Learning is an Art

I just go a mail on Subversion mailing list informing that today is the 6th Birthday of Subversion for those who are thinking what is Subversion they can read here.

So Subversion has come along pretty well since it’s 1.0 release 6 years back and this day I would like to thank all developers who have put in there efforts to make Subversion what it is today; it stands tall among the many commercial tools.

So how I celebrated the Subversion’s birthday I went to svnforum.org and tried to answer queries of people as thats the way you can help Subversion adoption to grow among masses.

Developers are doing there bit by developing it; so we as a part of community need to contribute in what ever way possible.

This day I wish many many happy returns of the day to Subversion and want to congratulate Subversion community for great going looking ahead to more great accomplishments in future.

I have been thinking for a while to write about my experiences but was refraining from this for some reason or the other but today something happened that I can’t hold my anger against the shity policies of companies in India; I will illustrate three experiences of my life which forced me to believe that India is still way behind when it comes to serving customers and listening to there problems and promptly providing the solutions.

So I will start with my latest problem which I am facing with Tata Indicom broadband connection; I took a plan of RS 1300 6GB combo so it was all goody goody initially when I didn’t used much of my allotted usage of 6 GB free usage which I had; than this month I had to download some linux operating system from internet that started the nightmare experience it was 3.7 GB of download still less than my free usage and the very next morning I got a call that you have exceeded your credit limit of RS 1700 and you need to make a payment of 3100 I am still trying to figure out from where the hell that figure of RS 1700 came from; if I do a flat conversion of 6 GB of free usage with their rate 0.9/MB it leads out to some where around RS 5000 I was totally surprised with this and in next two days my line was barred; I didn’t had a option there I went to the office of tata indicom and made the payment of RS 3100 and it took them more than a day to restore my line.
I haven’t even realized that my lines are fully working I got another call next day asking me to make a payment of RS 1500 and since I was afraid that my line might get disconnected again which may cause another day of work getting lost so I made the payment immediately but I was even not able come out of two back to back payments I got another call from them today that I have exceeded my credit limit again and I need to make a payment of RS 2300 I was so badly pissed of I called the customer care and asked what the hell is going on and he had more shocking news for me sir your billed amount is RS 7000 and your unbilled usage is RS 11000 I was shocked WTF and than I asked him to check again than he started saying actually the correct unbilled usage will start reflecting in 48 hours that amount is not correct that was some respite for me.
I was told I can check my unbilled usage from tataindicom site but there another tragedy stuck me I had registered to that site already but never received the first time pin I told the same thing to customer support executive, he tried to reset password but it was in vein he kept getting that customer not registered and I kept getting you are already registered and you can not registered again.
He told me sir you will have to write a mail; now I have to write two mails the first one which I had already written for increasing my credit limit and another to fix my account on tataindicom.
I hope they will not f*** any more as I have already given up with them. I am planning to change ISP any suggestions are welcome.

I think this story has taken quite out of me I will write other two experiences some other day, keep watching this space for more bitter realities of customer support in India.

Few days back I was trying to create database in a batch job using createdb.exe and I encountered this problem createdb.exe prompts for password and there is no way of passing password from command line as an argument.
I tried to find out how to do this but in vein there is no way you can pass password as an command line argument; than after looking around for sometime and googling I found two ways of doing it

• Setting up PGPASSWORD environment variable : This method is not preferable as it’s insecure more for further details read this

• Setting up password in pgpass.conf : The file is to be located
  %APPDATA%\postgresql\pgpass.conf

File has the following format

Each of the first four fields may be a literal value, or *, which matches anything.

For further details the postgres documentation

As always Google silently launched new version of its one of most popular application that is Orkut, I don’t exactly when it was made available as I noticed it couple of days back.
I thought to give it a try and realized that the features added are heavily inspired from Facebook, its quite apparent that Orkut was seem to be loosing to Facebook on many aspect so redoing the Orkut was always on the cards and they have done a fare job but still I haven’t found many people using the newer versions and you can not use newer features like commenting which is one of the most used features of Facebook unless other person is also on newer version.
Lets see what strategy is used by Google to move to people to newer version. One thing I liked and appreciate about changes done by Google is that you always have option to hop between versions very unlike Facebook which shifted me to lite version without my knowledge and only switched back after a month or so.

Here I am going to talk about three useful and handy subcommands  of svnadmin that will help us in dumping our data and restoring it back to fresh repository, restore is very critical, as we keep on backing up data but we seldom bother about restoring it to check if everything is working fine or not. I would like to put one point across that your backup is as good as your last restore.

Dumping the repository

This is the simplest option for dumping the repository for purpose of backing up your repository. If you want to do an incremental back than you need to specify –incremental flag and give start and end revision for same.

