Postcards from Semaphore

Compiling and installing software for the uBITX v6 QRP amateur radio transciever

2021-01-13T11:09:10Z

The uBITX uses an Arduino internally. This article describes how to update its software.

Required hardware

The connector on the back is a Mini-B USB connector, so you'll need a "Mini-B to A" USB cable. This is not the same cable as used with older Android smartphones. The Mini-B connector was used with a lot of cameras a decade ago.

You'll also need a computer. I use a laptop with Fedora Linux installed.

Required software for software development

In Fedora all the required software is installed with sudo dnf install arduino git. Add yourself to the users and lock groups with sudo usermod -a -G users,lock $USER (on Debian-style systems use sudo usermod -a -G dialout,lock $USER). You'll need to log out and log in again for that to have an effect (if you want to see which groups you are already in, then use the id command).

Run arduino as your ordinary non-root user to create the directories used by the Arduino IDE. You can quit the IDE once it starts.

Obtain the uBITX software

$ cd ~/Arduino
$ git clone https://github.com/afarhan/ubitxv6.git ubitx_v6.1_code

Connect the uBITX to your computer

Plug in the USB cable and turn on the radio. Running dmesg will show the Arduino appearing as a "USB serial" device:

usb 1-1: new full-speed USB device number 6 using xhci_hcd
usb 1-1: New USB device found, idVendor=1a86, idProduct=7523, bcdDevice= 2.64
usb 1-1: New USB device strings: Mfr=0, Product=2, SerialNumber=0
usb 1-1: Product: USB Serial
usbcore: registered new interface driver ch341
usbserial: USB Serial support registered for ch341-uart
ch341 1-1:1.0: ch341-uart converter detected
usb 1-1: ch341-uart converter now attached to ttyUSB1

If you want more information about the USB device then use:

$ lsusb -d 1a86:7523
Bus 001 Device 006: ID 1a86:7523 QinHeng Electronics CH340 serial converter

comments

Blocking a USB device

2020-11-13T12:07:40Z

udev can be used to block a USB device (or even an entire class of devices, such as USB storage). Add a file /etc/udev/rules.d/99-local-blacklist.rules containing:

SUBSYSTEM=="usb", ATTRS{idVendor}=="0123", ATTRS{idProduct}=="4567", ATTR{authorized}="0"

comments

Jupyter notebook and R

2019-03-21T12:26:22Z

This has become substantially simpler in Fedora 29:

sudo dnf install notebook R-IRKernel R-IRdisplay

comments

wpa_supplicant update trades off interoperation for security

2019-03-07T14:06:00Z

In case you want to choose a different security compromise, the update has a nice summary:

wpasupplicant (2:2.6-19) unstable; urgency=medium

  With this release, wpasupplicant no longer respects the system
  default minimum TLS version, defaulting to TLSv1.0, not TLSv1.2. If
  you're sure you will never connect to EAP networks requiring anything less
  than 1.2, add this to your wpasupplicant configuration:

    tls_disable_tlsv1_0=1
    tls_disable_tlsv1_1=1

  wpasupplicant also defaults to a security level 1, instead of the system
  default 2. Should you need to change that, change this setting in your
  wpasupplicant configuration:

    openssl_ciphers=DEFAULT@SECLEVEL=2

  Unlike wpasupplicant, hostapd still respects system defaults.

 -- Andrej Shadura <…@debian.org>  Sat, 15 Dec 2018 14:22:18 +0100

comments

Finding git credentials in libsecret

2019-03-05T01:30:25Z

To find passwords in libsecret you need to know what attributes to search for. These are often set by some shim but not documented. The attributes tend to vary by shim.

For git's libsecret shim the attributes are: protocol, server, user.

