Ben Kevan’s Blog

RedHat.com Servers Infiltrated - Time for Novell’s big move?

I was going about my day at work and went to go on www.redhat.com, only to find out that “The site is temporarily unavaliable. Please try back later.”

I thought nothing of it, until I tried to go on 30 minutes later and got the same message. My initial thought was “they got hacked” as they sent out the critical notification about the rhn leak they had the other day.

I checked a few hours later, to find the same message with an update static page that gave me support numbers.

I later found this article explaining the infiltration. So why doesn’t redhat.com have anything on their front page? (vmware did this when they had their Aug 12th blow up!). I guess a great question would be “What would Novell Do?”. They are the other major players in the Enterprise environment. Do we believe that Novell would have been honest and upfront and notified all users ASAP? I have yet to get an email from my RedHat Rep (I got one within HOURS of the Aug 12th discovery from my VMware rep) and I don’t see any details on the redhat.com page.

I think this is a huge opportunity for Novell to step in, and talk about the security of their online updates etc.

ESX 3.5u2 Upgrade doesn’t show Health Statistics for all Hosts in VC

So, you upgrade your Virtual Center to 3.5 u2 and then upgrade your 3.5u1 hosts using the upgrade tarball.

You reboot the host, check it in VC and the build number is right, however, you do not get the “Health Statistics” under the “Configuration” Tab.

It seems as though during the upgrade of the Virtual Center Database from Version 4 to 5 doesn’t actually add either the required tables, or the required views to the hosts that are still pre u2 (this shouldn’t be the case, as you should ALWAYS upgrade Virtual Center before going to the new version of ESX).

One of the main reasons for moving to u2 for me was this health status feature. So I needed to find a plan. VMware was unable to give me any solution, or any reason of why it happened, but I found that disconnecting then reconnecting the host in virtual center will give you the Health Status option.

I’ve tried bouncing the management daemon, bouncing the box, and neither worked (so to me it wasn’t ESX related, but what the host was allowed to “show” in its view).

Oh, and if you really want to, you can put the host in maintinence mode, remove it and re-add it, but why do that when you can disconnect and reconnect without moving any of the virtual machines?

Hope this helps someone out there.

Ben

Major VMware ESX 3.5/3.5i Update 2 Bug Discovered

Have you upgraded to VMware ESX 3.5 already? If so be sure not to power off ANY of the virtual machines on there. A new bug in VMware ESX 3.5/3.5i Update 2 will keep the powered off virtual machine to power back on, what could cause this type of behaviour you might wonder? A bug that incorrectly expires your license on August 12th (which happens to be today).

Other services that check licenses (ie Vmotion / HA / DRS) may also not work.

Please see the knowledge base from vmware:
http://kb2.vmware.com/kb/1006716.html

Here is a copy of the Symptoms from Vmware’s website:
A virtual machine fails to power on, fails to leave suspend mode, or fails to migrate with VMotion and the following message is displayed in the vmware.log file for the virtual machine:
This product has expired. Be sure that your host machine’s date and time are set correctly.
There is a more recent version available at the VMware web site: http://www.vmware.com/info?id=4.
————–
Module License Power on failed.

Currently running virtual machines will continue to run but will not be able to perform migration with VMotion, or power on virtual machines.

The workaround is to disable NTP if set, and change the date to a previous date (I would do the 8th .. so you have time to upgrade with the new ESX 3.5u2 packages due out Aug 13th at NOON PST.