Two thinks happened at the same time. I had again problems with my Samba/SMB setup and found and old article about OpenAFS.
AFS still sounds good but most of the advantages are not important on a home network. Also all the good features are now part of NFSv4. As it is still not widely used it probably is more difficult to find a already existing solution on the net. This is also true for all the other protocols found on wikipedia.
Therefore it still comes down to SMB or NFS for me. As the new file server I build will use ZFS and run on OpenSolaris NFS might be the way to go. At least I will give it a try and see if I have less problems or run into other disadvantages compared to my use of SMB.
Montag, April 20, 2009
Samstag, April 11, 2009
output caller as Morse code
In CQ DL 4-2009 was a entry about CW for Windows Mobile.This will play information about the caller as Morse code. As I don't have a Windows Mobile phone, I searched for a program like this for my phone. But I couldn't find one.
There is a new phone project for me. The above program started with WAV snippets and then a MIDI file. He was not satisfied with the result and generates now the whole WAV file to play.
Never the less I will start generating a MIDI file. This should be the easiest way to start. Finding the inforamtion about MIDI was not difficult:
I found also a page that generates Morse code via MIDI. The perl script that dose the work is available for download. This will be the starting point for my own program (perlscript repository).
The first attempted, to find information on how to run a program if the phone receives a call, was not successful.
Copying a MIDI file onto the Centro is described here.
During my search I found the following interesting page about learning Morse code with the Koch method.
There is a new phone project for me. The above program started with WAV snippets and then a MIDI file. He was not satisfied with the result and generates now the whole WAV file to play.
Never the less I will start generating a MIDI file. This should be the easiest way to start. Finding the inforamtion about MIDI was not difficult:
I found also a page that generates Morse code via MIDI. The perl script that dose the work is available for download. This will be the starting point for my own program (perlscript repository).
The first attempted, to find information on how to run a program if the phone receives a call, was not successful.
Copying a MIDI file onto the Centro is described here.
During my search I found the following interesting page about learning Morse code with the Koch method.
Sonntag, März 15, 2009
start zfs pool with one disk
Some first test using the OpenSolaris live CD in VirtualBox. It helps testing the things directly you read about.
zpool create mypool c3d0
start pool with ony one disk
Created in /mypool a directory and some testfiles.
zpool attach mypool c3d0 c3d1
add a second disk as mirror
zpool detch mypool c3d0
now only the second drive exists in the pool
the data was mirrod to this drive
after reboot "mount" the pool with
zpool import mypool
use -f if the pool was not properly "exported" before
zpool create mypool c3d0
start pool with ony one disk
Created in /mypool a directory and some testfiles.
zpool attach mypool c3d0 c3d1
add a second disk as mirror
zpool detch mypool c3d0
now only the second drive exists in the pool
the data was mirrod to this drive
after reboot "mount" the pool with
zpool import mypool
use -f if the pool was not properly "exported" before
Sonntag, März 08, 2009
some new Java build tools
Reading an article about The future of Java build tools, I started to think what this means for me.
So far I needed only to modify build processes using Makefiles and perl scripts. Our "new" (4 years old) company project I'm involved, in uses ant for the complete build. Down the chain where my code (C++) is located, still Makefiles are used.
On Java projects at home, eclipse is doing all the work for me.
Some of the new build tools define the rules in a script language like Gradle in Groovy or Raven in Ruby. Looking deeper into one of this makes only sense for me in the case
This build tools are very new and only time will tell if the concept of defining the build rules in a script language will succeed. For me it makes more sense to use a declarative build rule description every one knows and add if needed scripts in a language also every one in the projects knows.
My conclusion is to look deeper into ant as we use it in the project at work. It might also get handy in my private Java projects with eclipse.
Some wikipedia links to start
So far I needed only to modify build processes using Makefiles and perl scripts. Our "new" (4 years old) company project I'm involved, in uses ant for the complete build. Down the chain where my code (C++) is located, still Makefiles are used.
On Java projects at home, eclipse is doing all the work for me.
Some of the new build tools define the rules in a script language like Gradle in Groovy or Raven in Ruby. Looking deeper into one of this makes only sense for me in the case
- we use one in a project
- I know already the used script language
- it becomes the successor of make and ant
This build tools are very new and only time will tell if the concept of defining the build rules in a script language will succeed. For me it makes more sense to use a declarative build rule description every one knows and add if needed scripts in a language also every one in the projects knows.
My conclusion is to look deeper into ant as we use it in the project at work. It might also get handy in my private Java projects with eclipse.
Some wikipedia links to start
Montag, Februar 02, 2009
Darlington-Arrays ULN2x0y
To easily switch loads from a microcontroller the Darlington-Arrays ULN2803 from SGS-Thomson are used. But there exists different types like ULN2003 or ULN2801. What are the differences?
ULN200y has 7 (16 pin) and ULN280y (18 pin) has 8 switches. Both have a integrated suppression diode for each switch. Each switch is able to provide 500mA.
The "y" is for the different logic families:
There are also the ULN2064, ULN2066, ULN2068, ULN2070, ULN2074, ULN2076. This contain only four Darlington switches, but are able to switch up to 1.5A at 50V.
ULN200y has 7 (16 pin) and ULN280y (18 pin) has 8 switches. Both have a integrated suppression diode for each switch. Each switch is able to provide 500mA.
The "y" is for the different logic families:
- ULN2x01
- General Purpose, DTL, TTL, PMOS, CMOS
- ULN2x02
- 14-25V PMOS
- ULN2x03
- 5V TTL,CMOS
- ULN2x04
- 6–15V CMOS, PMOS
There are also the ULN2064, ULN2066, ULN2068, ULN2070, ULN2074, ULN2076. This contain only four Darlington switches, but are able to switch up to 1.5A at 50V.
Abonnieren
Posts (Atom)