VMware has released several patches on May 28-2009. The patches for ESX 3.5 with their descriptions are listed below:
ESX350-200905405-BG – PATCH
This patch updates kernel-source and kernel-vmnix to support the aacraid software driver update.The driver update fixes an issue where ESX 3.5 might fail on IBM, Fujitsu and SUN servers installed with Adaptec aacraid_esx30 driver. The failure occurs when the ESX host is running management agents and heavy I/O load on the volumes owned by aacraid_esx30 driver. Issues fixed in this patch (and their relevant symptoms, if applicable) include:
ESX350-200905404-BG – PATCH
This fix updates the tzdata package. The tzdata package contains data files with rules for various time zones around the world.This updated package addresses the following changes to Daylight Saving Time (DST) observations:* Future transition dates for Brazil were fixed
* Several Argentinian provinces decided not to observe Daylight SavingUsers in the affected locales, or users interacting with people or systems in the affected locales, are advised to upgrade to this updated package, which adds these enhancements.
ESX350-200905403-BG – PATCH
Issues fixed in this patch (and their relevant symptoms, if applicable) include:
ESX 3.5 might fail on IBM, Fujitsu and SUN servers installed with Adpatec aacraid_esx30 driver. This issue occurs when ESX host is running management agents and heavy I/O load on the volumes owned by aacraid_esx30 driver.
ESX350-200905402-BG – PATCH
Issues fixed in this patch (and their relevant symptoms, if applicable) include:
This fix includes an updated VMX RPM which addresses a robustness issue in VMX. Issues fixed in this patch (and their relevant symptoms, if applicable) include:
ESX350-200905401-BG – PATCH
Virtual machines using a NFS data store might fail after a HA failover event. This issue might result in two scenarios:1. When you have memory overcommitment with virtual machines on a NFS datastore, it results in the creation of a vswp file, which is a non-zero size swap file. In this scenario if HA failover events occur and the HA has the setting “Leave VM powered on”, then you might have a virtual machine failure on the host where the virtual machine was originally running before the HA event.2. If you do not have memory overcommitment with virtual machines on a NFS datastore, if HA failover events occur with the setting “Leave VM powered on”, further vmotions of the virtual machine running on the original host might fail.
- When an invalid license is specified to a license server or if the network is intermittent, the ESX server goes to an unlicensed state. The ESX server does not try again to acquire the licenses from the license server. This fix allows the ESX server to try to acquire licenses after a few minutes.
Issues fixed in this patch (and their relevant symptoms, if applicable) include:
- ESX 3.5 might fail on IBM, Fujitsu and SUN servers installed with Adpatec aacraid_esx30 driver. This issue occurs when ESX host is running management agents and heavy I/O load on the volumes owned by aacraid_esx30 driver.
Do you know exactly what a “robustness issue” is? Is that code for a security flaw, a potential guest crash, or host crash, or…?
Cheers,
Dave
Hi Dave,
Thanks for you comment. However, what exactly do you mean with your remark in regard to the ESX patches available post?
Are you questioning the robustness of ESX? And if so in what way do you think it has a “robustness issue”. Hope to hear from you.
Aah I see what you mean. This info is directly derived from the VMware website. No credit for me there. I was wondering the same thing.
.
As soon as I find out, I will get back to you
Frank