Being a big corp, it's almost a requirement that your reliance on all things microsoft has to be solid and deep rooted. Of course we have open source equivalent for most of the stuff but the pressure is on for me to perform and to present some tangible results as soon as possible thus I barely had time to put some basic best practices in place. But finally after 2 days of googling and tweaking, I've got a small pc running ubuntu server serving trac for ticketing and serving out our main mercurial repo. We don't even have the basic infra yet and still management says "these are things are nice to have, but you need to deliver those reports soon. Our deadline was last month".
Then when starting to do development, I start to see how deep the rabbit hole goes. Never mind about the ton of spaghetti code we have to tangle with, database management (this is just our internal development database mind you) consist of importing excel file into access and using odbc to push those tables to ms sql server. Ha. Seamless I tell you.. >.<
So of course there is nothing in the open source world that can match that kind of right click send seamlessness. So for now I concede I have to use windows. At least until most of our database stuff is fully in the ms sql server and I can connect to that directly. But the pain of working in windows.. oh my.. Even looking for text in all the source code is not as easy as 'grep -ir something'. I think I miss grep the most. But of course I miss all the rest of our little treasure trove of command line tools to make dev so much easier. How I wish I had a 'tail -f' in one window of terminator while I restart the server and redo the query in another.
But I'm already here now and I should square my shoulders and get in line.
Some interesting thing I've learned already is how to set the default gateway for windows for example. You just do a:
route change 0.0.0.0 mask 0.0.0.0
That is to change the default gateway if it already exists. Want to know whether it already exists? Print it out:
route print
And also for setting up tomcat6 cgi, edit the conf/web.xml file in the tomcat root directory. Inside there you would see already commented out parts for cgi config. Just remove tags from 2 parts:
<servlet>
<servlet-name>cgi</servlet-name>
<servlet-class>org.apache.catalina.servlets.CGIServlet</servlet-class>
<init-param>
<param-name>debug</param-name>
<param-value>0</param-value>
</init-param>
<init-param>
<param-name>cgiPathPrefix</param-name>
<param-value>WEB-INF/cgi</param-value>
</init-param>
<init-param>
<param-name>passShellEnvironment</param-name>
<param-value>true</param-value>
</init-param>
<load-on-startup>5</load-on-startup>
</servlet>
*note - you should add the passShellEnvironment part to allow the cgi to use perl and stuff.
And from:
<servlet-mapping>
<servlet-name>cgi</servlet-name>
<url-pattern>/cgi-bin/*</url-pattern>
</servlet-mapping>
And then you have to change the conf/context.xml file to enabled privileged like this:
<Context privileged="true">
Once that is done you've got it made baby.. you can now download strawberry perl and awstats and have awstats even for a tomcat6 server. Follow the instructions here for more details: http://www.wrenbeck.com/flowbuilder/$$download.xsp/blog/3eab0461f928f/awstats_tomcat.html
Just one more note on changes to tomcat to enable awstats. You need to enable the combined log for it to work. For that edit the file conf/server.xml and at the end of the file you might find something like this:
<Valve className="org.apache.catalina.valves.AccessLogValve" directory="logs"
prefix="localhost_access_log." suffix=".txt" pattern="combined" resolveHosts="false"/>
Release it from it's remarked prison and change the pattern from "common" to "combined".
Learned quite a lot these past few weeks so I guess that's very good. Still have to use windows so that's pretty bad. But plus minus it all, it's still pretty good.