Re: [mh] Fun data: insteon address corruption

[Marc M. <ma...@me...>]

On Sun, Nov 29, 2009 at 06:52:45PM -0500, Gregg Liming wrote:
> Marc MERLIN wrote:
> > I'm going through the fun of resetting my entire network to link against
> > the old now dead PLM and bind with the new one (oh boy, with 71 addresses,
> > 20 of which are wireless, remotelinc and motion sensor, that's not fun)
> > 
> > But the interesting thing I noticed:
> > aldb 0ed9ee031 [0x0FD0] contlr(03) record to 0ed9ee(01), (d1:fe, d2:00, d3:00)
> > aldb 0ed9ee041 [0x0FC8] contlr(04) record to 0ed9ee(01), (d1:fe, d2:00, d3:00)
> > (...)
> > aldb 12d9ee021 [0x0FC0] contlr(02) record to 12d9ee(01), (d1:fe, d2:00, d3:00)
> > 
> > My old PLM was indeed 12.D9.EE
> > 
> > Notice how 0ed9ee is different from 12d9ee by just one bit, which probably
> > got lost/corrupted either during my programming a while ago, or in the
> > remotelinc.
> > 
> > Doh... I guess it's only so reliable.
> 
> I'm assuming that this is after a rescan of the device memory?  If so, 

Yes. I do enough debugging that I never use log links, I always use scan all
links.

> then the adlb records get wiped and fully repopulated with whatever the 
> device responds with.  So, it's as "real" as the device knows.  However, 

Right.

> add/deletes/modifications all result in mh's mirror.  So, there's always 
> some slight possibility that mh's perception of reality is out of sync 
> w/ the device's memory until a rescan occurs.

Totally. That was a rescan and the bad data was definitely in the device.
Since it was a remotelinc the "easiest" was to factory reset it, but that
was a lot of work to reprogram everything (maybe 30mn or so since I1 and
remotelincs suck).

Marc
-- 
"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/