I’m going to talk technical neepery here, so if that doesn’t interest you, you’ll probably just want to wait for the next entry.
So, yesterday marked a year of using WordPress, and generally speaking it’s been a pretty positive experience. The last month I’ve been encountering problems with internal server errors, but I have a more than sneaking suspicion that this relates to my host provider not to WordPress itself. Some of you may recall that I found out some time ago that the reason I was having so many problems with Movable Type here last year was because 1&1 put a time limit on the length scripts could run, which inevitably caused problems with the software. I rather strongly suspect that something similar is happening again, as there’s no other reason the site should be choking and crawling as much as it is — the last month was down about 7% in terms of traffic from some of the earlier months, and had no “big posts” that would have caused congestion. So it’s not the software, I suspect. It’s the hardware.
This pretty much confirms that (dramatic but real sigh) I’m going to have to find another hosting home for Scalzi.com sometime in the near future. I do have some interesting options, which I will be pursuing, and which could possibly ultimately be of benefit to me and the site. But it’s still a pain in my ass, and it’s a shame, since up until the last month or so I’ve been generally happy with 1&1. But me having server errors every single day for the last month and them unable or unwilling to do anything about it is just not acceptable.
I don’t know that this should mean anything to you in the short run, since I don’t intend to make any moves on the site until I know where I’m going and until I can build out the backend so that when the domain transfers, you the reader have a fairly seamless experience with it (i.e., no wondering where all the content went, etc). That takes time and effort, and as long as I’m going to do all that, I might as well try to do some improvements on the site as well. So don’t expect massive changes instantly, but do expect some changes, possibly by the end of the year.
In the meantime, I’ll be doing what I can to lessen calls to the database, which will probably cut down the server errors a bit. The first step is that for the short term at least I’m going to close comment threads older than 14 days. 99% of comments on a post happen within the first three or four days anyway, so by and large I don’t think people will have a problem with this. But it will make a difference in the backend, since then people can see cached pages. There are other things I will tweak as well, if I can. Basically, I want to make sure you don’t have a problem getting to the site and working with it. In the short run, this means throttling the back end as much as possible. Alas.
So that’s where we are on the tech front of things.