[Build] Doorbell "fixed", computer gate control and keypad need some love.

Wladyslaw Zbikowski embeddedlinuxguy at gmail.com
Fri May 4 19:16:59 UTC 2012


[note moving from -disgust to -build because of secret URL]

Thanks for working on this Asheem. I am not able to trace this problem
to anything in the software on minotaur. Certainly we have never seen
this new unexpected "fail open" mode, where the lock is stuck open and
the buzzing won't stop until you hit the doorbell again. We need to
get together and go over the new wiring ASAP, as this is going to be
very annoying for people.

I am also not opposed to disconnecting the doorbell until this
annoying effect is fixed. AFAICT the keypad is working fine now.

In short, either this is a problem with the new wiring, or two wires
got shorted which we didn't know about until the new wiring exposed
the problem, or some other weird interaction between the wiring and
the rest of the system.

As far as resetting the keypad server: we noticed the server can go
bad after running several hours. Last night I put in a cron job to
restart it hourly, and (knock on wood) have seen no problems since
then. I am working on the root cause analysis, but in the meantime you
can reset the server on minotaur if needed by calling
/home/jesse/bounce-gate

If you don't have an account on minotaur, you can also restart it with
this secret URL:

http://minotaur:8000/foo.esp

On Fri, May 4, 2012 at 5:01 AM, Asheem Linaval <xormic at gmail.com> wrote:
> None of the wiring I touched is actually responsible for triggering the
> gate, so I'm reasonably sure the problem isn't there (I could be wrong). The
> keypad malfunction and the gate sticking started roughly at the same time,
> so that's a link that's worth investigating
> Someone who knows what they're doing will have to restart the payphone
> daemon; I personally wouldn't know where to start... once that happens, if
> the problem persists, we can start isolating the node that is causing
> trouble.
> .atl
>
>
> On Fri, May 4, 2012 at 12:10 AM, Jake <jake at spaz.org> wrote:
>>
>> Are you sure that the gate buzzer is not being kept buzzed by the doorbell
>> system, which you just worked on?
>>
>> The two ways the buzzer can be activated are the doorbell system (the DOOR
>> button by the top of the stairs), and the computer called minotaur.
>>
>> If the payphone does not make any sounds other than short beeps when keys
>> are pressed, it is an indication that the computer's keypad server is not
>> running properly.  It should send the payphone the command for a longer beep
>> or buzz after a wrong code or a timeout on an incomplete code.
>>
>> However, you did just work with the wires for the doorbell/buzzer system.
>> So I am not betting on which system is causing the buzzer to be stuck.
>>
>> But it should be easy to restart the computer daemon that runs the
>> payphone.
>>
>> Anyone?
>>
>> -jake
>>
>>
>> Asheem Linaval wrote:
>> Buzzer buttons have been reinstalled, and the doorbell functions normally.
>>
>> Unrelated, but noticed in the course of testing:
>>
>>   - It appears that the computer gate control is failing in such a way
>>
>>   that the gate gets stuck open/buzzing (bizarrely, pressing a doorbell
>>   button seems to stop it).
>>   - Keypad is malfunctioning. Unsure what to make of this. It beeps on
>>
>>   keypress, but not much else.
>>
>> That's all.
>> .atl
>
>
>
> _______________________________________________
> Noisebridge-discuss mailing list
> Noisebridge-discuss at lists.noisebridge.net
> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
>



More information about the Build mailing list