tbr | I've taken an alternative route for now on the device side. | 11:24 |
---|---|---|
tbr | The original problem is: you want apps to use mqtt, but you don't want one connection and the associated connection management overhead per app. | 11:25 |
tbr | well, I'm now successfully running a MQTT bridge on the device | 11:26 |
tbr | this way apps can connect to a local mqtt instance (instead going through a mqtt-dbus proxy) | 11:26 |
tbr | this /might/ also address timeout issues, but that's still tbd | 11:27 |
tbr | the only downside is: each app needs to actually use mqtt instead of some sort of dbus messaging. → TODO: still write a dbus proxy in the long run, but less critical to get a working end to end setup for app developers | 11:29 |
tbr | server temporarily unavailable while I upgrade | 12:14 |
tbr | hmhm, some sort of protocol problems, will need to look into that, connections might fail from python for now | 12:36 |
tbr | so it looks like latest mosquitto builds a broken server on wheezy... | 13:56 |
tbr | it's really weird | 13:56 |
*** lbt has quit IRC | 16:05 | |
tbr | lots of debugging later... | 16:44 |
tbr | things are now working | 16:45 |
tbr | I also rewrote the notification script to use a local bridged mosquitto server | 16:45 |
kimmoli | tbr lots of monolog here i see :) | 18:23 |
tbr | yeah | 18:38 |
tbr | today was a bit beyond my basic debugging horizon and I had to ask the author of mosquitto for help | 18:39 |
tbr | he was nice and pointed out some things, in the end it was again all about the ACLs. :-/ | 18:39 |
kimmoli | ok | 18:41 |
tbr | my current plan would be to have a mosquitto server on every phone, it is small footprint and would serve as a good proxy | 18:42 |
tbr | authentication to the central server based on certificates | 18:42 |
tbr | once I get around to set up TLS | 18:42 |
kimmoli | concur the world ? | 18:54 |
tbr | like every night, pinky, like every night. | 19:15 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!