Copying the list.

-Dave

---------- Forwarded message ----------
From: David Frascone <dave@frascone.com>
Date: Jun 19, 2007 1:32 PM
Subject: Re: lircd "delay" that lasts a minute or so?
To: Christoph Bartelmus <lirc@bartelmus.de>

Debug doesn't show much.  Log for the last two days is at http://www.frascone.com/lircd.broked.gz (1.1Mb)

Here's the broken section from this morning:

Jun 19 04:31:14 myth-fe lircd: trying "Hauppauge_350" remote
Jun 19 04:31:14 myth-fe lircd: <s4
Jun 19 04:31:14 myth-fe lircd: sync
Jun 19 04:31:14 myth-fe lircd: timeout: 1097
Jun 19 04:31:14 myth-fe lircd: <s4
Jun 19 04:31:14 myth-fe lircd: pulse expected
Jun 19 04:31:14 myth-fe lircd: failed on bit 1
Jun 19 04:31:14 myth-fe lircd: failed on code
Jun 19 04:31:14 myth-fe lircd: failed "Hauppauge_350" remote
Jun 19 04:31:14 myth-fe lircd: decoding failed for all remotes
[Wife pressed power button on box]
Jun 19 13:24:52 myth-fe lircd: removed client
Jun 19 13:24:52 myth-fe lircd: caught signal
Jun 19 13:24:53 myth-fe lircd: started server socket
Jun 19 13:24:53 myth-fe lircd: "begin" "remote"
Jun 19 13:24:53 myth-fe lircd: parsing remote
Jun 19 13:24:53 myth-fe lircd: creating first remote
Jun 19 13:24:53 myth-fe lircd: "name" "Hauppauge_350"
Jun 19 13:24:53 myth-fe lircd: parsing Hauppauge_350 remote

I have mythfrontend running in a script like this:

while [ 1 ] ; do
   /etc/init.d/lircd stop
   rmmod lirc* (I have the module names -- doing this from my head)
   mv mythfrontend.log mythfrontend.log.old
   /etc/init.d/lircd start (loads modules)
   mythfrontend >mythfrontend.log
done


The Power button is set to do a killall mythfrontend (which --- restarts and fixes the remote)

-Dave


On 10 Jun 2007 19:34:00 +0200, Christoph Bartelmus < lirc@bartelmus.de> wrote:
Hi David,

stopping lircd and running mode2 seems to be enough to make it work
again. The mode2 data looks ok and can be decoded to RC5 codes.
Please compile lircd in debug mode and start it with -D3. If the problem
happens again please send me the kernel output and the output of /var/
log/lircd for the last couple of minites before the problem happened.
Please also add your /etc/lircd.conf file.

Christoph

on 10 Jun 07 at 12:07, you wrote:
> Duh -- figured it out -- mode2 attached.
>
> -Dave
>
> On 6/10/07, David Frascone < dave@frascone.com> wrote:
>>
>> What's mode2 output?  (Sorry -- bit lirc ignorant still).
>>
>> BTW -- this death definitely looks different than the 1min death.  Remote
>> never comes back.  irw is silent.  strace on lircd is slient.
>>
>> -Dave
>>
>> On 10 Jun 2007 15:40:00 +0200, Christoph Bartelmus < lirc@bartelmus.de>
>> wrote:
>>>
>>> Hi David,
>>>
>>> on 10 Jun 07 at 08:28, you wrote:
>>>> Different behavior now -- remote stops working, then stays dead.
>>>
>>> Very strange. The sense parameter is not touched anymore. And according
>>> to the logs there is still data received. So the effect should be the
>>> same as commenting out the sanity check.
>>> Could you please repeat the test and send me the mode2 output once the
>>> remote stops responding. And please send me the lircd.conf that you are
>>> using.
>>>
>>> Christoph
>>>
>>
>>
>>
>> --
>> David Frascone
>>
>> Oxymoron: Safe Sex.
>>



> --
> David Frascone
>
> Oxymoron: Safe Sex.




--
David Frascone

Make like a drum and beat it!


--
David Frascone

Oxymoron: Safe Sex.