[OneRNG-Discuss] OpenSuse installation woes

[Paul Elliott]

Hi all,

I recently switched from Ubuntu to OpenSUSE, and am trying to get the
OneRNG working again.

I initially attempted using the RedHat RPM, which fails (lightly) due
to a difference in naming of udevadm, however even after reloading the
rules it does not seem to work.

 I have also tried the tarball method, and this installs seemingly
correctly (and I am yet to see a difference between this and the rpm),
however it still does not seem to work. This is 3.7, and the led on the
device stays solid orange / red, no blinking.

I have noticed the following in the journal after plugging in the
device:

---
Jan 12 11:03:05 niburu kernel: usb 1-4: USB disconnect, device number
11
Jan 12 11:03:07 niburu kernel: usb 1-4: new full-speed USB device
number 12 using xhci_hcd
Jan 12 11:03:08 niburu kernel: usb 1-4: New USB device found,
idVendor=1d50, idProduct=6086, bcdDevice= 0.09
Jan 12 11:03:08 niburu kernel: usb 1-4: New USB device strings: Mfr=1,
Product=3, SerialNumber=3
Jan 12 11:03:08 niburu kernel: usb 1-4: Product: 00
Jan 12 11:03:08 niburu kernel: usb 1-4: Manufacturer: Moonbase Otago
http://www.moonbaseotago.com/random
Jan 12 11:03:08 niburu kernel: usb 1-4: SerialNumber: 00
Jan 12 11:03:08 niburu kernel: cdc_acm 1-4:1.0: ttyACM0: USB ACM device
Jan 12 11:03:08 niburu systemd-udevd[7370]: 1-4: Process
'/sbin/onerng.sh 1-4' failed with exit code 1.
Jan 12 11:03:08 niburu systemd-udevd[7370]: 1-4:1.0: Process
'/sbin/onerng.sh 1-4:1.0' failed with exit code 1.
Jan 12 11:03:08 niburu systemd-udevd[7371]: 1-4:1.1: Process
'/sbin/onerng.sh 1-4:1.1' failed with exit code 1.
---

and also the following if I attempt to run the script the same way:

---
> bash -x /sbin/onerng.sh 1-4                                         
+ '[' 1-4 = daemon ']'
+ '[' 1-4 = kill ']'
+ '[' 1-4 = feeder ']'
+ '[' '!' -c /dev/1-4 ']'
+ exit 1
---

Now, whilst I haven't sat down to completely understand the script, it
would seem to me that the script is not being invoked 'correctly'. It
passes %k as a parameter, which should resolve to the kernel name,
however this is passing 1-4, NOT ttyACM0, which is what I suspect the
script wants. I must admit I am wondering if this should be %N, the
name of the device node, but I want to get something in the way of
advice before I tinker too much.

This is with kernel version 6.1.3 by the way.

Have you ever seen anything like this before, or managed to cure it?

Many thanks in advance,

Paul