Thanks for writing cogently on this. I do want to add my comment - change control is NOT bad, except when it becomes a symptom of cover-my-a$$. Inadequate care given to changes (i.e., a lack of change control) is the number one cause of downtime I've experienced throughout my career in medium-to-large companies.
The number two cause is flat-out human error. (Removing the wrong cable, punching the wrong button, etc. - one can attribute many, if not most, of these to not validating changes in an appropriate test environment or not being capable of repeating test environment actions in the production environment reliably.) Regards, Michael B. Smith, MCITP:SA,EMA/MCSE/Exchange MVP My blog: http://TheEssentialExchange.com/blogs/michael Link with me at: http://www.linkedin.com/in/theessentialexchange -----Original Message----- From: Ken Schaefer [mailto:[EMAIL PROTECTED] Sent: Sunday, October 05, 2008 10:11 PM To: NT System Admin Issues Subject: RE: Microsoft's Free version of Hyper-V Ships! & a random Friday Q > -----Original Message----- > From: Martin Blackstone [mailto:[EMAIL PROTECTED] > Subject: RE: Microsoft's Free version of Hyper-V Ships! & a random Friday Q > > Obviously this is an outsourcer issue, not a VMware issue. > But the bigger question is why throw out the baby with the bath water? > Wouldn't the best solution for the customer be for them to keep the existing > VMware and switch to better outsourcer, or is the whole enchilada (systems, > servers, hardware, etc) outsourced? When you are a quarter of a trillion dollar a year business, even the tender process to get a new outsourcer takes a year or more. Remember, this is a global 500 company - their outsourcer (at least for infrastructure, is one of those 3-letter outfits) Provisioning a VM involves a fair few teams - SAN storage, getting the SAN sync setup to the DR sites, working out which subnet to put the machine into and getting IP addresses from the network team, getting FW rules in place from the security team, getting sign off on the costings from the business owner, getting the outsourcer to put the base build down and any addons/patches/etc required for whatever application software is doing on the box. All of this needs to be done during the regular change windows, which means waiting weeks as you have changes being done by different teams. For our Ops Manager rollout (it's reasonably big - 18 management servers) we have NLB web console boxes. Just to get NLB allowed took about 4 weeks, with a number of meetings between the outsourcer, us, the networking team and the business. They need to make changes to VMWare configuration, and to the physical Cisco switches that the virtual switch is connected to. Because they don't have any other NLB boxes they are worried about port flooding and all sorts of other things. We have four MS clusters in this project - getting cluster heartbeat network cabled has taken nearly three months. Even the person who lays the cable is different to the person who patches it into the box - and they have different change windows. For an outsourcer - they get paid to keep things running, and to meet SLAs. Change introduces the risk of things not working, so they are resistant to changing anything. If they can get the changes in, whilst taking the longest possible time, they will. Cheers Ken > Is there a mechanism available that will allow you to import the existing > VM's into Hyper-V or is the customer not able to take the VM's with them? > > -----Original Message----- > From: Phil Brutsche [mailto:[EMAIL PROTECTED] > Sent: Sunday, October 05, 2008 10:25 AM > To: NT System Admin Issues > Subject: Re: Microsoft's Free version of Hyper-V Ships! & a random Friday Q > > Agreed, they either need a new outsourcer or the in-house knowledge to > do it themselves. > > Don Ely wrote: > > 2-3 weeks to provision a VM?!?!?!?! WTF is that about? > > -- > > Phil Brutsche > [EMAIL PROTECTED] ~ Finally, powerful endpoint security that ISN'T a resource hog! ~ ~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/> ~
