anyway.. i found a nice solution to nearly all the proxy auth problems out there, cause usually a proxy without any authentication should work without any problems in nearly all cases

check out the cntlm authentication proxy website, download the small package, unzip it somewhere and run the installer script.

now you just need to put the correct settings in the config file.
in the configuration hints paragraph on the cntlm webbie you’ll find the following guidance:

Configuration hints

After installation, you have to locate the configuration file. The default for Linux packages is /etc/cntlm.conf, for locally compiled source distribution (“./configure; make; make install”) it’s /usr/local/etc/cntlm.conf and for Windows installer it’s %PROGRAMFILES%\Cntlm\cntlm.ini (usually X:\Program Files\Cntlm\cntlm.ini, where X is your system drive).

When you have found it, fire up your favourite editor (not a word processor) and open the file. First a few rules, though – lines beginning with a hash, #, are comments: completely ignored. There is no required formatting and option names are case insensitive. Option values are parsed literally: a quote means a quote and is taken as part of the string, so do not quote, escape, etc. Anyway, you need to set these core options:

* Username - your domain/proxy account name
* Domain - the actual domain name
* Workstation - NetBIOS name of your workstation; Cntlm tries to autodetect it, but you might want to set it explicitly should dialect detection fail (see below)
* Proxy - IP address (or ping-able hostname) of your proxy; if you use several alternative proxies or know of backup ones, use this option multiple times; if one stops working, Cntlm will move on to the next
* Listen - local port number which Cntlm should bind to; the default is OK, but remember you can’t have more than one application per port; you can use netstat to list used up ports (lines with LISTEN)

Next, we need to find out which NTLM dialect your proxy understands. It’s a jungle out there and it can be quite challenging (i.e. boooring) to find a working NTLM setup – thank Bill. Good thing Cntlm has this magic switch to do it for you – thank me. Save the configuration and run the following command; when asked, enter your proxy access password:

$ cntlm -I -M http://test.com
Config profile 1/11… OK (HTTP code: 200)
Config profile 2/11… OK (HTTP code: 200)
Config profile 3/11… OK (HTTP code: 200)
Config profile 4/11… OK (HTTP code: 200)
Config profile 5/11… OK (HTTP code: 200)
Config profile 6/11… Credentials rejected
Config profile 7/11… Credentials rejected
Config profile 8/11… OK (HTTP code: 200)
Config profile 9/11… OK (HTTP code: 200)
Config profile 10/11… OK (HTTP code: 200)
Config profile 11/11… OK (HTTP code: 200)
—————————-[ Profile 0 ]——
Auth NTLMv2
PassNTLMv2 4AC6525378DF8C69CF6B6234532943AC
————————————————

You see, NTLMv2 – I told you to use it, now it’s official. BTW, here you can see all tests running – it’s just for demonstration purposes. Normal version finishes when it finds the first (i.e. most secure) working setup.

When you get your options (might be more than just Auth and Pass* here), remove all previous password settings and paste the profile into the configuration file and save it. (Re)start Cntlm and it should work. To use it in your applications, replace the old proxy settings with “localhost“, port same as you chose for Listen.

This was just a simple 101 lesson to help you kick-start the proxy. You should still RTFM.

many thanks to the cntlm team for developing such a smooth running and easy-to-go solution!

btw: if you are using cygwin on your windows system and put the cygwin/bin directory to your path environment variable for easy integration, you should delete the cygwin1.dll delivered with the cntlm application – usually found under c:\programme\cntlm – or at least rename it and check whether everything works this way.

so i installed the restful_authentication plugin, that let me generate my user model… further on i installed the role_requirement plugin, but when i tried to do the ruby script/generate roles Role User i got the following error:

No such file or directory – …/app/controllers/application.rb

after some investigation i found the cause and solution to the problem..
beginning with rails version 2.3 the the name of the application controller changed from application.rb to application_controller.rb, and it seems like the role_requirement plugin still misses the needed change in one of its generators.

to solve the issue, just update line 52 in …/vendor/plugins/role_requirement/generators/role_generator_helpers.rb to point to application_controller.rb instead of application.rb and the generator will do the job for you again.

a little addon:
if your project is created before rails version 2.3 and you get the following error in your browser:

