Most servers have updates marked as critical or recommended over the life of a server. Each of them is categorised to ensure their importance is duly noted by relevant IT support staff. However what happens if these are not updated or a categorisation is incorrectly defined?

We recently encountered a Dell PowerEdge 2900 server which took two of it's SAS disks offline - thus stopping the Raid array. In laymans terms: preventing the server from booting into Windows 2003, rendering the complete network unable to access data and emails on the server. It turned out that a Recommendation issued at the end of July 2008 recommended a special update revision of all of the SAS disks installed.

Without this update it is possible for the disks to be taken offline if write/read accesses to the disks exceeded the allowed time period. The result - without warning your server goes offline, and you need to call a Dell service engineer to put the disks back online and rebuild the Raid array.

Does it apply to a single disk manufacturer? From what we can see it applies to numerous manufacturers including: Hitachi, Fujitsu, Maxtor, Seagate, Western Digital, . In the words of one Dell engineer, "This is a time bomb waiting to happen."

For details on this see the Dell SAS Hard Drive Firmware Utility readme document on the Dell Drivers and Downloads site.

Our recommendation is to review both recommended and critical updates for Dell servers on a monthly basis. This can be achieved by signing up using the Dell Managed Subscriptions service. For more information, check the current updates available for your server...

(c) Protocol IT 2008