It was a technically challenging migration for me, though having worked through it I could probably do the entire thing in two evenings. Below is a list of the key posts and explanations if you have to ever do this. Note that the posts are updated as I learned things, so you really do need to read ‘em through to avoid my mistakes.
- Migrating a domain with related Google Apps and hosting files to a new service: This one worried me the most, but turned out to be straightforward. I learned a lot about how DNS works.
- The DreamHost migration moving the first domain
- The DreamHost migration getting a transfer authorization code: This stuff works well, I recommend creating a spreadsheet/table to track dates and migration stages for each domain. (Ex: review email settings on domain, get TAC, unlock domain, fill out DH form, get email from DH, authorize, get email from domain host, authorize, get confirmation from DH)
- DreamHost a better Google Apps choice than Google: You get everything Google Apps provide, with more control, a consolidated domain management environment, and the ability to add additional services.
- How to move files from one web site hosting service to another: This was where I floundered the most. I could have done it all in a few minutes of my time (hours of computer time) if I’d understood the archive transfer process.
I’m very pleased with DreamHost – all of the issues I had with Lunarpages seem to have been resolved.
I do want to mention that Lunarpages was very cooperative during the migration; Network Solutions resisted one domain move a bit, but Lunarpages never kicked. They also provided the ability to download an archive of my entire Lunarpages site collection – a good example of data freedom.
Lunarpages is not, by any measure, the worst company I’ve worked with. For several years they provided reasonable service at a good price. My read is that Lunarpages failed to invest in their core business and people, and focused too much on short term returns and cash flow versus long-term health. In the end they weren’t providing enough DNS control to customers, they didn’t stay on top of Google Apps integration, they didn’t provide WebDav services, they weren’t quite honest enough with customers (ex. automatically upgrading customer plans to reflect decreasing costs would have been wise) and they got trapped in a spiral of decreasing quality and increasing support burdens. They’re in a tough spot now, where they need to focus aggressively on customer retention and quality improvement, while taking a large cash flow hit.
Now for my next set of projects: upgrading our main workstation from 10.4 to 10.5, implementing iPhone 2.1, then returning to the synchronization and calendaring wars. (Hint: Our family google calendar is really working – with help from spanning sync).