Martin's questions

When booting my personal code server (Debian Squeeze), the boot screen freezes almost immediately, but when I press a random key (including useless keys such as NumLock) the boot process continues. This happens several times during boot - a random process will stop, continuing only after I press a key.

Once GDM shows up (requiring at least another key press) and I log in, the mouse changes to the "waiting" icon, stopping a couple times until Gnome starts up - from then on, the computer works normally... unless I try to switch to/from a console, in which case I have to (yeah, you guessed it) press a key. This also happens while using gnome-terminal (extra diagnosis: if I try cat /proc/cpuinfo, for instance, the command will hang the first time, but after that it will work flawlessly). This also happens when logging through ssh - if I issue a command on my netbook, and press any key on the server keyboard, the command will get unfrozen immediately.

Finally, this happens too when shutting down the computer (it will show something as "Shutting down Samba", but it continues only after I press a key).

Any ideas about why could this be happening?

Important detail: Sometimes if I don't press any key, after about 15 seconds the system continues doing whatever it was trying to do at the moment (particularly noticeable through ssh, when I'm not near the keyboard and, thus, have no alternative but waiting). This would also explain why Gnome is not as fast as it should (slight delays showing menus, switching programs, and so on).

Extra info:

• The computer has a 2.6.32-5-amd64 kernel.
• The video card is an ATI Radeon, but the problem was present both before and after configuring xorg.conf to properly recognize it. Both the video card and the motherboard are Gigabyte.
• The keyboard is PS/2 (I tried another PS/2 keyboard with no success, and I don't have any USB keyboard available). Plus, ...
• The computer works fine under Windows XP (32 bits).

I have Nagios 3 on Debian Lenny up and running, but trying to install the NagiosGrapher plugin has proven to be frustrating to say the least - the plugin doesn't work by itself, there seems to be close to no documentation about how to make it work under Debian, the service refuses even to start (it throws some Perl-related errors instead), the official installation guide included in the package is a 1-line file pointing to a non-existent directory and my chair is uncomfortable (that last thing may be unrelated, though).

I tried searching online, without luck. While I did found a couple guides, all of them start with "download the source code from...", but I really wish I could install it the "Debian" way if possible - don't forget that, since those guides are for non-Debian systems, the files are located in different places, making every step harder.

Anyone knows where can I find an installation guide/howto? Or at least a first-hand experience of painless installation under Debian with source code?

Where I work at, we have a bunch of Cisco Catalyst 3560 switches (plus two routers), which no one knows how to administer (the guys who installed them, it turns out, don't like doing tech support for free), so I'm stuck trying to figure out how to do even the simplest of tasks.

Currently, I'm trying to find out which workstations are sucking our bandwidth, and the Port Statistics interface seems like a good start (too many packets =~ lot of traffic). However, while I can see the statistics in the Device Manager (the web interface), I can't find the command-line version of said statistics, which I need if I want to write a script. I have the root password for the switches, so I can ssh into the devices, but I can't find which commands are the ones I need.

So far, I've tried:

• show interfaces counters: I realy hoped this one would work, but it seems those are statistics of dropped packets, not regular traffic, so it doesn't work
• show interfaces: This one actually shows a "5 minute input/output rate" field, but 5 minutes is too short - after 5 minutes of downloading, the offenders would be free to go again
• Copy and paste the information from the web interface, one switch at the time, pasting them in an Excel spreadsheet and working from there. That works, but every time I do that by hand I die a little inside

Any ideas? In case you ask: yes, it would be smarter to check this in the router instead of the switches, but I know nothing about Cisco, so toying around with the router scares me like a little girl (I do know about regular networking, though), and yes, there is a proxy on the way, but those things take time (check the "How do I plug a proxy into a Cisco router?" question next month).

everybody! Maybe you can help me with this: I have set up my daily backup rotation in Cron to run every day at 4 am. The script is not complicated (see below), and when I execute it myself it works without problem. However, when I schedule it on Cron, the script runs for about half an hour and then stops (no mail, no warning, it just stops copying files at some point).

Any ideas about why this could be happening? The backup folder is about 38Gb big, and it takes almost an hour for the script to run. I'm guessing some kind of run-time limitation, but is just an idea.

Last, but not least, this is the script. It keeps an "archive" folder with 7 daily backups:

#!/bin/bash
NEW=`date +%Y%m%d`
NUMBER=7

function erase_old()
{
    while (( $# > $NUMBER )); do
        rm -rf "$1"
        shift
    done
}

mkdir /backup/archive/$NEW
cp -vrp /backup/current/* /backup/archive/$NEW
erase_old /backup/archive/*

At my current work I have this situation: every time we install some new software (9 out of 10 times the app is written in PHP), it is installed in its own server, consisting of a Debian OS, an Apache server and a PostgreSQL database. Because of this strategy, we currently have about 10 servers (note: when I say server, I say "regular PC without a screen", and not "we spent enough in this server to feed a small country"), most of them sharing the same software requirements.

This situation lead me to think: shouldn't we create a cluster, instead of adding individual servers for each app? I can attest that, while some servers need better hardware (RAM, mostly), some other ones have been idle for no less than three weeks, so with a cluster we could use those resources better. At the same time, I'm not sure the results are worth the time and effort needed to put together and mantain the aforementioned cluster (I always wanted to use the word "aforementioned").

After all this intro, the main question is: Would you suggest, in this situation, creating a cluster? Or would you keep the servers as they are now?