Site was down for a bit around 2330 CST, with a mysql error ("too many connections"; maybe the database was down or inaccessible?) Clearly ABC does not appreciate complaints.
I went for a walk and it was back up when I came back, so I'd like to take all credit for it working now. Also, I've alerted the authorities and sounded the funk alarm.
Originally posted by thecubsfanSite was down for a bit around 2330 CST, with a mysql error ("too many connections"; maybe the database was down or inaccessible?) Clearly ABC does not appreciate complaints.
I went for a walk and it was back up when I came back, so I'd like to take all credit for it working now. Also, I've alerted the authorities and sounded the funk alarm.
OK, SO.
I have no idea what happened. It was working when I left!
The logs say our MySQL database was inaccessible from 23:35 until I stopped and restarted it at 23:54. Apache was never down, so during those 20 minutes a lot of you saw the "we're up, but we're not up" message. And NONE of you told me!
Also, it looks like the new best way to find me is to invoke my name on Twitter. (You may have noticed that Cubs is pretty passive aggressive, though)
I was able to ssh in remotely from my unbelievably amazing phone and do the do, but I wasn't able to check out the logs until now - well, to be more precise, I didn't feel like futzing around on my phone's keyboard any more than I had to, so I waited until now to check. (Also, I made Cubs make the announcement. Thanks, Cubs!)
The MySQL process HAD been running for months (dunno how long, but the server said "2009" so, you know, at least four months and probably quite a lot more) so maybe we hit some sort of overload there. Or maybe we've been hacked and I haven't noticed it yet. (I'll double check.)
On the plus side, I totally take credit for Cubs having to get off his (could be, how would I know) fat ass and take a walk, so there's that.
On the minus side, I'm not sure the other guy even noticed we were down, since he posted that bizarre thing about stocks nobody owns but nothing about us being down...
It was actually down-up-down right around 31.3.10 0243.27 (my time) too. I got the error, gave it a couple of minutes to self-resurrect, got in, then got the error again. Since I was just fixing a typo, I didn't try again.
Should I have attempted to get the funk out at that point?
Originally posted by emmaIt was actually down-up-down right around 31.3.10 0243.27 (my time) too. I got the error, gave it a couple of minutes to self-resurrect, got in, then got the error again. Since I was just fixing a typo, I didn't try again.
Should I have attempted to get the funk out at that point?
No, THAT time was ME trying to reproduce the error. Although I apparently succeeded, I'm not sure HOW I succeeded.
It didn't help that the server was already tied up doing our daily backups at the same time, which I didn't notice, because I'm not usually up at 0330.
ANYWAY. At that point, I stopped and restarted BOTH MySQL and Apache and everything was right as rain once again. It probably, hopefully, definitely won't happen again. Maybe.
Originally posted by Guru ZimI'm not sure what's going on.
They are failed hack attempts which return no sensitive information, but nonetheless do take down the MySQL server during the attempt. I'm configuring some new 403s to head 'em off at the pass.
Thread ahead: Facebook Like button takes over the Internet Next thread: thread "tag" icons in sort results? Previous thread: the-w.com topics of interest: March 2010