Where is GitLab installed on Ubuntu 14.04? - ruby-on-rails

GitLab was installed by following instructions at https://about.gitlab.com/downloads/ on Ubuntu Server 14.04 but I can't find the install directory. It's commonly said to be in /home/git/ or /home/gitlab/ but neither exist.
I need to find the install directory so I can setup SMTP (RE https://gist.github.com/petermanser/6117494) and migrate existing repositories into GitLab.
Does anyone know where GitLab would be installed on Ubuntu Server 14.04 if not /home/git/ or /home/gitlab/?

Config files are in etc/gitlab
Data is stored in /var/opt/gitlab
Thanks to Bjorn for finding the locations and Pavel S for providing cat /etc/passwd | grep git for finding the home directory of a given user!

To find a user's home directory do
$ cd ~git
you will be in that users directory.
:D

There is a Directory structure section in the official README.md file of Omnibus-GitLab repo:
Omnibus-gitlab uses four different directories.
/opt/gitlab holds application code for GitLab and its dependencies.
/var/opt/gitlab holds application data and configuration files that
gitlab-ctl reconfigure writes to.
/var/log/gitlab contains all log data generated by components of
omnibus-gitlab.
/etc/gitlab holds configuration files for omnibus-gitlab. These are
the only files that you should ever have to edit manually.
But, if you install the GitLab alone from source (rarely, but useful for unsupported systems like *BSD). Then the GitLab installation location will be different, and it is described in this official doc.

Related

Install libc6-dev, libgdiplus and libx11-dev using manifest.yml file on Cloud foundry

I'm trying to deploy an asp.net core 3.1 API on cloud foundry. I don't have admin rights, i just have developer rights. Is there a way to specify the URL of these libraries (libc6-dev, libgdiplus and libx11-dev) (maybe git or some official repository) so that i can execute the manifest.yml file during deployment and install these dependencies? Also to mention, i cannot turn on support for docker file on cloud foundry, as i get a message (insufficient rights)
I would suggest you give the apt-buildpack a try. You can give it additional Ubuntu package names, and it will install those for you.
You do that through an apt.yml file. Check out this post for instructions.
It's important to understand that the apt-buildpack will install these packages into a non-standard location. Since it also runs as a non-root user, it cannot install them into standard locations.
To work around this limitation, it sets variables like $PATH and $LD_LIBRARY_PATH to point to the locations where it has installed items. Most build tools will pick up these env variables and be able to locate what you install.
It's not perfect though, and some tools require additional env variables to be set. If you still get errors when building, look at your build tools and check if there are ways you can point to where apt-buildpack is installing stuff. The path it writes to can vary based on your buildpack order, but if you print out $PATH you can see the location. It's often /home/vcap/deps/0/... but the index can change based on your buildpack order.

Change /tmp folder for jenkins in debian

I recently had a low disc space issue in jenkins. Although the home directory has sufficient space, the /tmp did not and it brought my master down.
I tried many ways to fix this
Discard old build and deleting old workspaces
Restarting jenkins
Tried to setup a node with same config and starting it, but that failed saying i dont have sufficient permission for copying remote.jar
Since my server is debian, I could not find any file where I could set set -Djava.io.tmpdir
Many people talked about jenkin file or config file, but I could not find it.
The only thing that worked was https://updates.jenkins-ci.org/download/plugins/tmpcleaner/ , this plugin hpi, which I installed through Advanced tag in plugins.
Also, I went manually to the /tmp folder in root and then deleted it contents to start the jenkins again. So my question is:
How do i change the /tmp folder in jenkins?
The jenkins home directory is /mnt/data/jenkins, however the tmp it uses it root/tmp. Why is it so? The /mnt/data/jenkins has lots of free space and i manually created a /tmp folder there, but it only takes the /tmp from the root
My colleague found the answer out
open '/etc/defaults/jenkins' file using command 'less /etc/defaults/jenkins'
Add argument
JAVA_ARGS="$JAVA_ARGS -Djava.io.tmpdir=yourNewtempLocation"
Once you restart jenkins, you will be able to see the storage of the new space

