Are multiple instances of KiCad the same process?

I had two different instances of KiCad (tonights nightly on fedora 37) open, with different projects.

I did some moves (add a resistor) in one of the schematic editors, and suddenly BOTH instances of KiCad crashed.

Are the different instances in the same process, such that everything crashes at once?

They are different processes. This is easy to see with something comparable to Windows Task Manager: whatever your desktop environment provides, or top/pstree…

Ok, thanks.

I will reword my question: What makes both instances of KiCad crash at the same time, and is this absoultely necessary?

I have no idea.

This is a strange question. What do you say, would it be possible to answer “yes, it’s absolutely necessary that they crash at the same time”?

What I wonder is what design decision in the code makes the both processes interdependent so that both crash at the same time, and consequently when the mechanism is known, if it can be solved such that the result of a crash can be limited to only one instance.

Hope the wording is at least a little bit clearer…

Surely you want to investigate the crash and not whether they crash together or it crashes alone, that’s a distraction.

I think both aspects are of interest.

That manifestation may feel outré but it’s just another piece of evidence. More important is to gain ample evidence through a backtrace or similar, and if the bug could plausibly cause both crashes that another point in favour of that cause. As an example, both processes could be watching a file and both mishandled the notifications.

At the moment it seems the crash does not produce a core dump.

I seem to be able to reasonably consistently be able to reproduce it by the following steps:

Start KiCad nightly
Create new project
Open schematic editor
Start second instance of KiCad nightly
Create another new project
Open schematic editor → Causes crash of both instances immediately

Application: KiCad x86_64 on x86_64

Version: 7.99.0-1.20230624git1c1849e.fc37, release build

wxWidgets 3.2.1
FreeType 2.12.1
HarfBuzz 5.2.0
FontConfig 2.14.1
libcurl/7.85.0 OpenSSL/3.0.9 zlib/1.2.12 brotli/1.0.9 libidn2/2.3.4 libpsl/0.21.1 (+libidn2/2.3.3) libssh/0.10.5/openssl/zlib nghttp2/1.51.0

Platform: Fedora release 37 (Thirty Seven), 64 bit, Little endian, wxGTK, KDE, wayland

Build Info:
Date: Jun 24 2023 17:15:43
wxWidgets: 3.2.1 (wchar_t,wx containers) GTK+ 3.24
Boost: 1.78.0
OCC: 7.6.3
Curl: 7.85.0
ngspice: 40
Compiler: GCC 12.3.1 with C++ ABI 1017

Build settings:

FWIW: according to the devs, nightlies are extremely unstable right now and through the next week or so due to architectural changes.
Might be a factor.
You yourself commented on that :wink:

Yes, I’m aware that there can be / is likely to be instabilities in the nightlies.

But I did not spontaneously think that two different instances of the application would interact in the way that if there is a problem in one of them, both crash.
Evidently I was wrong…

I would have taken that for very unlikely as well …
And maybe it IS unrelated to the changes.

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.