Yes once again I had nothing better to do then move servers, fuck and today it wasn’t even raining. Ok I came across a lot of problems as usual, one was that I decided not to use debian testing anymore, I wanted only stable ! One of the problems was that stable had mysql4.0 and my old database was 4.1 and of course there where problems while trying to import the database, but hey, there is a nice parameter –compatible=mysql40 sounds like exactly what I needed … After I finished the server almost I found out what “compatible” means for the sickheads who implemented this parameter. “compatible” means that this line: `object_id` int(11) NOT NULL auto_increment, will be transformed into this: `object_id` int(11) NOT NULL default ’0′,

Finaly ! I have waited soo long for this one

I’m happy to write this because video has been the number one thing requested in Skype for Mac since… well, pretty much ever since we released Skype a few years ago.

I should think twice before doing stuff with lvm on a raid ! I am setting up a new server, rebooted it with a rescue system so I could create the raid 1, this was quite simple mdadm –create …. and raid was finished, then the lvm over it, pvcreate, vgcreate, lvcreate and ready to go … but alas I had to reboot the rescue system ( shit happens ) after the reboot ( since it was a rescue system which is read only ) the raid didn’t start, but I could start lvm which of-course complained that things are double ( on sda and sdb, because of the raid) I actually ignored that and started installing xen, after a short time I noticed what big shit I did. so … I stoped lvm again, starte raid ( mdadm –assemble /dev/md0 /dev/sda2 /dev/sdb2 ) and the raid is reconstructing. I have NO clue what it will look like after it’s finished, no idea where the lvm acutaly write to ( sda or sdb? ) and if after the rebuild of the raid the data will still be there … maybee I should just start from scratch.

oh no, my internet is not working today, blizznet seems to have some DNS problems, almos al .at domains are being resolved, and .. heise.de …. strange. anyway, this gives me some time to relax and forget about computers, go smoke a cigar, drink some cuban rum and write a blog. write a blog? without a internet connection? well thanks to my neighbor which hasn’t set up his wireless router correct I do have internet but I won’t abuse it, I will only use it to read my mails and write this blog, I promise

oh my good …. I never would even imagine that I would do more the 100gigs until the end of the month … and today is only the 07th !!!

ipacsum -s 200607010000
IP accounting summary
Host: foo.bar / Time created: 2006/07/07 12:48:36 CEST
Data from 2006/07/01 00:00:00 CEST to 2006/07/07 12:48:36 CEST
  forwarded incoming GREEN (eth0)                 :             75G
  forwarded incoming RED (eth1)                   :             44G

this means 119 Gigs of transfer this month ! Blizznet is going to kill me for this

yes, Blizznet has offered it’s customers as a kind of “thank you” a real flatrate for July and August and this with normal Blizznet speed ( 10Mbit ) The result after the first weekend ( Today is the 3rd as some might have noticed ) is this:

ipacsum -s 200607010000
IP accounting summary
Host: foo.bar / Time created: 2006/07/03 13:38:19 CEST
Data from 2006/07/01 00:00:00 CEST to 2006/07/03 13:38:19 CEST
  forwarded incoming GREEN (eth0)                 :             23G
  forwarded incoming RED (eth1)                   :             13G

that would be 13G of download and 23G of upload … in only 2.5 days!

I should know it better, especially that this isn’t the first time it happens. I installed once again after a long time nagios on my server so I have something that keeps an eye on the server. Anyway … I installed nagios-mysql ( as I did the last time ) but it looks like the database has to be created before installing, or … who knows, I still didn’t RTFM. well the result was that the 5 min. nagios was running I dumped about 2 gigs of syslog messages that where saved in the syslog, message, users AND the nagios.log … I had over 8 Gigs of logs, good performance test for syslog-ng by the way. Anyway now I finally cleaned up everything, as .gz the file isn’t that big at all only 8.5 MB, OK now that logrotate is finished I should get back to my server.