…uninitialized constant ApplicationController…

you can either rename the application.rb to application_controller.rb manually or just run the command rake rails:update:application_controller.

i found an explanation in the eclipse wiki and a workaround for the problem aswell:

Like other parts of ECF, the filetransfer API has a provider architecture, allowing multiple implementations to be used for filetransfer (i.e. via http, ftp, and others). In ECF 3.0/Eclipse 3.5 the primary provider is based upon Apache httpclient 3.1. This was introduced in the ECF 3.0/Eclipse 3.5 cycle because the previous provider that was based upon the JRE URLConnection implementation proved insufficiently reliable (i.e. see bug 166179).

Unfortunately, the Apache httpclient implementation, although more robust than the URLConnection-based provider, does not support NTLMv2 proxies directly (for an explanation of why, see here).

so just add the following line to your eclipse.ini and the proxy settings should work again.
-Dorg.eclipse.ecf.provider.filetransfer.excludeContributors=
org.eclipse.ecf.provider.filetransfer.httpclient

in the last weeks i had some spare time, so i installed the rubystack package from bitnami and played around a little bit.
first i upgraded ruby, gems, etc..
after that i started a project and tried to create the databases.

and booom… the first thing went wrong
when i wanted to do a db:create:all, it failed with the following error msg:
Rails requires RubyGems >= 1.3.1 (you have 1.2.0). Please `gem update --system` and try again.

seems the command gem update --system doesn’t work on all versions of ruby gems. when you run the command you simply get a message, that theres nothing to update. but bless google, i found a work-around on the following site.

here the steps to follow:
1. Open a terminal window and change (cd) to the JRuby or Ruby bin directory.
2. Type ruby -S gem --version
If the version is 1.3.1, then that installation has already been updated and you are done. Otherwise, continue.
3. Type ruby -S gem install rubygems-update
4. Type ruby -S update_rubygems
5. To ensure that the update happened, type ruby -S gem --version

after the fix, it wasn’t a problem anymore to create the databases and i added some models and controllers aswell… but more on that in the next days.. and weeks

links:
http://blogs.sun.com/divas/entry/fixing_the_rails_requires_rubygems

Sheep overflow!

(via)

so here the steps i went along to get things working again:

i downloaded an up-to-date version from
http://ftp.gnu.org/non-gnu/cvs/binary/stable/x86-linux/RPMS/i386/
and installed it on the new machine with
rpm -i cvs-1.11.18-cvshome.org.9x.1.i386.rpm
after on you can check if installation works, with
cvs -version

add a new group: cvsusers
groupadd cvsusers
add a new user: cvs
useradd cvs -g cvsusers -c CVS-Owner -d /home/cvs
create a new directory for the repository: /home/cvs/repository
mkdir /home/cvs/repository
and set it as CVSROOT:
CVSROOT=/home/cvs/repository
export CVSROOT

switch to cvs user and init repository:
su - cvs
cvs init

copy resp. create passwd file in CVSROOT.
copy resp. create readers file in CVSROOT to add the read-only users. (that’s the “black list” way… you can go the “white list” way aswell, you just have to add the writeable users to a writers file. more infos available here)
a quite nice addon perl-script for adding users easily, you can find in the mentioned adminguide.

configure xinetd.conf for using pserver (just change to the servers ip adress instead of 10.10.10.10):
service cvspserver
{
disable = no
socket_type = stream
wait = no
user = cvs
group = cvsusers
log_type = FILE /var/log/cvspserver
protocol = tcp
env = '$HOME=/home/cvs/repository'
bind = 10.10.10.10
log_on_failure += USERID
port = 2401
server = /usr/bin/cvs
server_args = -f --allow-root=/home/cvs/repository pserver
}

kill and restart xinetd

check from another machine, whether the installation went ok:
cvs -d :pserver:testuser@cvshost:/home/cvs/repository login
cvs -d :pserver:testuser@cvshost:/home/cvs/repository co CVSROOT
# it works, so remove it again...
rm -rf CVSROOT

and finally migrate the old repository. you just have to copy the existing contents into the new one…

VOILÀ… DONE!

