pixelserv Looks like it is back into action!

[SomeWhereOverTheRainBow]

Yes, you've sharpened the focus nicely.

I just looked, and at 30 hours current uptime I'm at 2,700 successful responses out of 566,000. Still a negligible amount (although I will admit 90% of that 566,000 represents calls to permutive.com).

I've been using pixelserv for almost 7 years now, so I don't remember the old days when the single pixel was a big deal. Pages seem to display perfectly acceptably with large white space--they are even improved design-wise.

If the principle benefit of pixelserv-tls then is snappy loading, what more do we need?

Maybe there can still yet be improvements with the mechanisms that handle these loading/ certificate handling process. maybe there is away to reduce the memory usage footprint. it would require testing and new beta's, but I hope this is the future pathway @kvic is on.

 

[thelonelycoder]

Maybe there can still yet be improvements with the mechanisms that handle these loading/ certificate handling process. maybe there is away to reduce the memory usage footprint. it would require testing and new beta's, but I hope this is the future pathway @kvic is on.

When I received the GitHub notification for the new release I was at first annoyed since I planned on abandoning the pixelserv-tls option in Diversion. Then it dawned on me that @kvic sees a future for his project.
We will have to wait and see if v2.4 will be added to Entware. At the moment there is no request placed by kvic to the Entware team.

 

[CriticJay]

The analysis done several years ago by kvic, pertaining to null redirection (0.0.0.0), is no longer true I believe.

I think it was someone on the pihole forum who re-analyzed and found that under modern device/OS/browser use cases, there was no noticeable difference in page loads comparing 0.0.0.0 to pixelserv-TLS IP. In addition, pixelserv-TLS redirection can cause issues with mobile devices, IoT, and media streamers (Chromecast, Apple TV, etc). This is a big issue because those devices are rapidly becoming the majority in people's houses.

I know for me, the reason I have 23 devices on my LAN is because of IoT devices and streamers. Sure you can manually whitelist one-by-one to get problem apps like "Amazon Shopping" to work on your android phone through pixelserv-TLS, but honestly that's a waste of time. In the previous SNB forum, the Diversion thread contained DOZENS of people complaining about "Amazon Shopping" not working on Android/pixelserv-TLS, and many man-hours were wasted trying to discover the correct whitelist-host combination to get it to work.

Meanwhile, disabling pixelserv-TLS prevented those problems from occurring in the first place.

 

[SomeWhereOverTheRainBow]

The analysis done several years ago by kvic, pertaining to null redirection (0.0.0.0), is no longer true I believe.

I think it was someone on the pihole forum who re-analyzed and found that under modern device/OS/browser use cases, there was no noticeable difference in page loads comparing 0.0.0.0 to pixelserv-TLS IP. In addition, pixelserv-TLS redirection can cause issues with mobile devices, IoT, and media streamers (Chromecast, Apple TV, etc). This is a big issue because those devices are rapidly becoming the majority in people's houses.

I know for me, the reason I have 23 devices on my LAN is because of IoT devices and streamers. Sure you can manually whitelist one-by-one to get problem apps like "Amazon Shopping" to work on your android phone through pixelserv-TLS, but honestly that's a waste of time. In the previous SNB forum, the Diversion thread contained DOZENS of people complaining about "Amazon Shopping" not working on Android/pixelserv-TLS, and many man-hours were wasted trying to discover the correct whitelist-host combination to get it to work.

Meanwhile, disabling pixelserv-TLS prevented those problems from occurring in the first place.

That is because instead of the response being "imitated" like it does with MITM pixelserv-tls, all your blocked responses hit a wall of 0.0.0.0. If you are confronted with too many of these, you may experience your app not even opening. So whitelisting would still definitely be a highly necessary process in some cases to reduce some of those 0.0.0.0 responses.

 

[bluepoint]

That is because instead of the response being "imitated" like it does with MITM pixelserv-tls, all your blocked responses hit a wall of 0.0.0.0. If you are confronted with too many of these, you may experience your app not even opening. So whitelisting would still definitely be a highly necessary process in some cases to reduce some of those 0.0.0.0 responses.

I have not experience any app or websites not opening or slowed down without pixelserv, with pixelserv, I can't open some websites including my ISP's. If any, credit to diversion, this script is very well written that it doesn't depend on pixelserv to function effectively.

 

[SomeWhereOverTheRainBow]

I have not experience any app or websites not opening or slowed down without pixelserv, with pixelserv, I can't open some websites including my ISP's. If any, credit to diversion, this script is very well written that it doesn't depend on pixelserv to function effectively.

I have had apps stall loading or not load at all with 0.0.0.0 blocking. Not trying to keep the wheel spinning though.