Just a reminder I’ll be delivering part two of my two-part presentation on how we do web content management on MOSS 2007 this Tuesday at 12:30pm. We had a great turnout for the first presentation so it will be interesting to see who comes back!! Honestly, it’s amazing how many of you SharePoint/MOSS guys I’ve met through the first presentation and I’m really looking forward to bumping into a few more of you—please come up and say hello!
The blurb on the user group site is a straight copy of the part one blurb so here’s what I’ll be talking about tomorrow:
westernaustralia.com was one of the first public-facing MOSS 2007-based internet sites launched in Australia and is billed as the Western Australia Tourism Commission’s flagship web site. Two years on and thirty MCMS 2002-based tourism sites are now being migrated to the MOSS 2007 platform. In the second segment of this two part presentation, Michael Hanes, the Development Coordinator/Tech Lead at Tourism WA, talks about the backend MOSS environments. In this presentation Michael presents the existing and replacement hardware environments, virtualisation, environment structure, farm configuration, security, site collection structure and variations, performance, tooling, content delivery (Akamai), and content deployment.
Jeremy’s aiming to record the presentation again so, all being well, the part two webcast will be available after the event in case you’re unable to come along.
Here’s part one in case you missed it. See you down there for part two!!
[Update: a vodcast of the presentation is now available here with a PDF of the slides and notes here.]
[Update: the original PowerPoint deck is now available.]
Great presentation again mate, I'm just encoding the web cast...will be up by end of night!
ReplyDeletethanks Jeremy for the upload and Michael for sharing, watched the whole thing.
ReplyDeleteA couple of things I was left wondering afterwards were;
why you used the IP lookup instead of the default behaviour to determine the Variation site page? and
regarding the content deployment by backup of authoring content and restore into production was confusing. I thought that MOSS publishing workflows on production would be able to manage the flow of draft content -> published content?
@Ben - thanks for watching the presentation and tracking down the blog!!
ReplyDeleteWe actually do both the IP lookup with a fallback to user agent language--the latter of which is helpful for internal users with a private IP address. The IP lookup still gets it wrong occasionally but the thinking at the time was it would be more accurate than a potentially mis-configured browser language.
Sorry for the confusion with the backup/restore! It's clear to me but expressing my understanding isn't always easy. The short answer is we've got two separate farms (one for authoring and one--well two actually--for prod); MOSS Content Deployment was *supposed* to bridge the gap but it never worked in our environment. Draft to published therefore only happens within the authoring environment; whatever's published by 10am every morning is stsadm -backuped from authoring and stsadm -restored to production... a painful, error-prone process! We'll be moving to a single-farm environment in the near future to eliminate this hassle and bypass CD altogether.
Two great questions--hope that answers your questions!
Hi Michael, I know that this may be off topic but considering that you're moving the authoring and production to 1 (one) farm. Have you considered an issue in this kind of set-up when a current document (it can be an image, pdf, etc.) has been tagged as approved/published and in the near future would be tagged again as draft? The end-user/anonymous user would be able to see the draft version of the document instead of the published.
ReplyDeleteI'm not sure if I have made my self clear but am I making sense?