# Another mystery in IJulia

**URL:** https://discourse.julialang.org/t/another-mystery-in-ijulia/135017
**Category:** General Usage
**Created:** [January 13, 2026, 8:35am UTC](https://discourse.julialang.org/t/another-mystery-in-ijulia/135017 "2026-01-13T08:35:00Z")
**Posts on this page:** 4
**Page:** 1

<div class="post-metadata">

### Author: ![lewis](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/lewis/32/5217_2.png) [@lewis](https://discourse.julialang.org/u/lewis)
#### Post date: [January 13, 2026, 8:35am UTC](https://discourse.julialang.org/t/another-mystery-in-ijulia/135017/1 "2026-01-13T08:35:00Z")

</div>

What does this mean? This occurs in a jupyter notebook when I try to use a local package added to my packages with Pkg.develop().

I do `using mypkg` and the result is:

```julia-auto

SYSTEM: caught exception of type :MethodError while trying to print a failed Task notice; giving up

```

What’s weird is that the package has been pre-compiled successfully and no packages have changed, been updated, been removed. It is the environment that is active that causes the problem. In one case, the base environment works, But the project environment seems not to work. But, both environments contain almost the same environment. I think the difficulty is that it is impossible for Julia to compile the package in the package’s development environment. Just a theory.

Well, I have to reject my theory. I have run the package in the base project (“@v1.12”) in the project’s own local development directory with that environment. It seems like Julia has to attempt precompile a couple of times and then it is ok.

For each experiment, I restart the kernel and then go cell by cell except when I get to Pkg.activate: there I have a choice of Pkg.activate()–\>the base environment. Or Pkg.activate(“.”)–\>the project’s own development directory. I only run ONE of these; then using CovidSim\_Ilm.

Eventually, when restarting the kernel, I can activate either environment and “using” is successful and everything works. As I said–another mystery. I think Julia seems to treat local packages whcih are developed rather than added as a stepchild with some undefined behavior.

Is it possible to add a local project? Might this work better?

I have been stymied a lot by environments since version 1.11 of Julia.

---

<div class="post-metadata">

### Author: ![nilshg](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/nilshg/32/2283_2.png) [@nilshg](https://discourse.julialang.org/u/nilshg)
#### Post date: [January 13, 2026, 9:21am UTC](https://discourse.julialang.org/t/another-mystery-in-ijulia/135017/2 "2026-01-13T09:21:11Z")

</div>

That’s an IJulia bug:

> <https://github.com/JuliaLang/IJulia.jl/issues/1230>
>
> This happens whenever a package is precompiled, I believe since I've switched to… Julia 1.12.
> \`\`\`
> In \[1\]: using Combinatorics
> 
> \[Info: Precompiling Combinatorics \[861a8166-3701-5b0c-9a16-15d98fcdc6aa\] (cache misses: incompatible header (10))
> 
> SYSTEM: caught exception of type :MethodError while trying to print a failed Task notice; giving up
> \`\`\`
> I'm using IJulia v1.33.0.
> \`\`\`
> Julia Version 1.12.2
> Commit ca9b6662be4 (2025-11-20 16:25 UTC)
> Build Info:
> Official https://julialang.org/ release
> Platform Info:
> OS: Linux (x86\_64-linux-gnu)
> CPU: 4 × Intel(R) Core(TM) i3-10110U CPU @ 2.10GHz
> WORD\_SIZE: 64
> LLVM: libLLVM-18.1.7 (ORCJIT, skylake)
> GC: Built with stock GC
> Threads: 1 default, 1 interactive, 1 GC (on 4 virtual cores)
> Environment:
> JULIA\_DEPOT\_PATH = /usr/local/julia-depot:
> JULIA\_LOAD\_PATH = /home/matthias314/Julia:@:@v#.#:@m3-dev:@m3:
> \`\`\`

which shouldn’t affect your usage of whatever package is precompiling.

---

<div class="post-metadata">

### Author: ![lewis](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/lewis/32/5217_2.png) [@lewis](https://discourse.julialang.org/u/lewis)
#### Post date: [January 13, 2026, 9:28pm UTC](https://discourse.julialang.org/t/another-mystery-in-ijulia/135017/3 "2026-01-13T21:28:08Z")

</div>

Any idea how to stop it? Build a new kernelspec?

---

<div class="post-metadata">

### Author: ![nilshg](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/nilshg/32/2283_2.png) [@nilshg](https://discourse.julialang.org/u/nilshg)
#### Post date: [January 14, 2026, 8:50am UTC](https://discourse.julialang.org/t/another-mystery-in-ijulia/135017/4 "2026-01-14T08:50:31Z")

</div>

Well it’s an open issue, I haven’t looked into it but I think your options are:

- Ignore and wait for it to be fixed;
- make a PR to IJulia fixing it; or
- revert to an older IJulia (and presumably Julia) version that predates the bug.
