Provides various tools to extract measurement data from a system and then uses the TrueSight Pulse REST API to send measurement data to your account.
The TrueSight Pulse ETL Toll is installed via pip:
$ pip install tspetl
The accompanying tool requires your login email and api token.
These can be specified on the command line with the -e and -t command line arguments respectively or
specified by setting the following environment variables:
TSP_EMAILTSP_API_TOKEN
in your shell like the following:
export TSP_EMAIL=dude@example.com
export TSP_API_TOKEN=api.558c71066a-0000
The list of tools available can be listed by running the following:
$ tsp-etl -h
usage: tsp-etl [-h] {weather,apachelog,twitter,csv,log} ...
Tool to extract/transform/load into Pulse
positional arguments:
{weather,apachelog,twitter,csv,log}
commands
weather Collects weather measurements from a city and optional
country code. (Future Release)
apachelog Parses apache logs for page status. (Future Release)
twitter Collects tweet data from Twitter (Future Release)
csv Import CSV file
log Collects measurements from log files (Future Release)
optional arguments:
-h, --help show this help message and exit
The follow sections provided usage on each of the tools
The CSV (Comma Separated Value) tool allows the loading of data stored in a file formatted as a CSV.
usage: tsp-etl csv [-h] [-e email] [-t token] [-a hostname] [-s {api,rpc,std}]
[-f file_path] [-b batch_count] [-o origin] [-p id]
[--skip-first-line]
optional arguments:
-h, --help show this help message and exit
-e email, --e-mail email
-t token, --api-token token
-a hostname, --api-host hostname
-s {api,rpc,std}, --sink-type {api,rpc,std}
-f file_path, --file file_path
Path to file to import
-b batch_count, --batch batch_count
How many measurements to send in each API call
-o origin, --origin origin
Origin to associated with the measurements
-p id, --application-id id
Application Id to associate with the measurements
--skip-first-line Skip header line in file
You can load data from a CSV with the following format:
"metric", "value", "source", "timestamp"
CPU, 0.1,"etl-test-red",1466406000
CPU, 0.2,"etl-test-red",1466413200
CPU, 0.3,"etl-test-red",1466416800
CPU, 0.4,"etl-test-red",1466420400
CPU, 0.5,"etl-test-red",1466424000
CPU, 0.6,"etl-test-red",1466427600
CPU, 0.7,"etl-test-red",1466431200
CPU, 0.8,"etl-test-red",1466434800
CPU, 0.9,"etl-test-red",1466438400
CPU, 0.1,"etl-test-green","2016-06-23 1:00AM"
CPU, 0.2,"etl-test-green","2016-06-23 2:00AM"
CPU, 0.3,"etl-test-green","2016-06-23 3:00AM"
CPU, 0.4,"etl-test-green","2016-06-23 4:00AM"
CPU, 0.5,"etl-test-green","2016-06-23 5:00AM"
CPU, 0.6,"etl-test-green","2016-06-23 6:00AM"
CPU, 0.7,"etl-test-green","2016-06-23 7:00AM"
CPU, 0.8,"etl-test-green","2016-06-23 8:00AM"
CPU, 0.9,"etl-test-green","2016-06-23 9:00AM"
CPU, 0.1,"etl-test-blue","2016-06-23 13:00"
CPU, 0.2,"etl-test-blue","2016-06-23 14:00"
CPU, 0.3,"etl-test-blue","2016-06-23 15:00"
CPU, 0.4,"etl-test-blue","2016-06-23 16:00"
CPU, 0.5,"etl-test-blue","2016-06-23 17:00"
CPU, 0.6,"etl-test-blue","2016-06-23 18:00"
CPU, 0.7,"etl-test-blue","2016-06-23 19:00"
CPU, 0.8,"etl-test-blue","2016-06-23 20:00"
CPU, 0.9,"etl-test-blue","2016-06-23 21:00"The timestamp can be entered as unix epoch time or the formats shown above.
With a file named test.csv with the contents above the data can be loaded and associated with the application DOG_ONE by executing the following command:
tsp-etl csv -f test.csv --skip-first-line -p "DOG_ONE"