown:

The subsystem needed to support the image type is not present.

A quick Googling led me to this Technet post.  The relevate parts of the post:

WOW64 is not present in Microsoft Hyper-V Server 2008 R2.  This change was made because Microsoft Hyper-V Server is supposed to be a single purpose OS, and users have asked for it to be as small as possible.  So, whenever possible, we will not include general Windows components that are not needed by Hyper-V.

And the resolution:

Issue this command to install WOW64 support:

start /w ocsetup ServerCore-WOW64

And of course: reboot.

Server Core Patches

Need to see what patches/hotfixes are installed on Server Core?  There’s a command for that:

WMIC QFE

You can also send this to a text file for easier viewing:

WMIC QFE > [FILENAME].txt

2008 R2 High Availability

Michael and I are probably the only two people in the world that would find this even remotely amusing, but…

We’re testing 2008 R2’s Failover Clustering.  In particular, we’re going to test Hyper-V and highly available machines.  So, in Michael’s infinite wisdom, he asked me what to name the first test cluster.  I suggested the following:

Installation Results

When you run the script mentioned here (to updated a Server Core 2008 install), you’ll see a part of the output called “Installation Results.”  In the infinite wisdom of Microsoft, they don’t seem to feel the need to plainly document these results.  They have instead included a number (1-3) to indicate the results of the installation.  Here are these codes in case you’re looking:

Exit Codes:
0 = scripting failure
1 = error obtaining or installing updates
2 = installation successful, no further updates to install
3 = reboot needed; rerun script after reboot

More here.

How Google Routes Around Outages

No, that’s not my title.  The folks over at Data Center Knowledge have an interview with Urs Holzle, Google’s Datacenter Operations Manager:

Making changes to Google’s search infrastructure is akin to “changing the tires on a car while you’re going at 60 down the freeway,” according Urs Holzle, who oversees the company’s massive data center operations. Google updates its software and systems on an ongoing basis, usually without incident. But not always. On Feb. 24 a bug in the software that manages the location of Google’s data triggered an outage in Gmail, the widely-used webmail component of Google.

Just a few days earlier, Google’s services remained online during a power outage at a third-party data center near Atlanta where Google hosts some of its many servers. Google doesn’t discuss operations of specific data centers. But Holzle, the company’s Senior Vice President of Operations and a Google Fellow, provided an overview of how Google has engineered its system to manage hardware failures and software bugs.

Read more at Data Center Knowledge: How Google Routes Around Outages.