search

deanhiller / databus

time series data in cassandra with visualization(NREL's opensource databus project)
Other
53 stars 21 forks source link

Demo "mode" vs "script" [comment/discussion] #14

Open hopcroft opened 11 years ago

hopcroft commented 11 years ago

It appears that the "initial turn-on" for the Databus code that external users can download has a sort of "demo mode" built-in, with some fake users, and fake data. Is this correct?

This seems like it is asking for trouble- you have to have a build system which supports production and test builds, or modes or something, and the code contains a lot of stuff that is not related to the core functionality.

Instead, is it possible instead to a "demo script" which would provision a Database, make a few Tables, and load them with fake data? A new user could run this script after installing Databus for the first time. The script could even have "pause and press any key" features, where the user is invited to look at the web interface and see the new database listed, now click on Tables and see the tables, now click on Charts and see the data, etc.

This would make for cleaner code, illustrate the API capabilities, and serve as a programming example.

deanhiller commented 11 years ago

How are you starting the system? ./startProduction.sh starts production mode which I thought populates no data(I have not tested this in a while though but it should not be populating data while ./play run does populate data)

hmmmmm, ok, looking at the documentation, this looks like a documentation bug that we need to clarify. I will have Vikas make changes here for more clarity.

thanks, Dean

easility commented 11 years ago

Documentation updated

deanhiller commented 11 years ago

Vikas, the documentation still says ./play run in two places I see. In those places, we really need to document there are two option play run and ./startProduction. play run will popuplate data and startProduction should not populate data. Do me a favor and make sure startProduction does not populate any data as well as we want a clean install and startup with no tables for when people put in production.

easility commented 11 years ago

Ok. I re-test that using ./startProduction.sh is not creating any demo data (except 4-5 meta Columnfamilies by Playorm). Document is updated as well. ON 3 node cluster page, user is recommended to use this shell script to start. While on single node cassandra, he is given both option