DocScrutinizer05 | xmn gone? A pity. Would be worth investigating options to work around or fix slide sensor | 10:09 |
---|---|---|
DocScrutinizer05 | (supl-proxy) can't we run a supl server on maemo.org? | 10:10 |
sixwheeledbeast | that was my thought | 10:11 |
DocScrutinizer05 | *maybe* we should use ssh or ssl with individual keys users obtain from maemo suply management orga for a donation of arbitrary amount, valid 1 year. Would avoid any exploitation of the supl server infra as well | 10:14 |
DocScrutinizer05 | s/suply/SUPL/ | 10:14 |
infobot | DocScrutinizer05 meant: *maybe* we should use ssh or ssl with individual keys users obtain from maemo SUPL management orga for a donation of arbitrary amount, valid 1 year. Would avoid any exploitation of the supl server infra as well | 10:14 |
DocScrutinizer05 | to test, I'd offer free test certs/keys that expire after 3 days | 10:16 |
sixwheeledbeast | i believe the proxy will only work with the modified certs so unlikely people will use it for anything other than maemo. | 10:16 |
DocScrutinizer05 | and also get void when overused, by sth like fail2ban with a threshold of 100 connections / hour | 10:17 |
DocScrutinizer05 | sixwheeledbeast: I'm talking right outa my rear, but maybe some ideas in there worth getting refinement | 10:17 |
DocScrutinizer05 | oooh, now I get it, SUPL already _has_ cert protection, right? | 10:19 |
DocScrutinizer05 | which I guess was one of the major sources of any troubles with SUPL during the last 10 years | 10:20 |
DocScrutinizer05 | re xmn's "broken" device: I wonder what the heck actually broke. And I also wonder how much real trouble user would experience from a kbd slide that's considered always open. Better than always closed anyways | 10:22 |
Maxdamantus | Why would it need modified certs? | 10:22 |
Maxdamantus | Wasn't the issue just that Google started rejecting old TLS clients? | 10:23 |
DocScrutinizer05 | could the magnet have latched up between display slide and main PCB / hallswitch and have damaged the latter? if so then in which way? just a few pins torn off the PCB? | 10:24 |
sixwheeledbeast | it was hosted at nokia, google one didn't work they rejected requests IIRC | 10:24 |
DocScrutinizer05 | or does the magnet stick right on top of the hall chip and nobody noticed? | 10:24 |
Maxdamantus | Oh. | 10:24 |
DocScrutinizer05 | thetre was a time google SUPL worked with N900 | 10:25 |
DocScrutinizer05 | and there was a point in time when it stopped to work | 10:25 |
sixwheeledbeast | i say was i believe nokia still works if you fix yo TLS hence proxy | 10:33 |
DocScrutinizer05 | sorry, didn't parse | 11:07 |
DocScrutinizer05 | ;-) | 11:07 |
sixwheeledbeast | s/\<was\>/& ,/ | 13:31 |
sicelo | bencoh: yes supl-proxy works fine, and i agree it is a good solution. the slight problem we have is that - if the gps fails to lock (this does happen even after getting valid supl response), then even re-running the proxy doesn't help, as no further supl requests are made by the device. | 14:45 |
sicelo | what i tried at some point is to stop and restart location-proxy, but i couldn't get something conclusive. | 14:45 |
sicelo | however it seems that a new supl request does get emitted if the cellid or some such changes. | 14:46 |
sicelo | i'll do a brief writeup on wiki so others can also test. | 14:47 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!