This repository is for the download of BGP advertisements and infering conclusions from them over the years.
Steps involved in the process are:
- Download the required RIBs using the script scrapper.py.
- Extract the ribs first using:
tar -xvf <rib-file.bz2>
- Extract the extracted ribs using zebra-dump-parser.
- The extracted rib-dumps should be combined, to find the combined list of ribs for a particular year through multiple places.
- The combined ribs must have many duplicate nodes. This means that multiple route-collectors might be capturing a single bgp-advertisement. The duplicates must be removed using the following command:
sort -u <combined-file-name.txt>
- The sorted files now have the problem of AS-prepending. This must be reduced using PATH-CLEANING. The file required is file-path-cleaning.py.
- Now we require the AS-Relationship file for the particular years. This means that for the analysis of an year, we require the relationship file for that year.
- Based on the relationship file, we find the valleys in the file using the valley-in-file.py