Am 21.08.2019 15:20, schrieb Schmid, Carsten:
>>
>> Different computer but same cables i guess the device is ok.
>>
> But maybe the USB  port of the computer is broken.
> 
>> NTL I found that little gem:
>> https://www.fclose.com/linux-kernels/594677/usb-io_ti-add-heartbeat-to-
>> keep-idle-ep-416-ports-from-disconnecting-linux-4-3/
>>
>> The behavior would fit (any way to check that the heartbeat is active ?).
>>
> Which kernel are you running on?
> use "uname -a" to get it.
> This patch was introduced in 4.3 kernel, not really hot ...
> And, yes, looks like what you experience.
> 

mea culpa, i had a typo in the first msg to greg so you did not see my first post:

I tested with: Opensuse 15.1 on a DELL latitude E5400

 uname -a
 Linux omnfrmo10 4.12.14-lp151.28.10-vanilla #1 SMP Sat Jul 13 17:59:31 UTC 2019 (0ab03b7) x86_64 x86_64 x86_64 GNU/Linux

 I did some experiments (changing cables etc) but always the same. But
 when tested with a windows system it worked all fine.  later i used a
 single port USB->Serial and all worked as expected.



>> Casten Schmidt ask me to look into
>> /sys/kernel/debug/dynamic_debug/control before i start to
>> debug the whole usb part, is that below expected behavior ? (i tried to find
>> what version of edgeport the driver suspects).
>>
>> drivers/usb/serial/io_ti.c:1507 [io_ti]do_boot_mode =_ "%s - STAYING IN BOOT MODE\012"
> ...
>> drivers/usb/serial/io_ti.c:884 [io_ti]i2c_type_bootmode =_ "%s - read 2 status error = %d\012"
>>
> That says that we have dynamic debug messages available for the driver.
> If you enable those by doing
> echo -n 'module io_ti =p' > /sys/kernel/debug/dynamic_debug/control
> 
> Then you should see at least some of those messages appear in dmesg (kernel log).
> 

i see a lot, see attached file