I have installed gitlab and gitlab_ci (with runners) on the same server running CentOS 6.5. Problem: Builds to not start.
Both gitlab and gitlab_ci pages render fine. I have "checked" things several times (i.e., redis, the ports, iptables, fully disabled selinux, and ran "bundle exec rake gitlab:check RAILS_ENV=production" as the git user for sanity sake), but the builds do not start. I verified the project settings on gitlab -> project -> settings -> services ... and everything checks out.
Is there anything that I'm missing? Is the above setup actually an acceptable setup?
Thanks!
J
Related
TIA for your help.
I recently started experimenting with Google App Engine, and I have been able to set up a project successfully.
However, I made a mistake with the location of my local files and I would like to change it.
This is the output from my console when I deploy:
jnkrois#dev:~/Development/My_Project$ gcloud app deploy
Initializing App Engine resources...done.
You are about to deploy the following services:
My_Project/default/1234567890 (from [/home/jnkrois/Development/My_Project/app.yaml])
Notice that the local folder is /home/jnkrois/Development/My_Project/app.yaml
I want to change the gcloud settings in order to pull the files from my /var/www/html/My_Project/
That way I can run the project locally via my Apache server.
Thanks for your help.
That way I can run the project locally via my Apache server.
In the vast majority of cases you won't be able to run your GAE project through apache. Except, maybe, for a totally static website with a very particular config.
The proper way to run your GAE project locally is using the development server, see Using the Local Development Server
But to answer your question - there is no extra dependency of the project outside the project directory, so just move the project directory to where you want (up to you to check address any permission issues, assuming all permissions are met in the example below) and run the gcloud cmd from the new project location:
mv /home/jnkrois/Development/My_Project /var/www/html
cd /var/www/html/My_Project/
gcloud app deploy
Again, donno if this will help you run it through apache or not.
Guys I've spent last 24 hours continuously trying to install Redmine 3.x on the WHM/Cpanel server but failed to do so. Can some please guide me with one proper way or tell me about a relatively new article regarding this issue because every solution i find on the internet is old.
There is no out-of the box solution for Cpanel.
Redmine is complex Ruby on Rails based web application, and as such it relies on lot's of 3rd party libraries, defined in Gemfile.
Although it migth be possible to install all Gems manually via Cpanel and avoid bundle install command, you would still need to run some commands like
bundle exec rake generate_secret_token
bundle exec rake db:migrate RAILS_ENV=production
So to install it via WHM/Cpanel, I suggest you create a cpanel account for your Redmine, then enable SSH access for that Cpanel user, and follow official installation tutorial from Redmine. (Which is too big to write here inside answer).
Just make sure to unpack your Redmine outside public_html!
And you can deploy your Redmine via Cpanel as FastCGI, by pointing your application path to Redmine's public directory, just make sure to have dispatch.fcgi in your public folder of Redmine
mv dispatch.fcgi.example dispatch.fcgi
mv htaccess.fcgi.example .htaccess
I cloned universal-starter (webpack version) and have it up and running on my local machine using npm start and npm run watch per the instructions
Now stuck after npm run build and attempting to deploy to Azure (and Google Cloud) via the github integration - can't figure out how to set up either to work.
Anyone have a recipe on how to get the webpack bundled files to fire up on an external host with express.js? Do I need to run commands via a CI integration? The files in /dist don't seem to stand on their own.
At Netlify you can connect your git repo and tell them what build commands you want them to use. If you specify the "dist" directory, then they will deploy anything that gets in there (after they have compiled your application).
Edit: the lowest tier is free.
Edit2: I am not associated with Netlify. I just used them in my latest deploy, and found the process extremely easy.
Note: This has changed dramatically since Angular 2. While I'm now moved on to SSR, docker, and all kinds of other things, the simplest answer was to
1) Production build
ng build --prod
2) Transfer files to a static web host (i.e., I used awscli to connect to a s3 bucket when it was just a static site...I know use SSR so I need to use a node server like express)
3) Serve files (there are some complexities for redirect requirements for index.html for error and for 404...and of course setting the status for both redirects to 200)
4) Put something on the frontend for performance/ ssl/ etc. nginx or a CDN would make sense.
I'm getting started with Vagrant and spent some time installing packages, setting up my DB and adding some data to the DB. Now that I have a base working box for my development environment, I would like to share this image with colleagues, that they can use as local VMs.
Is this not possible with vagrant? I just tried vagrant package and then destroyed it and did a vagrant up with my config.vm.box_url pointed to that packaged box location. To my dismay none of my installed packages, or files and configurations were included with my packaged vm.
Am I misunderstanding what vagrant is for or perhaps expecting vagrant to do something it's not designed to do here? If installed packages aren't the purpose of vagrant package, then what use cases is it for?
I've read through the docs and not found answers to these questions there.
Of course I can provision everything, and I'll get there too, but it's not what I'm getting at in this question.
I've since come to the conclusion that vagrant is basically for provisioning. If I need this I can just use virtual-box, but provisioning is a better approach because system packages and application dependencies etc. are a moving target.
Using provisioning allows future provisions to stay up to date and potentially expose incompatibilities.
I'm trying to deploy openMRS v.1.9.2 to a local VM running CentOS & Glassfish 2 for work. Unfortunately, I could not get it to work. Normally, I just download the standalone found at source forge. I just double-click the jar, and I'm good to go.
I normally just SSH into the the VM, so I first tried doing everything through a terminal. Here are the steps I took:
Using wget, retrieve the .zip
Create a dir (I just called it /openmrs), cd into the new directory, and then expand the .zip.
cd into the directory.
At this point, there are two options to start openMRS.
Run the bash script: ./run-on-linux.sh
Run the .JAR: java -jar [insert_jar_name].jar -commandline
When I run the .JAR, I get a stack trace.
When I try to run the bash script, I get another error.
Anyways, I thought I found a potential solution in an openMRS JIRA ticket, but it seems aimed at Glassfish 3, and not Glassfish 2 (which is what I need to use).
I then tried deploying the .WAR via the Glassfish admin UI. I thought it would work, but after going through the steps of selecting a language, whether or not to use demo data, etc. I received this.
Does anyone have experience deploying openMRS to Glassfish 2.1.1? Unfortunately Glassfish 3 doesn't seem to be a realistic option. I would really appreciate any help here. Thanks.
Although it doesn't solve my problem of not being able to successfully deploy openMRS to an instance of Glassfish v.2, I did manage to get myself further by just installing MySQL on the VM. Our work machines are all set up for postgres, so I think should have guessed earlier that not having a MySQL server installation was the problem.
Here is a tutorial I used to install MySQL