Bladewatch.com

Got involved in a procedural conversation with Fraser. Fraser is “mini me”, rather my little brother, who regularly reminds/corrects me about ITIL, the change process and everything else. He was discussing an issue he had experienced at work a few months ago.

A change was made the san storage, nothing serious, nothing that could be described as outside of ‘business as usual’. At the same time, users found that their shared departmental drives using that storage had a performance degradation, an investigation ensued and a debate then got raised regarding “MR SAN” who was working on the san storage.  Questions like:

• What work was done on the san storage
• Who approved that work
• What impact analysis was done
• What can be done to prevent this
• Was the change done to the san storage linked to the performance degradation.

The problem “MR SAN” had was that there was nothing to indicate why there was a performance degradation, therefore it was assumed that the configuration work on the SAN was the cause.  Discussions about the change process, about how we minimize this activity were had, what changes to the approval, the change mechanism and what best practices needed to be implemented.

As we move to virtualization, to common infrastructure, common platforms cross business lines and per business lines, we need to evolve the processes accordingly.  Can we agree what is a change on a virtual platform and what isn’t?  At the same time if we are to raise a change for every ‘transaction’ made on the infrastructure does that not de-value the change process? How would that fit within business as usual activity and extra-ordinary or planned work?

On the wintel side within the virtualization world, surely:

• Starting a service is a change to the current configuration and operating environment
• A reboot of a server that isn’t in use is a change to it’s operating environment – but if it’s not in use is it production?
• Clearing down disk space, or migrating a virtual machine from one host to another is a ‘change’ to configuration.
• Re-allocating a server to another business unit for billing, is that not a change?

1. We need the procedures for managing the platforms to evolve with the evolution of that platform.
2. We need to manage the ‘value’ of ‘change’ and accept what is a business as usual activity as they come on stream – as I get a san solution, we need to agree, what’s change and what’s business as usual.
   1. For example allocating new disk space is business as usual.
   2. Changing the performance settings is a change?
   3. Could we not discuss this in generic industry wide terms?

It’s difficult and it’s not going to be fixed overnight, we need to manage my ability to deliver – ‘Ken demands his server be rebooted at 3am’, and the audit/reporting process – do I really call a change analyst at 3am to discuss the impact and seek approval?

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.