News Archive

Older news and announcements

2013-07-18: leapingtiger.org DNS migration

This is it, I have begun migrating the leapingtiger.org domain from Gandi to hisoka. Until I get the configuration right and DNS servers update, the old domain may not be available. Please use tigris.fr instead (you really should anyway.)

— Kefen
2013-07-10: Now turning to DNS

Now that I’ve had my fun with HTTPS, I’m going to go and get my hands dirty with DNS/Bind. The good thing about this is that I’ll be testing stuff on leapingtiger.org which, I believe — well, I hope — isn’t used very much anymore.

The bad thing is that it may eventually mean that we’ll handle our own DNS zone for tigris.fr as well. Still, considering that most of the sites hosted on hisoka are using their own domain, this shouldn’t be much of a problem either.

That being said, as usual, sorry for any inconvenience this may cause.

— Kefen
2013-07-03: Websites outage

I’ve been tinkering with SSL and Lighttpd, which resulted in the latter stopping for about 45 minutes this morning. My bad! All is back to normal now — HTTPS is technically functional but we need a proper certificate before I actually use it.

Sorry for the inconvenience.

— Kefen
2013-06-27: Switching to virtualenv

After the last system update, some parts of the tigris.fr website started acting up (only in the administration section) so I moved the whole Django app powering the site to a virtualenv project, as advised by the great article about How To Properly Start a Django Project I only discovered recently.

Now that I’ve remembered to reinstall rstats inside that project *cough* *cough* the whole site should be back to normal, happily using Django 1.3 because I can’t be bothered to upgrade it all to 1.5.

As a side note, we are nearing 900 days of uptime, and this is ludicrous, and I love OVH! I have half a mind to migrate (again) to another flavor of Kimsufi server to expand disk space and slightly reduce server costs (or compensating with an extra IP) but that won’t be in the near future.

— Kefen
2012-07-02: High CPU use for mysqld fixed

Turns out mysqld using 100% of CPU was something due to the very recent leap-second and NTP not playing nicely.

Temporary fix was used, CPU usage is back to normal, I’ll keep an eye on it. MySQL-based websites should still be working properly, and there should be no more interruptions for now.

— Kefen

“Now, we’ve basically got it all worked out, except for small stuff, big stuff, hot stuff, cold stuff, fast stuff, heavy stuff, dark stuff, turbulence, and the concept of time.”

― Zach Weinersmith