Alohaha,
Has anyone tried playing with a TEMPer USB thermometer under raspian? I am completely unsure what its doing, I can see it pop up with a device id etc in lsusb, so far so good. But trying to install libhidapi-dev results in apt throwing a segfault. The lib is required for whatever the script is to read the device. Perhaps theres a way around this, I guess I can re write the SD card and just start again.

Spec


--- SuperBBS v1.17-3 (Eval)
* Origin: < Scrawled in blood at The Lower Planes > (21:3/101)

29 Jan 20 14:11, you wrote to Anybody:

Alohaha,
Has anyone tried playing with a TEMPer USB thermometer
under raspian? I am completely unsure what its doing, I can see
it pop up with a device id etc in lsusb, so far so good. But
trying to install libhidapi-dev results in apt throwing a
segfault.

that's unusual. Can you post the complete error message / apt output.

The lib is required for whatever the script is to
read the device. Perhaps theres a way around this, I guess I
can re write the SD card and just start again.

If you only installed packages with apt / from the debian/raspbian repository there should be no need to start again.

--- Garbage v1.12💩A44
* Origin: 🦄 🌈 (21:1/151)

If you only installed packages with apt / from the debian/raspbian repository there should be no need to start again.

Famous last words, the sdcard has a tendency to corrupt every now and then, usually during ungraceful power loss, so I'm not writing that lot completely off, but I keep an ISO of the installed image so I can have a quick fix.

that's unusual. Can you post the complete error message / apt output.

root@raspberrypi:/home/pi# apt install libhidapi-dev
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
libhidapi-hidraw0 libhidapi-libusb0
The following NEW packages will be installed:
libhidapi-dev libhidapi-hidraw0 libhidapi-libusb0
0 upgraded, 3 newly installed, 0 to remove and 97 not upgraded.
Need to get 0 B/65.0 kB of archives.
After this operation, 221 kB of additional disk space will be used.
Do you want to continue? [Y/n] y
Illegal instruction
E: Sub-process /usr/bin/apt-listchanges --apt || test $? -lt 10 returned an error code (1)
E: Failure running script /usr/bin/apt-listchanges --apt || test $? -lt 10 root@raspberrypi:/home/pi#

Thats actually a new result, similar sub process error, but not resulting in a segfault today.

Spec


*** THE READER V4.50 [freeware]
--- SuperBBS v1.17-3 (Eval)
* Origin: Scrawled in haste at The Lower Planes (21:3/101)

30 Jan 20 11:01, you wrote to me:

If you only installed packages with apt / from the
debian/raspbian repository there should be no need to start
again.

Famous last words, the sdcard has a tendency to corrupt every
now and then, usually during ungraceful power loss, so I'm not
writing that lot completely off, but I keep an ISO of the
installed image so I can have a quick fix.

Okay, if the fs is corrupted this might indeed be the easiest fix.

that's unusual. Can you post the complete error message / apt
output.

root@raspberrypi:/home/pi# apt install libhidapi-dev
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
libhidapi-hidraw0 libhidapi-libusb0
The following NEW packages will be installed:
libhidapi-dev libhidapi-hidraw0 libhidapi-libusb0
0 upgraded, 3 newly installed, 0 to remove and 97 not upgraded.
Need to get 0 B/65.0 kB of archives.
After this operation, 221 kB of additional disk space will be
used. Do you want to continue? [Y/n] y Illegal instruction
E: Sub-process /usr/bin/apt-listchanges --apt || test $? -lt 10
returned an error code (1)
E: Failure running script /usr/bin/apt-listchanges --apt || test
$? -lt 10 root@raspberrypi:/home/pi#

Thats actually a new result, similar sub process error, but not
resulting in a segfault today.

I never saw this error in the 20+ years I'm using debian based distros. Some people have fixed it with an reinstall of apt-listchanges. But if there were segfaults before this is maybe just a symptom of some other problem (like fs corruption or something completely different).


---
* Origin: 🦄 🌈 (21:1/151)

I never saw this error in the 20+ years I'm using debian based
distros. Some people have fixed it with an reinstall of
apt-listchanges

I'm unsure how you'd do that without apt, unless I go delving into dpkg.. and suck in a package manually... So far as I can tell I don't have any other symptoms going on, I think, the fs is in fact clean. But its so blinkin' hot here at the moment, I haven't looked sideways at it.

Spec


--- SuperBBS v1.17-3 (Eval)
* Origin: < Scrawled in blood at The Lower Planes > (21:3/101)

On 30 Jan 2020 at 11:01a, Spectre pondered and said...

root@raspberrypi:/home/pi# apt install libhidapi-dev
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
libhidapi-hidraw0 libhidapi-libusb0
The following NEW packages will be installed:
libhidapi-dev libhidapi-hidraw0 libhidapi-libusb0
0 upgraded, 3 newly installed, 0 to remove and 97 not upgraded.
Need to get 0 B/65.0 kB of archives.
After this operation, 221 kB of additional disk space will be used.
Do you want to continue? [Y/n] y
Illegal instruction

^^^^^^^^^^^^^^^^^^^

E: Sub-process /usr/bin/apt-listchanges --apt || test $? -lt 10 returned an error code (1)
E: Failure running script /usr/bin/apt-listchanges --apt || test $? -lt
10 root@raspberrypi:/home/pi#

I've highlighted the critical bit of your error: the "Illegal instruction" error means that your machine came across a primitive instruction that it didn't understand while it was running some program; apt-listchanges,
perhaps, though the error indicates that that is a shell script:
regardless, something invoked a program that had an instruction that the processor didn't understand. The processor will generate a synchronous exception in this case, and the OS will catch that and turn it into a
signal (SIGILL); the default behavior for handling that signal is to
terminate the process "with additional actions" (usually generation of a
core dump file for debugging).

There are several reasons one could encounter an illegal signal in a
program. Filesystem corruption is certainly one of them. Another might
be a mismatch between the kernel and hardware; for systems like the
ARM-based Raspberry Pi, where there are many different profiles of processors with different functionality (say, whether a system supports hardware
floating point...) it may be normal to trap and emulate instructions
present on a newer generation of processor if running on an older
processor.

I would check for corruption, though: check the `dmesg` log to see if it
has any indication of which binary was actually killed, and if so, checksum that binary (shasum /what/ever) and compare against the checksum on a known-good system.

Running all of this stuff on consumer-grade SD cards is a
bit of a bummer, though; they are fine for boot media or
on-demand write-mostly workloads like cameras, but they wear
out quickly under general-purpose use. The Raspberry Pi,
in particular, is rather unfortunate in this regard since it
doesn't have a great story for storage peripherals: you are
constrained to running over a USB bridge, which means that
you're bottlenecked by the performance of the USB hardware.

The Pi4 has enough juice to run a USB<->SSD bridge at reasonable
speed but another SBC with an M.2 interface or even native SATA
will be faster and more robust.

--- Mystic BBS v1.12 A44 2020/01/29 (Windows/32)
* Origin: Agency BBS | Dunedin, New Zealand | agency.bbs.nz (21:1/101)