Skip to content

Super Awesome way to manage health checks for VMWare, Palo Alto, network config, etc. and display captured data over time

License

Notifications You must be signed in to change notification settings

AnnAddicks/sendash

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

sendash

Super Awesome way to manage health checks for VMWare, Palo Alto, network config, etc. and display captured data over time

#Testing There are a couple tests that work with the remote git repository and code. These require a username and password passed in via command line for maven or set in the environment variables of the IDE. The following is required to be set to make the tests passs: *-Dgit.username=your.email@email.com *-Dgit.password=yourPass

Currently the remote URL for the repository is hard coded in the application.yml file.

Notes

request from service includes:

  • node identifier string (unique api key)
  • timestamp of last received configuration

response from server includes an array of objects that were updated since the timestamp in the request:

  • unique identifier of job, so the service knows if this is a net new job or one that will overwrite something it already has
  • (name of job) not sure this is necessary, really - does the service ever need to know the name of the jobs it is running? possibly for error reporting?
  • data (the powershell code that will be executed)
  • schedule data (when the powershell code will be executed. this is a cron string, or possibly "asap"/"now")

in a limited configuration such as what we are building, there's really just two jobs to speak of:

  1. the initial configuration job that handles node setup and deployment. this job's schedule is "asap" or "now," and fires as soon as it's received.
  2. the worker job, which has a defined schedule (every weekday at 7pm, or somesuch), and is the one that kicks off the actual work on a regular basis.

it's not strictly necessary to go any further than this, but this sort of mechanism could be used, for instance, to granularize the checks themselves. say, we set up a schedule to ring checks 1-10 daily at 7pm, but check 11 runs once every two hours, and check 12 runs every half hour. or we could set up a separate daily task to run a backup report.

in what format/method does the service need to send the response data back to the server?

Jobs vs Checks

Job: something that the client's service runs Check: something that the client's powershell runs

About

Super Awesome way to manage health checks for VMWare, Palo Alto, network config, etc. and display captured data over time

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •