Change color to default
Change color to red
Change color to blue
Change color to green
Change color to yellow
Change color to cyan
Change color to magenta
Change color to orange
Change color to b/w

Change color to default
Change color to red
Change color to dark red
Change color to blue
Change color to dark blue
Change color to green
Change color to dark green
Change color to yellow
Change color to dark yellow
Change color to cyan
Change color to dark cyan
Change color to magenta
Change color to dark magenta
Change color to magenta
Change color to dark orange
Change color to grey
Change color to b/w



On this page you will a collection of short notes about different pieces of hardware I somewhere had the possibility to try out, on which there was no documentation on the internet, and for which a short explanation is sufficient.

All comments, suggestions etc. are welcome. Please contact me at

polipo on openwrt does not cache to disk

Searching for a workaround took me quite some time, so I write it down here to remember it.

OpenWRT Chaos Calmer (15.05.1 Polipo polipo_1.1.1-1_brcm47xx.ipk Problem: I added a USB SSD to my OpenWRT router to have it cache to disk for my whole LAN instead of having to download everything for all clients or have a server running permanently as proxy server. Since I am using the same websites on different clients this should speed up things. In /etc/config/polipo I configured the cache section config polipo 'cache' option cacheIsShared '1' option maxDiskEntries '256' option dontCacheRedirects 'true' option chunkHighMark '10' option diskCacheRoot '/data/var/cache/polipo/' after an uci commit; /etc/init.d/polipo restart I checked the cache directory, but there were no files in it. After searching the internet for some time I came across this forum post. There is a bug in polipo that the disk cache is only activated if the Webserver root directory is set. Therefore I added in /etc/config/polipo config polipo 'general' option localDocumentRoot /data/var/www/ /data is the mount-point of my SSD, you can write here any existing directory (it can be empty), ran mkdir /data/var/www uci commit; /etc/init.d/polipo restart and checked the cache directory. Now websites were cached, everything was working. Links

Western Digital My Book World Edition II

The first problem was that the windows shares could not be accessed in windows. I couldn't reach the NAS via ssh, so I had to switch it off hard. After switching it back on I still couldn't access the shares and got the following error in the webinterface.

Warning: is_dir(): Stat failed for /DataVolume/_torrent_ (errno=5 - Input/output error) in /proto/SxM_webui/ctcs/ on line 24

I searched in google and only found descriptions to take out the hard disks, connect them to another computer and try to save the data. I didn't want to accept that fact, as long as I could ssh to the NAS and had a running Linux on the box.

So here is how to solve the problem step by step (at least how it worked for me):

Login as user root (not admin!). The default password is welc0me, this is not changed when you change the password for the webinterface.

You should see the mounted devices
In the normal admin account:
# mount
/dev/root on / type ext3 (rw,noatime,data=ordered)
proc on /proc type proc (rw)
sys on /sys type sysfs (rw)
/dev/pts on /dev/pts type devpts (rw)
securityfs on /sys/kernel/security type securityfs (rw)
/dev/md3 on /var type ext3 (rw,noatime,data=ordered)
/dev/md2 on /DataVolume type xfs (rw,noatime,uqnoenforce)
/dev/ram0 on /mnt/ram type tmpfs (rw)
/dev/md2 on /shares/xxxx1 type xfs (rw,noatime,uqnoenforce)
/dev/md2 on /shares/xxxx2 type xfs (rw,noatime,uqnoenforce)
/dev/md2 on /shares/xxxx3 type xfs (rw,noatime,uqnoenforce)
/dev/md2 on /shares/xxxx4 type xfs (rw,noatime,uqnoenforce)
/dev/md2 on /shares/xxxx5 type xfs (rw,noatime,uqnoenforce)
/dev/md2 on /shares/xxxx6 type xfs (rw,noatime,uqnoenforce)

In the root account:
/dev/root on / type ext3 (rw,noatime,data=ordered)
proc on /proc type proc (rw)
sys on /sys type sysfs (rw)
/dev/pts on /dev/pts type devpts (rw)
securityfs on /sys/kernel/security type securityfs (rw)
/dev/md3 on /var type ext3 (rw,noatime,data=ordered)
/dev/ram0 on /mnt/ram type tmpfs (rw)
/dev/md2 on /DataVolume type xfs (rw,uqnoenforce)
/dev/md2 on /ExtendVolume type xfs (ro,ikeep,noquota)

First check the status of the RAID just to make sure that the hard disks and the underlying RAID is OK.

