Anonymize IPs with Netconan #169
Conversation
chadell
requested review from
glennmatthews,
pke11y and
scetron
as code owners
There was a problem hiding this comment.
I noticed that at least one of the randomized IPs was still an actual IP address, I think that it would be good to limit it to the documentation IP addresses.
A second thought on anonymization would be that there are a bunch of email headers that appear to contain some information that is not necessary to the function of the library. Not sure if any of it might be sensitive however.
| X-SM-DEDUPING-ID: 7cc8bab7-00bb-44e0-a3ec-bdd1a5560b80-EMAIL--1012261942-036424c1a19ca69ca7ea459ebd6823e1 | ||
| Date: Thu, 6 May 2021 21:52:56 +0000 | ||
| Feedback-ID: 1.us-east-1.xvKJ2gIiw98/SnInpbS9SQT1XBoAzwrySbDsqgMkBQI=:AmazonSES | ||
| X-SES-Outgoing: 2021.05.06-62.46.238.146 |
There was a problem hiding this comment.
Is this a randomized IP? This is an actual IP according to WHOIS: https://apps.db.ripe.net/db-web-ui/query?searchtext=62.46.238.146
There was a problem hiding this comment.
The randomisation from netconan has no option to define the "usable" IPs to anonymise (or I haven't found it), so it randomly pick "valid" IPs.
This PR uses a different approach than #165 , using the
netconanlibrary. It makes it easier as less code is needed to maintain, but because it's randomly assigning IPs, it can't be used as a CI check to verify nothing is changing.Because this is simply a helper script, I think it is still a valid approach, for when anonymization is needed.