My virtualization project is 75% finished.

Somehow I don’t feel 75% more intelligent, well-informed or experienced.

Many people were aghast when they read the news about the BIG VMWARE BUG.  I read about it online that evening.  It was nothing serious - it just wouldn’t let any virtual machine start up on August sixteenth.  Fortunately we had been excused from this virtual nightmare (or so I thought).

The following day I let my cohorts know about the small situation.  Just in case they weren’t listening, which tends to only happen when I speak, VMware underscored the point by failing to let one of our virtual machines start.  Mind you, they could reboot; just not shut down and restart.

At this point I had two essential virtual machines and several less essential ones that needed to be rebooted due to Windows updates and resource changes.   I realized that I couldn’t make the resource changes because they required shutting down the machines.  If I shut them down, they’d never come back up (until I got the patch and installed it).

Today I hit the VMware site and discovered they had finally gotten around to releasing the patch.  They were most emphatic that you read ALL FOUR BULLETINS before downloading.  The first bulletin mentioned what they did.  They pulled the defective update from their site so no one else would download it.

I don’t know about you, but that made me feel MUCH better.  I almost forgot that I needed a patch for this update.

My very alert cohort told one of the ESX servers to update (Go Remediate Yourself!).  It responded promptly by firing off all sorts of alerts and failing to migrate a single virtual machine to a different host. He was most impressed.  After he got done being impressed, he manually moved the virtual machines.

Well, he manually moved all but one of the virtual machines.  This was the file server, so it couldn’t exactly be shut down during working hours.  So there it sat, holding up progress.  We also couldn’t shut it off because it would never come back up.   Catch 22.

Now it’s ten pm and I have shut down the machines that needed adjusting.  I gave one more RAM but can’t edit the other because it’s still on the host that’s in maintenance mode and won’t let me edit it.  I feel so safe now.

Since I started, about two hours ago, the host that’s updating is still in maintenance mode.  Gee, only three more to go!

While on the VMware site, I noticed they strongly recommended against manually adjusting the date on a host server (the suggested interim workaround for the bug).  That’s nice… especially as I got a virtual machine working precisely by manually adjusting the date.   Perhaps they shouldn’t present the information if they don’t want you to follow it.  I’m just silly that way.

I’m a bit surprised that VMware didn’t get more press on this.   I sure hope they enjoy this tiny bit of notoriety.