~ # cat /proc/mdstat
Personalities : [linear] [raid0] [raid1]
md1 : active raid1 sdb2[1] sda2[0]
256960 blocks [2/2] [UU]

md3 : active raid1 sdb3[1] sda3[0]
987904 blocks [2/2] [UU]

md2 : active raid1 sdb4[1] sda4[0]
973322752 blocks [2/2] [UU]

md0 : active raid1 sdb1[1] sda1[0]
1959808 blocks [2/2] [UU]

unused devices:


Obviously everything is OK there, so our only problem is that the data partition cannot be mounted.

So we have to stop all services accessing /dev/md2 = /Data/Volume so we can unmount it and check the filesystem. Although the mount was not successful the system thinks it is mounted, although it cannot access the partition.

/etc/init.d/S55mini_httpd stop
/etc/init.d/S80nfsd stop
/etc/init.d/S90smbd stop
/etc/init.d/S95winbindd stop
/etc/init.d/S97twonkyserver stop
/etc/init.d/S9iTune stop
/etc/init.d/S91atalk stop

NB: The integrated mount/umount in busybox always needs 2 parameters, device and mountpoint. Only one parameter results in the Error 'Invalid parameter'.

Then unmount the partition so the system realizes it is actually not mounted (successfully).
If there is an error that the log has to be replayed first mount the DataVolume with

# mount /dev/md2 /DataVolume -o rw


# umount /dev/md2 /DataVolume -o rw

From now on you can mount and remount the Device as often as you want. Although mounting works let's check and repair the filesystem just in case it got some damages.

# xfs_check -v

then run

# xfs_repair

This is the kernel version this NAS is running:
~ # cat /proc/version Linux version (hung@localhost.localdomain) (gcc version 4.2.4) #1 Thu Apr 1 16:43:58 CST 2010

Max! eQ-3 Heizungsthermostat


Das Wochenprogramm lässt sich für ein Zimmer nicht auf die Geräte (in meinem Fall 2 Heizkörper-Thermostate) in einem Raum laden. Alle anderen Räume funktionieren.
Im Interface erhält man die Fehlermeldung 'Bei der Datenübertragung ist ein Fehler aufgetreten'.

Schritte bei der Fehlersuche

An der Funkverbindung konnte es bei mir nicht liegen, da die Thermostate in Sichtweite zum Cube installiert sind.
Ich habe die Thermostate aus dem Zimmer rausgelöscht und einzeln wieder angelernt, hat nicht geholfen.
Die einzig mögliche Ursache war daher, dass das Wochenprogramm einen Fehler hatte, sodass es die Thermostate nicht verarbeiten und speichern konnten.
Ein weiterer Hinweis darauf war auch, dass ich im Zuge der Tests dann auch Darstellungsfehler in der Anzeige des Wochenprogramms hatte, dass die farbigen Flächen links und rechts aus dem Bereich der Stundenanzeige für einen Tag hinausreichten.

Lösung des Problems

Ich habe den Raum (und damit das Wochenprogramm) gelöscht, die beiden Thermostate neu angelernt und einem neu angelegten Raum zugewiesen, ein Wochenprogramm erstellt und erfolgreich an die Thermostate gesendet. Danach funktionierte wieder.
Vermutlich hätte es auch ausgereicht, ein funktionierendes Wochenprogramm von einem anderen Raum an diesen Raum zu senden (bitte um Feedback, falls das bei Euch funktioniert).
Ein Reset des ganzen Geräts, wie man im Internet bei einigen google-Treffern liest, war bei mir jedenfalls nicht erforderlich.

Nokia S60 ' no gateway reply '

You get the error 'No gateway reply' when browsing over wifi / wireless lan?

For me it helped to enter the DNS server (although it should have been configured by DHCP anyways).

