KiСAD drops out during tracing

[Dibaltic]

I’m having this issue where Kicad crashes when I place any track on any layer. I can place footprints and vias without an issue. And it happens in a brand new blank project as well. Below is the Kicad info and the error i get when launched from the terminal. Is there something I can do to fix this?

Application: KiCad x86_64 on x86_64
Version: 9.0.5, release build
Libraries:
    wxWidgets 3.2.8
    FreeType 2.14.1
    HarfBuzz 12.1.0
    FontConfig 2.17.1
    libcurl/8.16.0 OpenSSL/3.5.4 zlib/1.3.1 brotli/1.1.0 zstd/1.5.7 libidn2/2.3.7 libpsl/0.21.5 libssh2/1.11.1 nghttp2/1.67.1 nghttp3/1.12.0
Platform: EndeavourOS, 64 bit, Little endian, wxGTK, Wayland, KDE, wayland
OpenGL: NVIDIA Corporation, NVIDIA GeForce RTX 4080 SUPER/PCIe/SSE2, 4.6.0 NVIDIA 580.82.09
Build Info:
    Date: Oct 2 2025 07:18:54
    wxWidgets: 3.2.8 (wchar_t,wx containers) GTK+ 3.24
    Boost: 1.88.0
    OCC: 7.9.1
    Curl: 8.16.0
    ngspice: 45
    Compiler: GCC 15.2.1 with C++ ABI 1020
Build settings:
    KICAD_USE_EGL=ON
    KICAD_IPC_API=ON

/usr/include/c++/15.2.1/bits/stl_vector.h:1282: constexpr std::vector<_Tp, _Alloc>::const_reference std::vector<_Tp, _Alloc>::operator[](size_type) const [with _Tp = VECTOR2<int>; _Alloc = std::allocator<VECTOR2<int> >; const_reference = const VECTOR2<int>&; size_type = long unsigned int]: Assertion '__n < this->size()' failed.
Aborted                    (core dumped) kicad

[tama]

@Alex2

the program crashes.

When discussing crashes in a Windows environment on a forum, it would be a good idea to use the Windbg tool to show the function call stack trace of the crash dump to show which program is crashing, along with a clear string of characters. This will help open everyone’s eyes and ensure that everyone is experiencing the same issue.

Windows WinDbg · Wiki · KiCad / KiCad Source Code / kicad · GitLab

Unable to run kicad-nightly-9.99.0.2573.g5895c27f4a-x86_64

If you start cmd.exe with administrator privileges and attach the following execution results, you may be able to get more details. wevtutil qe Application /c:5 /f:Text /rd:true "/q:*[System[(EventID=1000)]]" dir %appdata%\kicad dir %LOCALAPPDATA%\CrashDumps sfc /scannow If you get WinDbg and provide the traceback information of the crash dump, you can visualize the sequence of function calls and the cause of the abnormal termination. Install only WinDbg from here. (Win10) Windows SDK - Wi…

Instead of posting to the forum, we can also report the issue directly to the developer. When doing so, as we have added this time, if attach a small circuit diagram or board diagram, it will be easier for the developer to reproduce the same environment.
We also recommend searching for similar issues before reporting.

Report an Issue

I found a bug. What now?

As this is a user forum, the users here can only try to offer a workaround for your problem not a fix. This means there might be a reason to post here but make it clear that you ask for help fixing your problem via a workaround. Bug in the software To help the kicad development effort it is suggested to post your bug description over at the bugtracker. A detailed explanation on how this is done can be found in the report a bug section of the KiCad website. Another detailed description about h…

[DeBa]

I have the same issue. With KiCad 9.0.5 the example completely crashes KiCad.

[RaptorUK]

@tama

When discussing crashes in a Windows environment on a forum

Even more important/helpful ensure you have Anonymous reporting enabled

[anemo]

Same issue for me running KiCad 9.0.5 on MacOS 12.7.6 Monterey. Downgrading to 9.0.4 solved it for now.

[Dr_Panda]

Did you use the provided test zip or created a new Project with the same Parts?

I just want to update to 9.0.5 ( MacOS ), but if it happens also with a new project I wait for 9.0.5.1.