some more links:
http://www.ahinc.com/linux101/users.htm
http://www.linux.ie/articles/tutorials/managingaccesswithcvs.php
http://www.network-theory.co.uk/docs/cvsmanual/
http://ximbiot.com/cvs/manual/

here the commands: to add users, to manage groups, to install apache+php+mysql+phpmyadmin, to add mysql users and databases

add a user:
adduser username

add a group:
addgroup groupname

add a user to a group (link):
adduser username groupname

change a user-is/group-id (link):
usermod -u 123 -g 456 username
groupmod -g 4321 groupname

create a softlink:
ln -s Ziel(verzeichnis) Linkname

links:

http://www.cyberciti.biz/faq/howto-linux-add-user-to-group/

http://polishlinux.org/console/file-permissions-in-linux/

http://it.toolbox.com/blogs/locutus/how-to-change-a-users-uid-and-gid-26368

http://www.everyjoe.com/newlinuxuser/topic/how-to/

http://www.linuxheadquarters.com/howto/index.shtml

installation of apache, mysql, php, phpmyadmin:
apt-get update
apt-get upgrade
apt-get install apache2 php5 mysql-server php5-mysql phpmyadmin
mysql_install_db resp. mysql_secure_installation
/etc/init.d/mysql restart
/etc/init.d/apache2 restart
to access phpmyadmin, we need to add the following line to /etc/apache2/apache2.conf: “Include /etc/phpmyadmin/apache.conf”
if apache fails to start -> apt-get install apache2-mpm-prefork

more useful utils:
apt-get install wget unzip php5-gd

add a new mysql-user and create a new mysql-database:

login as root:
mysql -h localhost -u root -p

create a new database:
mysql> create database testdb;
Query OK, 1 row affected (0.00 sec)

add a new user and grant full access to the previously created database:
mysql> grant CREATE,INSERT,DELETE,UPDATE,SELECT on testdb.* to testuser@localhost;
Query OK, 0 rows affected (0.00 sec)

set a password for the previously added user:
mysql> set password for testuser@localhost= password(‘irgendeinpasswort’);
Query OK, 0 rows affected (0.00 sec)

in the end flush the privileges and logout:
mysql> flush privileges;
mysql> exit;

i googled around and found 2 interesting pages [1 and 2], which gave me some understanding on the whole issue.

solution: let me quote a reply from someone called Bjorn Abelli on one of these pages, to explain the whole situation:

It’s no bug at all, though it’s not *convenient* for us Java developers.

Instead it’s quite logical, as java.sql.* pretty much conforms to…

SQL!

Those databases that really conform to “standard” SQL *doesn’t* have any time information in their DATE-fields (hence a java.sql.Date), instead they have it in TIME-fields (note there is also a java.sql.Time!).

And there’s also the datatype TIMESTAMP in the standard form of SQL, to have a combination of date and time (hence the java.sql.Timestamp!)…

Unfortunately there’s plenty of db:s that *doesn’t* conform to standard SQL, such as Oracle, MS Access/Jet, SQL Server, etc…

For the fields that includes both date and time in those databases, the *closest* thing to represent it is java.sql.Timestamp.

No bug…

so finally i went this way, to get the whole thing working:

statement.setTimestamp(1, new java.sql.Timestamp(date.getTime()));

there are some other issues, if you use Timestamp in java versions older than 1.4, but that wasn’t relevant in my case. if it is for you, you should check [2].

[1] http://www.keyongtech.com/4634511-java-timestamp-in-sql
[2] http://www.thunderguy.com/semicolon/2003/08/14/java-sql-date-is-not-a-real-date/
some more ideas and explanation on the whole case, unfortunately just in german language:
[3] http://www.hi-tier.de/Entwicklung/Technik/datetime001.html
[4] http://help.sap.com/saphelp_webas630/helpdata/de/6c/7784ac804be2419f957e8434be6a07/content.htm

so in case of a problem, usually you start to search around the internet, if there are other people, who solved the same or similar problems before and kept some hints and information about it.
i thought about keeping or posting the gathered information – or even both – before, but i never did it. so everytime i start it over again – searching my code and the net for the right solutions.

to stop this inefficient bullshit, i’ll start this weblog now. for me, for you and for everyone who cares about it