Content-type: text/html Set-Cookie: cookiehash=D8TIX1F9GFT8ITMG7PCWDC1UDL31CF7Q; expires=Mon, 23 Feb 2026 00:00:00 GMT; path=/; DMI News

DMI News

Previous Entry.. Next Entry..

The Second Server

April 18, 2010 02:16

As most of you don't know, I have two servers for the site. Only one of them is actually used. Yes, I pay full price for a second server that really isn't doing anything except acting as a backup for the first. It's not even a functional backup. It's just a hosted server that I copy all of the data from the site over to on a reasonably frequent basis. I purchased it the same week that my hosting provider had a transformer fire in their facility that ended up shutting me down for 6 days. Since I have paying customers which use the server for email and other things, I decided to always keep an active, functional backup at a separate location, which I could switch over to in a matter of minutes if I absolutely had to. Since then I have also done some work on the site to loadbalance it, such that it can always operate on multiple servers, and seamlessly handle any one of them dying without any impact to the users.

Of course, it costs money. Quite a bit of money each month that lets face it, in all honesty, doesn't really justify its existance. And while loadbalancing would be great for those few times that the site is REALLY getting slammed, the rewriting of most of the server applications has cut way down on the need for peak traffic balancing as well. It's nice to be ready to handle a triple slashdot effect, but it really doesn't occur frequently enough to justify the preparedness, and if I really got desparate, I could purchase a server and have it set up in a matter of hours anyway. If not for a long term, sustained pommelling, whatever do I need it for?

I have also decided to rethink my plans for the documentation project. I've got this lofty, web 2.0-esque idea on how I want it presented, linked to and from, etc. How new articles and documents get posted as news in a blog format, but get archived into a book type format, and have full site searching, automatic linking, banner suggestions, etc. It's all great, and I really think it'll look and work wonderfully. It's also taking a damn long time to implement. In the end, what difference does it make? I just want nice, organized, sharp looking pages that are appealing to those who are searching for them, and a profitable place for advertising.

My ultimate problem is that I've spent a lot of time focusing on how I can make this a functional system for people other than myself. No, I have no intention of marketing this as a tool for others to use in thier own documentation projects. This is all for this site. And I planned on being the only one writing documents anyway, with the exception of my wife. I wouldn't recruit anyone else, since my goal for a lot of this is to generate a bit of advertising revenue, and I can't really justify to anyone (including myself) having other people do work just so I can get paid for it. It also wasn't going to likely speed things up in the development process, although I somewhat deluded myself that it might be the case. But it won't. I know this now, and I've always known it. Any issue regarding the speed of development will 99% be involved with research and rhetoric. Layout, image manipulation, and indexing are activities that take so little time, relatively speaking, that I could do it manually almost as quickly as I would have something do it automatically, and it wouldn't require me to spend another year working on a system to do it to my satisfaction.

In any event, I'm going to go back to cranking out documents without any concern for the long term implications. My wife has her blog, and I will just have her write there and steal from it anything that would fit well into an independent document. We ARE a team afterall, no reason we can't work together on this. As new documents are generated, they will be referenced in the appropriate index file (the howto page, or some other page depending on their subject matter). I will first spruce up all of the documents I've historically had on this site, and finish all of the ones I've created a stub for but never done anything with. Then I will start on the long list of other documents I've been meaning to write for ages but never got started on.

And so, to begin, I've completed the Audio Streaming Howto. Enjoy!