From: Marc MERLIN <marc_mh@me...> - 2011-04-19 15:08:24
As an update mostly for Gregg, and others if you are curious.
I use sec.pl to look for patterns in print.log.
One is for Parsing serial data that is too long, although it seems the one I
caught below may not have been a hang afterall (usually something that long
is a hang).
For that, I use my 1-wire I/O board to power cycle the PLM.
Mon Apr 18 00:03:29 2011: 18/04/2011 00:02:52 [Insteon_PLM] Parsing serial data: 0261821300060258150250099425129dfb611381
Maybe that reset was premature, this may have been an 'ok' string.
I'll raise my 'too long' to 40 characters instead of 30.
Then, my PLM will occasionally trigger Gregg's new code:
Mon Apr 18 00:04:20 2011: 18/04/2011 00:04:20 [Insteon_PLM] Button Pressed. Ignoring.
Mon Apr 18 00:04:21 2011: 18/04/2011 00:04:21 [Insteon_PLM] Button Pressed. Ignoring.
This just fixes itself and it goes on after that.
I've however noticed that the PLM does miss a bunch of state updates from
other devices that it used not to, so it looks like it's going south.
I'm trying to hold off for the insteon branch to work well enough that I can
scan all links, sync all links and delete orphans (hopefully soon) and I'll
upgrade to a dual band PLM.
(I just can't see links to 100+ targets (due to keypadlincs with 8 buttons
each, and remotelincs), some manual like on my remotelincs, with the old
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems & security ....
.... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/
Get latest updates about Open Source Projects, Conferences and News.