Page head image

Viewing posts tagged computers

Maschin kaputt

Just a quick heads-up: My co-located machine martes.yaina.de keeps crashing after a reboot. I'm going to move to a rental machine ASAP (on Monday at earliest,) but it will take a couple of days until everything is working again. The current plan is:

  • Retrieve the broken machine from the data center
  • Recover the harddisk
  • Set up the base system on the new machine
  • Set up DNS (Secondaries: I'll send you an e-mail with the new IP number)
  • Restore e-mails
  • Set up e-mail services
  • Restore web services
  • Restore everything else
I'll send an update to all users as soon as service is back to normal.

One reason why I moved away from OX5

I don't understand much about SQL, but even I know it's not supposed to be done like this...

CREATE TABLE prg_knowledge (
creating_date timestamp with time zone,
created_from text,
changing_date timestamp with time zone,
changed_from text,
user_right text,
group_right text,
sid text,
tid text,
order_crit text,
timestampfield01 timestamp with time zone,
timestampfield02 timestamp with time zone,
intfield01 integer,
intfield02 integer,
intfield03 integer,
intfield04 integer,
intfield05 integer,
intfield06 integer,
field01 text,
field02 text,
field03 text,
field04 text,
field05 text,
field06 text,
field07 text,
field08 text,
field09 text,
field10 text
);

And that's just one of numerous tables where all of the relevant information (including the relations to other tables) is stored in intfield* and field*. Granted, OX6 has fewer tables like this. But it doesn't work for me. And there is no migration tool from the old open-xchange version to OX6, thus I would have had to do what I had done today anyway: getting data from the old database and convert it to something that can be imported by the new system. At least the data base schema of Horde is much better than the OX one.

More breakage...

I drove to the data center yesterday. The machine didn't even shutdown. All fans running, memory okay...Okay, no hardware issue after all. In fact, it was a broken maintenance script which was supposed to set up the chroot environment for postfix. Instead, it copied everything to the root directory and changed the permission of /tmp to 755 and ownership to user "postfix". Thus, a couple of processes just crashed. At least the machine continued to process incoming e-mail. Well, almost exclusively.

Making MacOS X usable

Luckily, MacOS X is based on some kind of BSD, in other words: it is a Unix system. It even comes with X11. However, as a desktop system many tools I like and need aren't present. I already had an ancient version of Darwinports running since 10.2. That old installation was terribly broken, though, and almost nothing compiled anymore on 10.4. Thus, I removed it and started again. There are several alternatives available, two of them being based on (Free)BSD ports, and one on Debian packaging.

yaina.de is back

The machine running yaina.de crashed twice within the last two months, so I decided to do some maintenance work. At the data center which is housing the system I discovered that my suspicions were right: Indeed, the CPU fan didn't move, and the power supply fans were stuck as well. Hence, I decided to take the machine home and fix it. First problem: I could not get the heat sink from the CPU (which I had to remove to get the fan off the heat sink),instead the CPU came out of its socket. Luckily, a colleague showed me how to get it off: take the blade of a box cutter, place it on the gap between the heat spreader and the heat sink and hit the blade carefully with a hammer. (Naturally, this will only work with chips that have a metal heat spreader plate. Try at your own risk.)