Did rather a lot of cleanup of the code today. Finished a number of "polish" elements (e.g. links upward in the hierarchy, retaining hierarchy names and comments through upload/download cycles, that kind of thing). Confirmed that email alerts do work when the email account is set up.

Last task of the day was sitting down to learn matplotlib (which is ridiculously easy to install on Windows, of course (Tim went to extreme lengths just to get it installed on cmon)). Fairly straightforward, though I have to do some hacking to have it give me back the SVG as strings instead of saving it to files. Haven't yet hooked it up to produce real-world graphs, as I'm still in the spike-testing stage.

Also spent a few hours spelunking through libsnmp.py as a possible speed-fix. Unfortunately, it doesn't have a get bulk request.

Have to sit down tomorrow and figure out how much power we need for a mid-sized plant. I'm tempted to say "whatever we can afford", but I know that multi-CPU isn't likely going to help us, so a single-CPU machine with scads of RAM seems like the best bet.

In a fit of faint hope I'm running the scanner all night to see if all the changes from the cleanup and audit have helped. It seems a pretty forlorn little hope. I didn't find anything even remotely incriminating. Did do an enormous amount of cleanup though. Probably dropped the lines-of-code count by 1/2, though I didn't check in most of those drops (the code may be useful some day, I tell myself).

Absolutely must go shopping first thing tomorrow. Rosey and I have been without any reasonable food in the house for... well... I've actually lost track of how long.

Oh, one thing I've just realised; it's possible that we're seeing a slowdown from not specifying the ... [hmm, forgot to post this earlier as I ran off to see if this could be done... cool, I can use the set-a-date feature for the first time]