-
This repository includes the following programs:
- measurement_creation.py --> creates user-defined DNS-measurement in Ripe Atlas
- renderlinegraph.py --> renders a line-graph for one or more nameservers between a specified timeperiod
- renderlive.py --> renders a live line-graph for on or more nameservers
- renderscatterplot.py --> renders a scatter-plot for one nameserver between a specified timeperiod Every timestamp in this timeperiod connects to all rtt values in the resultfile from that time
- renderNSID.py --> renders a linegraph where the median is based on NSID (uses runNSID.py instead of run.py)
- renderNSIDmap.py --> renders a Scatter map with rtt values grouped by NSID (uses runNSID.py)
-
Running any of the last tree programs does the following apart from rendering:
- Downloads and parses previously defined RIPE atlas DNS measurements
- Produces a CSV file with the results from one or more specified NameServer/s, and enriches the results with probemetadata
- Produces a .gz.file with probe meta data:
- probe's country
- probe's continent
- probe's subregion
- probe's firmware version
- Probe's Coordinates
#Clone the repository
$ git clone https://github.com/NoraOdel/dash.live.medianrtt.git
$ cd dash.live.medianrtt
#Activate your virtualenv
$ cd Files
$ pip install -r requirements.txt
#read to run:
$ python3 measurement_creation.py yyyy-mm-dd HH-MM-SS(or)
$ python3 renderlinegraph.py yyyy-mm-dd HH-MM-SS (or)
$ python3 renderlive.py yyyy-mm-dd HH-MM-SS (or)
$ python3 renderscatterplot.py yyyy-mm-dd HH-MM-SS (or)
$ python3 renderNSID.py yyyy-mm-dd HH-MM-SS
$ python3 renderNSIDmap.py yyyy-mm-dd HH-MM-SS
#required arguments:
first
--> initial start time in datetime format ie. yyy-mm-dd HH:MM:SS
#optional arguments:
$ -ns (option for renderlive.py aswell)
--> choose which nameservers to plot, ex. a.ns.se4 (the last element refers to IPv, 4 or 6)
renderlive.py, default is 'all4' which equals to every nameserver and its IPv4 traffic
$ -interval (is not always relevant)
--> interval between result fetching, default is 10 minutes
$ -numberofintervals (is not always relevant
--> number of intervals which defines the timeperiod for plot, default is 144. If numberofintervals is 144 and interval is 10 the timeperiod will be:
144*10 = 1440 minutes --> 24h
# EXAMPLE:
$ python3 renderscatterplot.py 2019-11-22 12:00:00 2019-11-23 12:00:00 -ns a.ns.se4 z.ns.se4 b.ns.se6 -interval 30
This would create a scatter plot with a timeperiod between 2019-11-22 12:00:00 and 2019-11-23 12:00:00,
for nameservers a.ns.se4, z.ns.se4 and b.ns.se6 with an interval of 30 minutes
-
Results file similar to this:
a.ns.se4-20191016-1571205731-1571206331-atlas-results.csv- Different components are:
measurementID for a.ns.se IPv4date=20191016start=1571184000stop=1571270399(each measurement has a duration of 10 min)
- Different components are:
-
Probe meta data file similar to this:
20191016-probemetadata.json.gz- This includes information (listed above in 'this program does the following') about relevant Ripe-Atlas probes
-
Running measurement_creation.py--> MeasurementID file consisting of measurementIDs and connecting NameServers
Result and probe meta data files will be continuously deleted.
This CAN be changed in the script however, it is NOT recommended since one or more files
are created every 6 seconds if renderlive.py is run --> A LOT OF FILES.