Go to Settings - Connectivity - Destinations - Internet - ESSID_OF_YOUR_WIFI - Options - Advanced Settings - IPv4 settings - DNS addresses and enter the IP Address of your DNS server (usually the address of your access point

After that it worked for me, setting the IP address of the phone by hand was not necessary.

Debian lenny - problem with keyboard in X / with xserver

After my dist-upgrade to Debian lenny setting the keyboard layout in /etc/X11/xorg.conf didn't work (on the console everything was fine). Probably this is related to the upgrade of the xkb-data package.
The fastest and easiest work-around is to add the following line to your .xinitrc / .Xsessionrc (of course adapting it to your correct keyboard layout):
setxkbmap -rules xfree86 -model pc105 -layout de -option nodeadkeys

Lexmark E210 GDI printer + Netgear PS121 printserver

With Windows

I wanted to have a small network print server in my LAN so I could switch off the computer the printer was connected to before.
I found information on some manufacturer's websites that their hardware wouldn't work at all with GDI printers, so I chose this thing from Netgear, because my printer wasn't on the incompatibility list (at least).

I bought the printserver, connected it to my LAN and ran the Setup-Program in windows. It found the printserver without problems and I assigned a fixed IP-adress.

Then I added a new (local) printer and chose the printer-share (LAN-Manager Printer Interface) of the printserver as the interface to be used. Then I finished the setup process and printed the test page - and it worked.

The other possibility is to use the IP-address as the interface instead of the printer share.
Select local printer, then TCP/IP-Interface, enter the IP-address, it will be converted automatically to IP_192.168.254.254, select LPR as the protocol (thanks for the correction :-) ) and enter lp as the name of the printer queue.

With Linux

With Linux it is just as simple.
Create a new printer in the CUPS web-interface, select LPD/LPR Host or Printer.
Enter as Device URI lpd:// (of course replace the IP-address according to your setup).
On the next screens select Lexmark and Lexmark E210 Foomatic/gdi (recommended) (en) as the Model/Driver.
Then it should work and you can print the CUPS testpage.

Some interesting links:

Large collection of hints for USB Print Servers

Omnibook XE3 GC hotkeys

Finally I upgraded my old laptop to Debian testing/lenny and kernel 2.6. So now I am using the great omnibook kernel module from Everything is working well, only the hotkeys took me some minutes' search and playing around.

The problem is that compared to the old kernel patches for kernel 2.4 you don't get the keycodes of the hotkeys directly to use them with the hotkeys package, instead you have to use setkeycodes (as root) first to map the scancodes.

The second and more severe problem is that keycodes higher than 240 cannot be used, so you cannot use the same keycodes as with kernel 2.4.

These are the error messages:

When you have loaded the omnibook module without setting any scancode/keycode (at least you see that the module works):
atkbd.c: Unknown key pressed (translated set 2, code 0xf2 on isa0060/serio0).
atkbd.c: Use 'setkeycodes e072 <keycode>' to make it known.

When you are setting scancodes/keycodes and use keycodes higher than 240:
setkeycodes e073 243 e074 244 e072 242 e071 241
Error message:
keyboard.c: can't emulate rawmode for keycode 242

Solution: use for example the following keycodes (check first with getkeycodes wheter these keycodes are not use yet):
setkeycodes e073 233 e074 234 e072 232 e071 231

The keycodes used for hotkeys in the .def-file are the following (small blue one touch keys from left to right):
<userdef keycode="142" command="firefox">Firefox</userdef>
<userdef keycode="213" command="xawtv">xawtv</userdef>
<userdef keycode="228" command="Eterm">Eterm@localhost</userdef>
<userdef keycode="218" command="Eterm -T Eterm@remote -e ssh remotehost">Eterm@remote</userdef>

Some interesting links:

Omnibook kernel module


Written 2007-02-07 - last update 2007-11-12

Change color to defaultChange color to redChange color to blueChange color to greenChange color to yellowChange color to cyanChange color to magentaChange color to orangeChange color to b/w

© by wf
webslave (at)
Wolfgang Fuschlberger, Salzburg