PJ Walters started the Twins off, pitching against the Phillies last night. He left the game after pitching to only four batters (all of whom got hits and eventually scored). His shoulder wasn't working right, and today he's going to get an MRI to see what's wrong. At the end of the inning, the Phillies had run up six runs.
I think I've got them all now. It seems my two-stepping test of earlier combined with some browser caching to give some failed successes.
In a little bit of a face-palm moment, I found and corrected the problem with my HTTPS redirection attempts.
In an effort to re-acquaint myself with the difficulties associated with running an SSL site, I've added HTTPS to this one. If you've arrived using an old link, you should be redirected to the new one that matches the certificate. In some cases, the redirection is failing, and I'm working on it; apologies if you're one of the afflicted.
Once again, an update and reboot stopped sendmail from working by deleting the /etc/init.d/sendmail script. Further, something's broken in makemap, so I can't adjust any alias, access, or other map sendmail needs to work. The way I see it, I can either outsource my e-mail, move to something package-managed (including a different MTA), or continue to deal with these occasional outages.
It seems to be the case that the break occurred on either 23 or 24 May, as there were no messages processed (according to the log files) on 24 May. Apologies if the outage affected you.
The Mac had just one update today. 700MB of v10.7.4 Lion updated goodness. The update was supposed to contain the Safari update, but the first thing the Mac did after rebooting from the Lion update was trigger an update for Safari; maybe that was the designed process. It seems the only visible change is that my tweaked background for the login page was reverted to the grey slate of the default Lion. Now to find the bit I wrote about how to do that again...and maybe to find a new splash image.
It seems the new Squid proxy server listens on a different port. Although it started up, it wasn't responding to client requests. A quick configuration change and all was fixed.