Mark Hindess

I've not thought about the correct fix but a workaround is to launch with: ``` LD_PRELOAD=/lib/x86_64-linux-gnu/libudev.so.0 node test.js ``` where the correct path to libudev.so.0 will be os dependent but...

I've made two fixes - workaround the AnyEvent one_event method disappearance and the Notifo module failure. I'd be interested to know what errors are still there after those fixes but...

Just a quick comment that one reason hollie saw fewer failures was because he must have pulled commit 1397ef56330589b4427decfa369c0a870478862a which I was a bit slow pushing to github before running...

"make test" problems should be fixed now. Can you confirm by closing this issue? Thanks, Mark.

Can you elaborate? For instance, what values are you seeing that you aren't expecting? Can you send xpl-logger output of for the unexpected values? Note that xpl-rfxcom-rx reports everything it...

What are the model numbers of the oregon sensor you have?

Can either of you - @giejo has fewer sensors so that might be marginally easier to figure out - send me some output from the command: ``` strace -tt -xx...

Thanks. Now I'm really confused as it looks like the bytes read from the tty are exactly what is being reported as being processed in the case of the incorrect...

Dan, Just to rule out an obvious change between 0.11 and 0.12, can you try 0.12 with Device::RFXCom::RX with the change in: https://github.com/beanz/device-rfxcom-perl/commit/819698ceefef1cfe3aba3518afc32207a5cb78f4 Regards, Mark

Thanks for the additional information. I can get hold of an THN132ES in a few days. So hopefully I'll be able to reproduce the problem and get to the bottom...