Restoring the repository back

Verify the restore

Habit of restoring the data regularly is very important for having usable backup.

This article covers how to setup domain authentication for Subversion with Apache using mod_auth_sspi , I am running Apache 2.2 and Subversion 1.5. First of all you need to get SSPI module from its new home at sourceforge , module can be directly downloaded from here. Once you have downloaded the module extract using some zip utility there will be a bin folder inside the extracted folder and inside bin folder there will a file named mod_auth_sspi.so take this file and drop it in Apache’s modules folder.

Once you are done with above steps we now need to modify Apache’s httpd.conf to include the domain authentication for our Subversion repository make changes to httpd.conf as shown below,  first we need to ensure that we have required modules loaded

To mod_dav_svn to work you need to uncomment the line which loads mod_dav module Apache look for it and uncomment if its commented.

Now the real magic configuration starts which will enable us to authenticate Subversion users against AD domains make changes in httpd.conf as shown below

Once done with above changes save the file and restart Apache and try accessing the URL http://<yourservername>/svn/myrepo you will be prompted for authentication saying “Subversion Domain” enter your domain name without domain as we have SSPIOmitDomain On and based on Subversion authorization file i.e authz in our scenario you will be able to see repository or you will get forbidden error.

We are done with setting up domain authentication, in-case you face any problem with above settings feel free to write to me.

LoadModule sspi_auth_module   modules/mod_auth_sspi.so
LoadModule dav_svn_module     modules/mod_dav_svn.so
LoadModule authz_svn_module   modules/mod_authz_svn.so

Yeah you heard me right today Atlassian JIRA 4.0 was officially released after doing many beta’s and release candidate…I was expecting that it will take people with a storm with opensocial gadget JQL and many other improvements yes it did took people with storm but for some other reasons and the reason was its licensing policy you can find details of same here and my favourite one is there reason of changing the licensing that’s really cool and pretty lame at the same time for reasons of increasing the licensing cost or changing it all together.

I really doubt the fact that any hundred people company will really like to pay $4000 JIRA and the biggest one is the Enterprise license now costs you double that is $8000 and the cherry one the cake is licensing is now based on number of active named users that means that if you have 100 users and you are now 101 ; you need to run behind Atlassian to upgrade your licenses.

I want to write more on this but may be some other lets wait and watch for reactions from other people; I will be back on this topic with more detailed analysis of pros and cons of same this was just an initial outburst.

I have been fiddling around with Apache for quite sometime now on Ubuntu and there were few questions which used to come to my mind as novice user initially like how do i install apache on first place, where are Apache logs, where are Apache conf files located, how do I add new site to Apache server, stopping and restarting the server so I thought that I would some up answer to all these question at a single place for my own future reference and for the help of my readers.

So first things first I will start with how to install apache here apt-get comes to your rescue and thanks to Ubuntu archives all you need to give simple on liner as shown below

So now that we are done with installation lets move on other things like for troubleshooting you need to know where are apache logs, apache logs can be found at below mentioned path here you can find access logs as well as error logs

we also need to know where are configuration related files to our web server for that we need to look at path mentioned below here you will find apache2.conf, httpd.conf, ports.conf and other configuration related files how to modify and update them is beyond the scope of this nifty article

Moving on to next FAQ this one is my favourite now that I have setup apache how do I go about hosting my site thats the whole purpose for you following along this article, all you need to do is follow the sequence mentioned below

Create a file mysite or what ever name you want to give and add configuration related to your site in this file

you are done with setting up your site.

Last of these faq’s is how do I start and stop the server for doing this all you need to know about is apache2ctl utility and you can start stop using it pretty easily

We are done with the faq’s which were in mind and I will keep on updating this list in future as and when things come to my mind. In next article I would covering how to use apache to host Subversion repositories along with domain authentication so stay tuned.

/etc/apache2/conf.d/

Few days back first release candidate of JIRA 4.0 was announced that means JIRA has moved pretty closer to one of the greatest release in the history of JIRA releases; there was a mixed bag in JIRA 3.13 which people were speculating to be 4.0 at that time but Atlassian decided to mark it as 3.13 as they were planning something really big for the mega release 4.0.

Two features which look killer to me are Opensocial gadgets and JIRA query langauge; I was really wondering why Atlassian went ahead with implementing Opensocial while missing out few popular issues like ldap authentication which has been around for six years now but after pondering on thought of Opensocial  I realized that Atlassian has cooked up something really big which is not easily comprehensible in current scenario of issue trackers what I feel it has a potential to cause a paradigm shift in the way we look at issue trackers and way we use them I can smell something really big coming out of it.

I would loke to end with a not to all those who have AMC to start getting ready for a Opensocial. I am also eagerly waiting for this release to happen.