|What can be learned from this move|
I assume all is back up now but what can we do
| 3:47 pm on Aug 24, 2007 (gmt 0)|
I saw this and can wonder what could these companies have done to keep from losing 100k or I would assume more.
One thing is I assume they have no disaster program set up well this in itself is a bad management descision.
What I see could have avoided all this is a backup system set up on another server and all they had to do when they recieved the email is make sure the system backup or disaster site was updated make the necessary ssl changes (could be a problem)and point the dns to the new IP address.
If the ssl certificate was going to be an issue always have a backup cart to take orders or at least a page letting the customers to call or come back
This would have kept them from being down a minute and when the move was completed change the dns back over to the origional server.
I am small potatos yet I have a plan set up just for this reason all I need to do is have the dns changed to the new IP and within hours I am back in business.
You know you can't put all the blame on the hosting company as some of this down time is from lack of proper planning....and you need to look at this as your fault...
| 4:01 pm on Aug 24, 2007 (gmt 0)|
|I saw this and can wonder what could these companies have done to keep from losing 100k or I would assume more. |
Certainly the companies who had the potential to lose that much money shouldn't have been relying on the hosting company to do all the work! Our co-lo center moved to the next building and out "90 minutes max." downtime was over 3 hours during prime time for us. I would assume comparable underestimation for a move that far.
First, if I'd received a letter like that I probably would have started looking for a new hosting company ASAP. If I had to have my servers moved anyway, I'd decide where they would go.
Second, if I decided to stick with the company, I would probably be involved with the move myself. unless I had too much hardware to replicate, I'd probably buy some servers to put in the new data center, mirror the sites, then switch DNS over to the new servers well before the move. So downtime would (should) be negligible.
Then again, you're right, if you're making that much money, you should have a disaster recovery plan in effect long before a move, including a completely cloned setup at a different data center. I'd have DNS switched over to that data center well beofre the move and would switch it back until everything was well tested at the new data center.
| 11:36 am on Sep 12, 2007 (gmt 0)|
When Rackspace threatened to do this to us, I started kicking and screaming (metaphorically speaking!), made a formal complaint, wrote to the US CEO and the UK MD, asked how much of a billing rebate we'd be getting, how many consulting hours they'd be providing for the extra work we'd need to do, etc etc.
So far that's been an effective strategy, and they haven't moved us. I get the impression a few other customers did the same thing.