Added output method to job class
This was required because job_fail will not allow us to return the output of the failed job back to the client (job.py). This is a problem because gearman protocol does not support it. So what we did is, we make the worker always return a success but add a failure return code to the output (json format) that is passed to the client. The client checks the rc from the output to determin is the job failed or succeeded.
It does not exit if fabric fails It instead sends a job fail signal to the job Also, it is now returning the output of the fabric job as well.
This is required so that if the gearman server goes down during an operation, an appropriate "Server unavailable" error gets printed.
All other modules will be using this module to print messages
It interacts directly with the job class
It was such a bad logic on my part -- Can't believe I could be this stupid
run > Runs the job poll > Poll for job to finish and get the output and return codes success > Whether the job was successful or not
Had to get rid of some server names
A simple plugin is written which simply streams the output of the jobs in a jobflow
Added option for the get_dict_val funtion to exit if exitonerror set to True
jobflow config file are definitely present
Removed all the code that was running Fabric jobs
* I have hardcoded a sample plugin as of now * Also made changes to variable names