Diversion Diversion - the Router Ad-Blocker

[Stephen Harrington]

@thelonelycoder @Jack Yaz et al

Have been experimenting with the YT blocking, had to do your reset as below to get it all working.

In d select "7. Reset Diversion ads counter or logs" and then reset Dnsmasq log files.
Browse some YT videos and then run b,8,2 again. I'm crossing fingers.

Don't know if it is coincidence but now uiDivStats is showing some pretty impressive ad-blocking numbers!

Anyone else seeing something like this?

Running a "Generate Diversion Statistics now" in at command line for uiDivStats didn't fix it, just made different spurious results ...

Not really sure if this is a Diversion issue or a uiDivStats issue?

Anyway the YT ad-blocking seems to be working well, thanks!

 

[heslo]

i got over 269 domains collected.

Mine seems stuck on 13. I'd be interested to see how you got over 200

 

[thelonelycoder]

Hi,

Is there some way to know/test/confirm whether Diversion YouTube ad blocking is working on your system, or not? I've got it enabled, but I'm still seeing a significant though un-quantifiable number of ads on YouTube. I see nothing in the firmware GUI - Addons - Diversion Stats that suggests YouTube ads are being blocked.

Thanks,
Anton

They don't count as ads blocked - yet. Technically all the r*.**-********.googlevideo.com domains that the YT block feature collects are forced to use the same IP address, instead of them individually resolving their own IP. That server behind the forced IP will not have the same content as the real server.
Or so I think this is how it works...

I have yet to build in the counter that includes this new feature.
To check if the YT blocking works, run the f command in Diversion, select one of the Unfiltered log options and view some YT videos.
Every r*.**-********.googlevideo.com that is directed to the forced IP address set and shown in b, 8 potentially shows no video ads.

Edit: And to add, every r*.**-********.googlevideo.com that is not resolved to the forced IP address should then be collected by this new YT blocking feature and added to the list.

 

[thelonelycoder]

@thelonelycoder @Jack Yaz et al

Have been experimenting with the YT blocking, had to do your reset as below to get it all working.



Don't know if it is coincidence but now uiDivStats is showing some pretty impressive ad-blocking numbers!

Anyone else seeing something like this?

Running a "Generate Diversion Statistics now" in at command line for uiDivStats didn't fix it, just made different spurious results ...

Not really sure if this is a Diversion issue or a uiDivStats issue?

Anyway the YT ad-blocking seems to be working well, thanks!

That's something for @Jack Yaz to figure out, I don't do percentages calculation in Diversion.

 

[SomeWhereOverTheRainBow]

They don't count as ads blocked - yet. Technically all the r*.**-********.googlevideo.com domains that the YT block feature collects are forced to use the same IP address, instead of them individually resolving their own IP. That server behind the forced IP will not have the same content as the real server.
Or so I think this is how it works...

I have yet to build in the counter that includes this new feature.
To check if the YT blocking works, run the f command in Diversion, select one of the Unfiltered log options and view some YT videos.
Every r*.**-********.googlevideo.com that is directed to the forced IP address set and shown in b, 8 potentially shows no video ads.

perfect answer.

 

[Jack Yaz]

That's something for @Jack Yaz to figure out, I don't do percentages calculation in Diversion.

I suspect resetting a counter messed with it. The total queries is far lower than the queries blocked. I'm not too concerned about fixing it because the current incarnation of uiDivStats is being thrown away soon. Trying to wrangle sqlite into a state where I'm happy with performance

 

[Jack Yaz]

Does YT ad blocking apply to the second dnsmasq instance if it is being used, out of interest?

 

[Stephen Harrington]

That's something for @Jack Yaz to figure out, I don't do percentages calculation in Diversion.

Figured it might be for @Jack Yaz but then I noted the "Total Queries = 9,733 Queries Blocked = 62,893" and wondered whether it was something in Diversion itself which Jack was then generating the percentages from?

 

[SomeWhereOverTheRainBow]

Does YT ad blocking apply to the second dnsmasq instance if it is being used, out of interest?

it probably would if it was using dnsmasq.conf.add and the default for users was set to "include" dnsmasq.conf.add

 

[thelonelycoder]

Does YT ad blocking apply to the second dnsmasq instance if it is being used, out of interest?

No, I don't think so. Actually that is a no.

 

[thelonelycoder]

it probably would if it was using dnsmasq.conf.add and the default for users was set to "include" dnsmasq.conf.add

Let me do that, otherwise there will be a bunch of files or entries Diversion has no control of.
And I'd have to sort it out.

 

[Jack Yaz]

No, I don't think so. Actually that is a no.

No worries. I customise the second instance so I'll poke around at any changes needed for my setup.

Thanks for this wonderful feature!

 

[thelonelycoder]

No worries. I customise the second instance so I'll poke around at any changes needed for my setup.

Thanks for this wonderful feature!

I'll get a fix out for that file soon.

 

[thelonelycoder]

Fix is out for Alternate blocking list
- Added YT blacklist
- Change to use the new location of the generated wildcard blacklist

Use uu to force update, then toggle ad-blocking or logging off, then back on to apply.

 

[Stephen Harrington]

I suspect resetting a counter messed with it. The total queries is far lower than the queries blocked. I'm not too concerned about fixing it because the current incarnation of uiDivStats is being thrown away soon. Trying to wrangle sqlite into a state where I'm happy with performance

No problem @Jack Yaz, just reporting it to you and @thelonelycoder in case it was relevant. Will await updates from you both.

 

[SomeWhereOverTheRainBow]

Mine seems stuck on 13. I'd be interested to see how you got over 200

relentlessly watching youtube videos for hours on in, (actually the kids did).

 

[thelonelycoder]

No worries. I customise the second instance so I'll poke around at any changes needed for my setup.

Thanks for this wonderful feature!

Could you point me to which file(s) contain that p7zip code you mentioned?
I still have to fix the missing real zip Entware package thing.

 

[thelonelycoder]

relentlessly watching youtube videos for hours on in, (actually the kids did).

My main router is at 57, but there is very moderate use of YT here
But every time an ad is shown the new domain gets picked up with the next run of the YT collector. It's nice to look at it actually working.

 

[SomeWhereOverTheRainBow]

My main router is at 57, but there is very moderate use of YT here
But every time an ad is shown the new domain gets picked up with the next run of the YT collector. It's nice to look at it actually working.

it seems at some point you hit a plateau and the ads collected slow down a bit. not that it is not work, i just suspect the addresses( and amount of address) are different based on users geo-location.

 

[thelonelycoder]

it seems at some point you hit a plateau and the ads collected slow down a bit. not that it is not work, i just suspect the addresses( and amount of address) are different based on users geo-location.

Exactly, the DNS Server in use might also matter and how much YT usage one has. I'm sure the Google people use cookies to determine how many and where from video ads are served.