[Resolved] 255.255.255.255

General discussion about the NoScript extension for Firefox
Post Reply
barbaz
Master Bug Buster
Posts: 10018
Joined: Sat Aug 03, 2013 5:45 pm

[Resolved] 255.255.255.255

Post by barbaz » Sun Nov 17, 2019 12:10 am

Code: Select all

v 11.0.9rc1
============================================================
[...]
x Use invalid IP rather than domain name to prevent offline 
  status from breaking sync messaging in Chromium
https://github.com/hackademix/noscript/ ... b555751584

In my experience, using 255.255.255.255 as a "invalid IP" can actually result in connection attempts, whereas using 0.0.0.0 seems not to.

1) Why use 255.255.255.255 instead of 0.0.0.0?

2) Is it just as safe to whitelist this IP-based implementation in the same way I had to whitelist the previous implementation?

I am uncertain enough about this that I'm keeping NoScript at 11.0.8rc1 until I better understand the consequences of this change. Thanks for any clarification.
Temporarily off forum staff at my own request
*Always* check the changelogs BEFORE updating that important software!
-

User avatar
Giorgio Maone
Site Admin
Posts: 9143
Joined: Wed Mar 18, 2009 11:22 pm
Location: Palermo - Italy
Contact:

Re: 255.255.255.255

Post by Giorgio Maone » Sun Nov 17, 2019 7:49 am

barbaz wrote:
Sun Nov 17, 2019 12:10 am
In my experience, using 255.255.255.255 as a "invalid IP" can actually result in connection attempts, whereas using 0.0.0.0 seems not to.
What's your experience exactly?
My experience: in Firefox and Chromium (on Linux at least), http://0.0.0.0 opens the web server on localhost, while http://255.255.255.255 fails as an invalid address before any connection attempt.
barbaz wrote:
Sun Nov 17, 2019 12:10 am
1) Why use 255.255.255.255 instead of 0.0.0.0?
See above.
barbaz wrote:
Sun Nov 17, 2019 12:10 am
2) Is it just as safe to whitelist this IP-based implementation in the same way I had to whitelist the previous implementation?
Yes, it won't connect anyway (it's the multicast address, has no meaning for HTTP requests).
Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:71.0) Gecko/20100101 Firefox/71.0

barbaz
Master Bug Buster
Posts: 10018
Joined: Sat Aug 03, 2013 5:45 pm

Re: 255.255.255.255

Post by barbaz » Sun Nov 17, 2019 12:13 pm

Giorgio Maone wrote:
Sun Nov 17, 2019 7:49 am
barbaz wrote:
Sun Nov 17, 2019 12:10 am
In my experience, using 255.255.255.255 as a "invalid IP" can actually result in connection attempts, whereas using 0.0.0.0 seems not to.
What's your experience exactly?
My experience: in Firefox and Chromium (on Linux at least), http://0.0.0.0 opens the web server on localhost, while http://255.255.255.255 fails as an invalid address before any connection attempt.
Oops, sorry, I just re-tried and realised this is my mistake :oops: I was confusing it with 255.255.255.0 that I had tried a long time ago following https://hackademix.net/2009/07/01/abe-w ... where-omg/

255.255.255.255 doesn't seem to make any connection attempt for me either.
Giorgio Maone wrote:
Sun Nov 17, 2019 7:49 am
barbaz wrote:
Sun Nov 17, 2019 12:10 am
2) Is it just as safe to whitelist this IP-based implementation in the same way I had to whitelist the previous implementation?
Yes, it won't connect anyway (it's the multicast address, has no meaning for HTTP requests).
Thanks Giorgio for clearing up my concerns! Image
Temporarily off forum staff at my own request
*Always* check the changelogs BEFORE updating that important software!
-

Post Reply