"This address is restricted"

[Mathwiz]

Hopefully this will be an easy one for one of the Firefox experts. Trying to learn about the IceApe browser, Serpent (55) gave me this:

Apparently Serpent (at least 55) blocks port 50100 for some strange reason, but that's where the IceApe changelog is. Is there a pref to control this? Searching for 50100 in about:config yields nothing....

[roytam1]

1 hour ago, Mathwiz said:

Hopefully this will be an easy one for one of the Firefox experts. Trying to learn about the IceApe browser, Serpent (55) gave me this:

Apparently Serpent (at least 55) blocks port 50100 for some strange reason, but that's where the IceApe changelog is. Is there a pref to control this? Searching for 50100 in about:config yields nothing....

50100 is not banned as it is not listed in https://github.com/roytam1/basilisk55/blob/master/netwerk/base/nsIOService.cpp#L98

you may check if there is pref `network.security.ports.banned` and try resetting it.

or you may add 50100 to string pref `network.security.ports.banned.override` in about:config.

[Mathwiz]

Thanks, but no dice. There's no network.security.ports.banned pref; yet it refuses to connect, even after I created string pref network.security.ports.banned.override and set it to 50100

[roytam1]

43 minutes ago, Mathwiz said:

Thanks, but no dice. There's no network.security.ports.banned pref; yet it refuses to connect, even after I created string pref network.security.ports.banned.override and set it to 50100

OK I think I found the root of issue, and a fix is pushed.

[Mathwiz]

Thanks! I'll be sure to download the next Serpent 55 update once you post it. Does this affect any other platforms, such as UXP? If so I'll download those too.

I looked at the code you linked to, and it looked pretty straightforward. There's a hard-coded list of `bad' ports, and you can add more ports to the list with the network.security.ports.banned pref, and/or remove ports from the list with the network.security.ports.banned.override pref. So it was quite strange that it wouldn't work no matter how I set those prefs!

[Skorpios]

NewMoon28: ok
K-Meleon: ok
Mypal: ok
Serpent52: ok

Serpent55: same result as you / "This address is restricted"
 

[XPerceniol]

Hmmm ... fine here with both Serpent 52 and FF 52.9.1?

https://git.hyperbola.info:50100/

Then I clicked ->software/icecrypt.git/log/

[Mathwiz]

Good - sounds like the issue was unique to the Moebius platform and doesn't affect UXP browsers. Which sort of makes sense, because IceApe is itself a UXP browser - I think they would've noticed if they couldn't browse their own change log!

[roytam1]

6 hours ago, Mathwiz said:

So it was quite strange that it wouldn't work no matter how I set those prefs!

yeah, because there are some bit width problems in variable, and problem is revealed in the conditions in the code after security patches applied. fix is simple as well: https://github.com/roytam1/basilisk55/commit/fc7f3915a7bd9d30615f0430c9c014f7fd758b76

EDIT: also reported to upstream: https://github.com/classilla/tenfourfox/commit/4f70dd5c954cfcfd55357527da98c15f8355fbe5#r59478686

Edited November 5, 2021 by roytam1

[Mathwiz]

On 11/4/2021 at 4:59 PM, XPerceniol said:

Hmmm ... fine here with both Serpent 52 and FF 52.9.1?

And fine with FF 53 too. Evidently a very unique bug.

23 hours ago, roytam1 said:

problem is revealed in the conditions in the code after security patches applied.

I see; the procedure is passed a 32-bit value, and wants to ensure it's a valid port number (in the range 1 to 65535), so that you'll get the error if you enter something idiotic like microsoft.com:0 or google.com:65536; but it was first assigning the 32-bit value to a 16-bit value, cutting off the high-order bits and rendering the check for being over 65535 meaningless. Worse, it was a signed 16-bit value, so anything over 32767 tested as less than 0 and was blocked as an invalid port number!

Edit: Roytam's fix works! BTW, I saw that he found and fixed the same coding mistake in FF 45 and the Tycho browsers (PM 27 and KM); yet it didn't cause the same bug, at least in the FF 45 build I was using. (Haven't tried the others.) Like later FF versions, FF 45 accessed port 50100 just fine.

Edited November 6, 2021 by Mathwiz