Serial Connection Not Working

[jcarlson23]

Evening

I've tried to connect using the Debug Console cable (USB to serial) to my raspberry pi and am not seeing anything sensible. Looks like a bad baud rate but I've checked several times.

I followed the tutorial (mostly) and when I try to login over ZOC I get a bunch of gibberish...

You can see the ZOC settings, login attempt and the cable connections at the following URL:

http://gototheboard.com/adafruit/

The driver I used was slightly different than the tutorial because that didn't work at all for me, (couldn't see the /dev/cu.PL2303-00001014 device) so:


$ kextstat | grep -i 303
109 0 0xffffff7f80935000 0x8000 0x8000 com.prolific.driver.PL2303 (2.0.0) <108 33 5 4 3>
122 0 0xffffff7f825d8000 0xb000 0xb000 nl.bjaelectronics.driver.PL2303 (1.0.0d1) <108 33 5 4 3>


That allowed me to see the serial device very nicely..

Any help would be greatly appreciated.

- jared

[adafruit_support_bill]

There is some kind of data being received, which would tend to indicate a baud-rate problem. Have you tried changing the baud-rate in ZOC to see if there is a setting that yields more intelligible output?

[jcarlson23]

Yeah, I've tried a bunch (see snippet below) and aside from a slight change in the form it's pretty similar. I guess the first thing I'd like to test is the cable/driver setup. Do you know of any way to test the cable/driver with something common that ought to give me a good signal? TX/RX coming out of an Arduino? Something similar?? I would think the driver would be an all or nothing; it's either working and seeing the device or not, but at this point I'm at a loss.

new-host-5:src jcarlson$ screen /dev/cu.PL2303-00001014 115200
[screen is terminating]
new-host-5:src jcarlson$ screen /dev/cu.PL2303-00001014 230400
[screen is terminating]
new-host-5:src jcarlson$ screen /dev/cu.PL2303-00001014 460800
[screen is terminating]
new-host-5:src jcarlson$ screen /dev/cu.PL2303-00001014 57600
[screen is terminating]
new-host-5:src jcarlson$ screen /dev/cu.PL2303-00001014 38400
[screen is terminating]
new-host-5:src jcarlson$ screen /dev/cu.PL2303-00001014 19200
[screen is terminating]
new-host-5:src jcarlson$ screen /dev/cu.PL2303-00001014 4800
[screen is terminating]
new-host-5:src jcarlson$ screen /dev/cu.PL2303-00001014 921600
[screen is terminating]
new-host-5:src jcarlson$

[adafruit_support_bill]

If you have an Arduino, it should be simple to set up a test. You can use SoftwareSerial to send serial data on any pair of pins. (If it is a Mega, you need to choose pins that support pin-change interrupts).

[jcarlson23]

Ok, that was a good test to do... So, it looks to me like something really screwy, I'm guessing there's something at the OS/driver layer going on. But let me know what you think? So, two screenshots, the first is just to verify the arduino test. http://gototheboard.com/adafruit/arduino-test.png pins 10,11 having RX/TX and Ground of course, I used the serial monitor from the Arduino tools to verify the serial test was working correctly. Then I "screened" over and saw a very strange looking dump, that looks like a Security.Framework issue... http://gototheboard.com/adafruit/photo.jpg

If you have any ideas, I'd love to hear them but starting to sound more and more like a Mac OS X driver issue; as I said, I tried the tutorial on 10.8.2 and that I couldn't even connect with...

- Jared

[csomerlot]

I also ordered the USB cable and am trying to connect with ZOC, from Windows7, and I get no response, irregardless of baud rate. I just see:

[SERIAL/DIRECT] CONNECTED TO PORT \\.\COM9 (19200-8N1)

And then nothing. I confirmed I have the right COM port, and I followed everything in the tutorial, but I do not have an arduino so I need some other way to test the connection - any suggestions?

[adafruit_support_bill]

That does look pretty strange. If the Arduino is only sending "Goodnight Moon". even a serious baud rate mismatch would not result in that much output. I agree, the problem is on the computer/driver end of things.

[jcarlson23]

Just to confirm so others can see the issue, or at least verify it.. do:

$ sudo dmesg | tail -n 40

And if you've recently tried to connect, you'll see a...

nl_bjaelectronics_driver_PL2303(0xffffff8051f04a00)::dequeueDataGated - INTERRUPTED
nl_bjaelectronics_driver_PL2303(0xffffff8049136e00)::dequeueDataGated - INTERRUPTED

Or something to that effect. This also explains why Zoc hangs on me. I'm going to try to find some driver source, but I'm on a 10.8.3 (Dec 20 update) on a retina and so there's some issue on the driver, doesn't appear to be cable related at all..

If I find a solution, I'll post.

[jcarlson23]

For anyone interested, the osxpl2303 driver can be used to reproduce the issue, from source. My problem is to properly debug I need the USB Prober and accompanying kernel extensions for my OS and having upgraded to 10.8.3 on the hopes that it might I now have to wait until the logging release accompanies the OS (you need the same version). So if this does happen to anyone with an older system you can install the 10.8.2 logging and the src for the driver and start in. Given my time frame I'll move over to an old machine, install Ubuntu and start over from there.