Cryptogen tool not found. Hyperledger fabric installation in Windows 10

I followed the Hyperledger fabric documentation to install and configure it in Windows 10. However when I run the command - "./byfn.sh -m generate" for first-network sample application, I get the following error,
I have gone thru all StackOverflow questions regarding this and made sure following steps are done,
Have set the $PATH variable correctly to include bin folder.
Have downloaded the platform-specific binary and my bin folder looks like this,
I have doubts about following steps,
I have installed Docker for Windows and was able to verify the docker installation by running hello-world image in Docker. However, I have not shared any of my local drives in Docker. Not sure whether this is the cause of this error.
Please note that this is my first question in StackOverflow. Forgive me for any mistakes/redundancies. Any help is greatly appreciated.
I'd suggest making sure that you run the script to download / install the binaries and images from within the fabric-samples directory.
The $Path is exported every time you run the byfn.sh script, confirm that the path configuration in the byfn.sh is correct and points to your correct bin location
# prepending $PWD/../bin to PATH to ensure we are picking up the correct binaries
# this may be commented out to resolve installed version of tools if desired
export PATH=${PWD}/../../bin:${PWD}:$PATH
export FABRIC_CFG_PATH=${PWD}

Getting started with Laradock on Mac

I'm just getting started with learning Laravel, as well as taking my first dive into Docker. I've taken the time to understand what Docker is and (more or less) how it works, and have also set up Docker Native on my Mac (Sierra).
I'm following along with the documentation at: http://laradock.io/getting-started/, but am a bit confused on how to proceed. The current version of Laradock is v5 which has a different setup to previous versions. I'd like to maintain a different docker environment for each project, and as per the documentation:
Follow these steps if you want a separate Docker environment for each project
So I proceeded with the section titled A.2) Don’t have a PHP project yet, where it mentions that the folder should look like:
+ laradock
+ project-z
So in my home directory I have a general Dev folder, and inside that I have a Laravel folder. So in ~/Dev/Laravel/ do I clone the Laradock git repo, and then alongside it (it being the laradock folder that is created) create my projects? So by my understanding I will therefore end up with:
~/Dev/Laravel/laradock
~/Dev/Laravel/project-1
~/Dev/Laravel/project-2
~/Dev/Laravel/project-...
If so, then this does not seem to be a separate Docker environment for each project.
Or should I instead go the route of creating a new empty Laravel project first and then clone the Laradock git repo into that?
I know this a very noobie question, but I'm unsure on how to proceed. Any help is humbly welcome!
Note: I have homebrew, composer and PHP7 already installed on my Mac.
If you want a separate Docker environment for each project, clone the Laradock repo inside each project directory. Your directory structure should look like:
+ project-a
+ laradock-a
+ project-b
+ laradock-b
Alternatively, if you want a single Docker environment for all your projects, your directory structure should look like:
+ laradock
+ project-1
+ project-2
I'd recommend to go with adding laradock as a submodule to an existing laravel app's folder with command: git submodule add https://github.com/Laradock/laradock.git first.
You have to initialize Git first ofc to do that.
To be concrete, you need to see a folder structure like this:

is symfony system wide?

i have a directory with some .php files in it...do i need to install symfony in that directory so that i can run symfony commands from that directory?? i tried: pear channel-discover pear.symfony-project.com
and got:
Channel "pear.symfony-project.com" is already initialized
the i tried: pear install symfony/symfony
and got:
WARNING: configuration download directory "/build/buildd/php5-5.3.2/pear-build-download" is not writeable. Change download_dir config variable to a writeable dir to avoid this warning
Cannot install, php_dir for channel "pear.symfony-project.com" is not writeable by the current user
how do i fix this please? must i run it as sudo??
thanks
If I remember well, Symfony is "system-wide" if installed through PEAR (like you are trying to do).
Now it is recommended by the manual to do a "standalone" install through SVN. See http://www.symfony-project.org/gentle-introduction/1_4/en/03-Running-Symfony
As well, the sandbox will bring a "standalone" installation of Symfony (you will have to be in the project root directory to be able to run the symfony commands)
EDIT
Example of quick install using SVN:
mkdir /path/to/symfony
cd /path/to/symfony
svn checkout http://svn.symfony-project.com/tags/RELEASE_1_4_0 .
but it should be better to follow the manual and configure the external SVN sources.
I run servers where I have applications that use different versions of symfony simultaneously. I found it much easier to deal with the SVN checkout than the PEAR insteall.
So what I've done is do an SVN checkout of each revision that I need
into it's own folder. In my case i check them out to
/usr/share/symfony/(version)
Then I configure the ProjectConfiguration.class.php to reference whatever version is needed, using the line:
require_once '/usr/share/symfony/1.x.x/lib/autoload/sfCoreAutoload.class.php';
If I need to switch to a different symfony version, I can just change that config.
That is, assuming all my code is compatible with that version

Resources