[davidsrsb]

9.0.6 should be the next release

Meanwhile there are daily Testing builds for MacOS, https://downloads.kicad.org/kicad/macos/explore/9.0-testing/download/kicad-unified-universal-20251009-0859-57309d4a62.dmg is the latest, which should fix this

[anemo]

This happened to a project that was initiated in 9.0.4. After downgrading all was fine again.

[paulvdh]

So there are at least 5 people who have already mentioned KiCad V9.0.5 crashing during routing.

@Alex2

When I route from P1 to P2 or from P2 to the resistor, the program crashes.

@Fllorente

After upgrading to 9.0.5, when start tracing on PCB editor, it crashes.

@Dibaltic

I’m having this issue where Kicad crashes when I place any track on any layer

@DeBa

I have the same issue. With KiCad 9.0.5 the example completely crashes KiCad.

@anemo

Same issue for me running KiCad 9.0.5 on MacOS 12.7.6 Monterey.

And all that happened even before the official announcement of V9.0.5 by Wayne Stambaugh: Stable version 9.0.5 available Has this already been reported on gitlab? I can’t / won’t do it myself because I am still using KiCad V8.0.6 myself.

If anyone reports it, please also create a post in this thread with a direct link to the gitlab issue.

[Claudio.Lorini]

someone has now,

[9.0.5] pcbmew CRASH related to ‘Highlight collisions' → 'allow drc...

i’m always reluctant to add possible noise on the tracker, but as far as i can see none has taken the nuisance to directly report this…

[DaveL]

With the detail that the router needs to be in Highlight collision mode, I can reproduce this crash on all the designs I tried it on with 9.0.5 on Win10. Shove and WalkAround seem fine which is why I couldn’t reproduce on earlier attempts on the example files here.

I downloaded and installed 9.0.5.67 testing build and it looks to me to be resolved. I had zero crashes under the same test conditions where I could crash 9.0.5 release pretty much on command.

[DeBa]

It think this is related to this issue:
Seg fault when pressing ‘x’ in PCB editor (#21883) · Issue · kicad/code/kicad

[RaptorUK]

But this was apparently fixed 3 days ago . . . I guess you didn’t try the build suggested ?

Crashes during routing

This should be fixed in post-9.0.5 testing builds if you can use those (flatpak?)

[Claudio.Lorini]

Yes i can confirm that with:

Version: 9.0.5-unknown-202510100552~baf4843a45~ubuntu24.04.1 release build

the problem seem to be resolved.

[hedgehog]

I wonder If such a “prominent” problem affecting lots of people (no edge case! ) could be addressed by a follow-up release without waiting for about a month or two until 9.0.6

Guess that won’t happen and I am fine with that. Just wondering… that could spare some more Forum posts on the very same topic

[gkeeth]

@hedgehog

a follow-up release without waiting for about a month or two until 9.0.6

The follow-up release would be 9.0.6, it would just come sooner than a month or two.

[paulvdh]

@hedgehog

I wonder If such a “prominent” problem affecting lots of people (no edge case! ) could be addressed by a follow-up release without waiting for about a month or two until 9.0.6

Yes, this sometimes does happen. A few years back there was such an egregious error in one of the bug fix releases, that it got completely pulled and the next bug fix release was released just a few days later. Whether this routing bug (apparently only in “highlight collisions mode”) is severe enough for that I do not know.

[tama]

I feel like if the RC had been announced and more people had used it early on, we could have fixed bugs like this that were immediately apparent after the release.
I think the root cause is that there is something missing in this flow.

Stable Release Policy | Developer Documentation | KiCad

By the way, since “September 30th” is already in the past, I feel like the file format probably won’t change in the nightly release, but I’m not sure.

[DaveL]

I did run 9.0.5 rc1 and 9.0.5 release. I tried to replicate the reported problem on both sample projects and couldn’t. It wasn’t until the detail about it only happening in highlight collision mode that made it easily reproducible. 98% of the time, I just don’t use that mode so in addition to more people testing, it would need people testing with workflows they don’t normally utilize.

[joojala]

The problem is that if you have a few constitutive releases with some pretty high profile bugs then you don’t get those testers.

The problem is that if most people find these versions do not upgrade, then they can not participate in bug hunting because they have a catch to this. If they find a bug they are instructed to upgrade to possibly another bug which puts bug reporting into counterproductive territory.