Note:- I had an issue testing my SFTP connection when I configuring the deployment location and received the following error;
“Conection to ‘192.168.0.10’ failed. Invalid descendent file name “Accept:”.”
It turns out that I’d managed to somehow create a file called “Accept:” in my home folder in Ubuntu and the WebStorm IDE doesn’t appear to deal well with filenames that contain ‘:’, at least I think that is why. All I know is I deleted the file and was then able to connect without issue. Hope this helps someone as it took a while to figure out this morning in my pre-caffeinated fugue.
So I’ve been trying to get to grips with scripting for git to allow me to commit then deploy via FTP on Windows. I discovered git-ftp which looked promising but it was Linux flavoured, I followed the rabbit hole all the way to installing Cygwin and attempting to build it for Windows before realising I could just use PowerShell with Posh-Git and PSFTP (PowerShell FTP) instead. More to follow on that in the next post however though I’d share a few things I learnt about Cygwin first, no point in my learnings going to waste, this post is mostly a link dump.
Git-FTP can be found here and information regarding installing packages into Cygwin can be found at the source and there is a packpage manager, apt-cyg, that I never tried but looked useful.
A few issues I hit were that I sit behind a proxy at work so following this guide to configure Cygwin to have internet access. The guide works for http access and if you need https access, just used this command;
To access my existing git folder which sits in my Documents folder I mounted my C: drive in Cygwin then created a symbolic link;
mount c: /c
ln -s /c/Users/Username/Documents
This creates a folder named Documents in your home folder which maps to your Documents folder in your Windows user folder. I found this in a cheat sheet on a Stanford users site, very useful so my thanks to them.
All fun learning, more info on my PowerShell solution in the next post.
Today’s post relates to my experience installing Atlassian Stash on a headless Ubuntu box. My first attempt ended with me trying all sorts of nonsense which left the server messy as hell, including an accidental installation of Unity…
The trickiest part for me was the installation of Java as the version available via the package manager, OpenJDK, didn’t seem to play ball with Stash. No idea why but one of the helpful fellows from Atlassian, Stefan Saasen, pointed me in the right direction with this article which I’ve used as basis for the steps below, the browser plugin steps were omitted for obvious reasons.
First of all you need to grab the tar file for Java from here then get it onto your server, the easiest way I found was to use an FTP application in SSH file transfer mode. Then connect to the terminal and run through these commands;
Decompress the tar file noting that the file name will reflect the version you download and may differ from mine;
tar -xvf jdk-7u2-linux-x64.tar.gx
Move the JDK directory to /usr/lib;
sudo mkdir -p /usr/lib/jvm
sudo mv ./jdk126.96.36.199_17 /usr/lib/jvm/jdk1.7.0
Now run these lines, no idea why, explanation in the comments please!
sudo update-alternatives –install “/usr/bin/java” “java” “/usr/lib/jvm/jdk1.7.0/bin/java” 1
sudo update-alternatives –install “/usr/bin/javac” “javac” “/usr/lib/jvm/jdk1.7.0/bin/javac” 1
sudo update-alternatives –install “/usr/bin/javaws” “javaws” “/usr/lib/jvm/jdk1.7.0/bin/javaws” 1
Correct the permissions for the executables;
sudo chmod a+x /usr/bin/java
sudo chmod a+x /usr/bin/javac
sudo chmod a+x /usr/bin/javaws
If you are installing on a fresh server, that should be all you need for Java to work. If you’ve a few JVMs present on your machine, see the original article for steps to configure the default. For Stash, you can now follow the instructions here. One thing I found quite useful was the wget command that a colleague showed me, it makes downloading via the terminal quite easy, getting the Stash installer for example can be done like so;
For editing the setenv.sh file, and text files in general, I’ve found nano really useful. I’ve not ventured into the murky depths of the text editor holy war yet but this does the trick for now. I had to run the same chmod against the Stash home directory I created but after that I was good to go. Open a browser to the path listed in the console and use the wizard to continue setup.
As I’m implementing Git at work, I’ve decided to start using it at home too and for the site. As such I’ve dusted off an old account I’d forgotten I’d created and you’ll find me on github as jedibowler. The first of my projects to be posted is the code for the network enabled temperature/humidity logger from a few weeks back which can be found here.