Global Cache Disconnects every 10 min

I'm running indigo 6.1.0 with Cynical Cache 1.6.7. I have a Global Cache GC-100-6 and its disconnecting every 10 min or so and does not reliably transmit IR. Any ideas? Any help is appreciated!

Some log data (everything starts up fine it seems):

Cynical Caché Global Cache starting
Cynical Caché Global Cache reset
Cynical Caché IR Blaser 2 reset
Cynical Caché IR Blaster 1 reset
Cynical Caché IR Blaster 3 (TV) reset
Cynical Caché Global Cache connecting to host
Cynical Caché Global Cache is now exploring
Cynical Caché Global Cache enabling idle probes
Cynical Caché Global Cache ready with 4 device(s)
Cynical Caché IR Blaser 2 host device Global Cache now available
Cynical Caché IR Blaser 2 is now ready
Cynical Caché IR Blaster 1 host device Global Cache now available
Cynical Caché IR Blaster 1 is now ready
Cynical Caché IR Blaster 3 (TV) host device Global Cache now available
Cynical Caché IR Blaster 3 (TV) is now ready
Cynical Caché Global Cache version 3.2-06

Now, when I try to send an IR command, I sometime get these errors (and then it seems to reset everything):

Action Group IR tv cursor right
Cynical Caché send <Symbol/generic:38000 [(172,172), (22,64), (22,64), (22,64), (22,21), (22,21), (22,21), (22,21), (22,21), (22,64), (22,64), (22,64), (22,21), (22,21), (22,21), (22,21), (22,21), (22,21), (22,64), (22,21), (22,21), (22,21), (22,64), (22,64), (22,21), (22,64), (22,21), (22,64), (22,64), (22,64), (22,21), (22,21), (22,64), (22,1820)]...[(172,172), (22,64), (22,64), (22,64), (22,21), (22,21), (22,21), (22,21), (22,21), (22,64), (22,64), (22,64), (22,21), (22,21), (22,21), (22,21), (22,21), (22,21), (22,64), (22,21), (22,21), (22,21), (22,64), (22,64), (22,21), (22,64), (22,21), (22,64), (22,64), (22,64), (22,21), (22,21), (22,64), (22,1820)]> to <IOM+ 1514159804=IR Blaster 3 (TV)> repeating 0
Cynical Caché Error Global Cache unavailable: [Errno 60] Operation timed out
Cynical Caché Error IR Blaser 2 unavailable: host device Global Cache is unavailable
Cynical Caché Error IR Blaster 1 unavailable: host device Global Cache is unavailable
Cynical Caché Error IR Blaster 3 (TV) unavailable: host device Global Cache is unavailable
Cynical Caché Error Global Cache connection lost - resetting
Cynical Caché Global Cache connecting to host
Cynical Caché Global Cache is now exploring
Cynical Caché Global Cache enabling idle probes
Cynical Caché Global Cache ready with 4 device(s)
Cynical Caché IR Blaser 2 host device Global Cache now available
Cynical Caché IR Blaser 2 reset
Cynical Caché IR Blaser 2 is now ready
Cynical Caché IR Blaster 1 host device Global Cache now available
Cynical Caché IR Blaster 1 reset
Cynical Caché IR Blaster 1 is now ready
Cynical Caché IR Blaster 3 (TV) host device Global Cache now available
Cynical Caché IR Blaster 3 (TV) reset
Cynical Caché IR Blaster 3 (TV) is now ready
Cynical Caché Global Cache version 3.2-06

Most of the time, it seems that the actual IR transmission does not get through (no blinking green light on the IR output port of my GC-100 that this command is going to). Sometimes it does seem to work (blinking light on my GC-100 and my TV responds correctly)

Thanks,
Darren

For what it's worth, I would say its a network issue, not the plugin.

The only time I have has issues with perry's plugging not working it always ended up being a network issue.

I'm using an iTach WIR mind you, not a GC100.

Is the GC100 at the latest firmware revision?

I wrote the following python script to test this further:

Code: Select all

import socket
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.connect(('10.0.1.70', 4998))
s.settimeout(1.0)

#tv on cmd
cmd = 'sendir,2:3,1,38000,1,1,172,172,22,64,22,64,22,64,22,21,22,21,22,21,22,21,22,21,22,64,22,64,22,64,22,21,22,21,22,21,22,21,22,21,22,64,22,21,22,21,22,64,22,64,22,21,22,21,22,64,22,21,22,64,22,64,22,21,22,21,22,64,22,64,22,21,22,1820\r'

s.sendall(cmd)
result=s.recv(1024)
indigo.server.log(str(result))

It works everytime (20 out of 20) I receive completeir,2:3,1 from the GC-100 which is the correct response and my tv turns on.

So, it seems like a issue with the plugin perhaps or some other more complex network problem...

Checked firmware version of my GC-100. It is 3.2-6, which appears to be the latest per the Global Cache website/release notes.

What if you try the other older Global cache plugin : https://github.com/IndigoDomotics/global-cache

I used that one before Perrys came out.

It looks to me like you possess the skills to modify it as is open source...

Thanks for the suggestion. I took a look at that plugin, but it doesn't support reading IR commands from a GC-IRL learner which I would like to do as well. Perry's plugins seem generally very robust and full featured so I was hoping to get it working... I did look at his code, but its complicated... a bit beyond what I can debug myself. Do you know if Perry is still active here?

I'm afraid ETIMEDOUT (error 60) really is your Mac's way of saying "I sent plenty of packets to the device but it didn't say anything back in over a minute so I'm giving up on this connection." This is not a timeout imposed by the plugin; it's the system network code deciding to give up on you.

Now sometimes these things just happen, so the plugin responds by terminating the connection and re-establishing it. That's meant to deal with occasional hiccups. Clearly your situation isn't occasional.

Check the signal path from you Mac to your GC-100. If there are any WiFi elements, those are your first suspect. Otherwise, unplug and carefully reseat the Ethernet plugs all the way through. Try to replace the Ethernet cables. You'd be surprised how "interesting" a cable with a loose or abused plug can make your network life. Don't entirely ignore your hub/switch; try different ports.

Reset the GC-100 to factory defaults and reprogram it. This is unlikely to help, but you should try it because the next step is...

Replace your GC-100. I've had one go bad exactly like this a few months ago. I didn't go to heroic measures to debug it - feel free to try yours - but it basically kept losing its connection like this. The replacement is working fine...

Cheers
-- perry

This happens to me aswell.

It started when upgrading to 6.10. It has been rock solid before this update

Sometimes it looses the connection to the iTach's until I reload the plugin. It also start to lag sending commands after some time.

This happens to me aswell.

It started when upgrading to 6.10. It has been rock solid before this update

Sometimes it looses the connection to the iTach's until I reload the plugin. It also start to lag sending commands after some time.

So what error are you getting in the log? Let's be clear on what we're seeing.

Cheers
-- perry

@Perry: There is no WiFi in my setup. My Macmini and the GC-100 are connected on my network together in the same cabinet via an 8 port Ethernet switch. As I mentioned earlier in the thread, sending single commands to by GC-100 via my own script works perfectly everytime, so I don't think its a network problem on my side.