A worked example, the account gdt on git.example.org:

$ secret-tool search --all 'protocol' 'https' 'server' 'git.example.org' 'user' 'gdt'
[/org/freedesktop/secrets/collection/login/123]
label = Git: https://git.example.org/
secret = CvKxlezMsSDuR7piMBTzREJ7l8WL1T
created = 2019-02-01 10:20:34
modified = 2019-02-01 10:20:34
schema = org.gnome.keyring.NetworkPassword
attribute.protocol = https
attribute.server = git.example.org
attribute.user = gdt

Note that the "label" is mere documentation, it's the "attribute" entries which matter.

comments

MacbookPro8,2 boots with very bright screen

2018-07-29T13:24:39Z

The screen brightness is held in a EFI variable.

hexdump /sys/firmware/efi/efivars/backlight-level-7c436110-ab2a-4bbb-a880-fe41995c9f82 
0000000 0007 0000 0101

comments

Installing NetCDF for R in Fedora

2018-04-13T06:43:40Z

$ sudo install netcdf-devel
$ R --no-restore --no-save
> install.packages('ncdf4')
> q()

There's a good summary of using NetCDF here.

comments

Jupyter notebook for Fedora 27

2018-04-11T12:44:12Z

Jupyter with Python

Fedora loves Python has the good oil:

$ sudo dnf install notebook
$ jupyter notebook

Jupyter with R

Firstly, install Jupyter with Python, as above.

IRkernel has the good oil. I've modified their advice to install RPM packages where they are available:

$ sudo dnf install czmq-devel libcurl-devel openssl-devel
$ sudo dnf install R-crayon R-pbdZMQ R-repr
$ R --no-restore --no-save
> install.packages('devtools')
> devtools::install_github(paste0('IRkernel/', c('IRdisplay', 'IRkernel')))
> IRkernel::installspec()
> q()

If you want to do a system-wide installation then run R as root, proceed as above, but the last step is:

> IRkernel::installspec(user = FALSE)

comments

Checking Meltdown and Spectre fixes, and not thinking about usage

2018-03-16T03:37:38Z

The Linux kernel has fixes for Meltdown, Spectre v1 and Spectre v2. Check the status with:

$ cat /sys/devices/system/cpu/vulnerabilities/meltdown
Mitigation: PTI
$ cat /sys/devices/system/cpu/vulnerabilities/spectre_v1
Mitigation: __user pointer sanitization
$ cat /sys/devices/system/cpu/vulnerabilities/spectre_v2
Mitigation: Full generic retpoline

You'd think that would be enough. But Linux has fucked it up. Here's the output from a Raspberry Pi 3 running 4.9.80. That ARM CPU doesn't have instruction speculation and this isn't vulnerable to Meltdown or Spectre-style bugs.

$ ls /sys/devices/system/cpu/vulnerabilities
ls: cannot access '/sys/devices/system/cpu/vulnerabilities': No such file or directory

That's right. We can't tell if the CPU is immune to the issue or if the Linux kernel is too old to have a fix for the issue. A program wanting to check this is going to have to carry around a table of CPU architectures and Linux versions. Which is likely going to be wrong, because who's going to know the data a fix for Spectre made it into a MIPS build. Classic. All Linux needed to do was to create /sys/devices/system/cpu/vulnerabilities for all architectures. Sigh.

comments

Finding Google DNS cluster

2017-08-28T11:06:51Z

From NANOG list:

dig o-o.myaddr.l.google.com -t txt +short @8.8.8.8

comments

dd

2017-03-31T23:08:55Z

The reason for using dd to write to USB flash sticks is historical.

On some other UNIX-like operating systems only "raw" block devices present the partition table, boot sector and unpartitioned space. Specifically, whereas Linux presents a /dev/sda block device containing all the bytes of a disk, in those other operating systems the equivalent would be a /dev/rsda raw block device.

In those other UNIX-like operating systems you must write to raw block devices in multiples of the sector size of the device. dd can do this, cp and cat cannot. How you discover a device's sector size was left as an exercise for the reader, it is traditionally 0.5KB, more recently 4KB, and three orders of magnitude larger again for flash devices.

Linux doesn't have raw devices, so using dd isn't needed to write an image to a disk. You can wget -O /dev/sd𝐱 … a Fedora .iso file directly onto the USB flash drive.

Note that some devices perform better when handed data is particular block sizes. Most USB sticks perform best if handed data in 4MB chunks. dd is useful if you want that optimisation: wget -O - … | dd of=/dev/sd𝐱 bs=4M status=progress. Note that if you do not set the bs blocksize then the default of 0.5KB is going to make writing a USB flash stick very slow.

comments