Hopefully this will be a new and united ansible for both our production and staging web environments. And we can stop maintaining 2 repos.
- Do not forget to add your ssh credentials to the machines!
> ansible-playbook -i staging_hosts --extra-vars "my_env=staging" deploy.yml
Make sure to the blastmachine and software install roles send things to the right/same/consitent places as the current setup and the staging setup
- Do not forget to add your ssh credentials to the machines!
> ansible-playbook -i production_hosts --extra-vars "my_env=production" deploy.yml
- create a salt
> openssl rand -base64 32
- Then hash the password with the salt
> openssl passwd -1 -salt <salt> <password>
Fresh machines from TSG have a fixed sudoers file and changes will be lost at midnight each day. Email requests with any permenant changes you want to make to /etc/sudoer
On bm1 and bm3 start a pool of 2 workers using ~/analytics_automated_alt/
celery --app=analyticsomated_project.celery:app worker --loglevel=INFO -Q low_localhost,localhost,high_localhost,celery,low_R,R,high_R,low_Python,Python,high_Python --pidfile=celery.pid --concurrency=2
celery --app=analytics_automated_project.celery:app worker --loglevel=INFO -Q celery,blast,low_blast,high_blast --concurrency=2 --pidfile=celery.pid
Don't for get to put things back on bioinfnew1
- copy index_production.html over index_production_alt.html on the fronend