Precompilation of OrdinaryDiffEq fails

General Usage
1

Hello,
I just noticed that adding the OrdinaryDiffEq to Julia fails at the precompilation step. It worked yesterday. But today, I cannot get it to work. It is not an issue of my machine as I have tested this on multiple ones.

Steps to reproduce:

  1. Delete Julia completely to make sure it is not a problem with my installation.
  2. Delete the .julia folder.
  3. Install Julia via winget install --name Julia --id 9NJNWW8PVKMN -e -s msstore(I am using Windows.)
  4. Open Julia in terminal via julia command.
  5. ]add OrdinaryDiffEq

I get the following output:

  ✗ OrdinaryDiffEqRosenbrock
  ✗ OrdinaryDiffEqDefault
  ✗ OrdinaryDiffEq
Precompiling packages finished.
  131 dependencies successfully precompiled in 338 seconds. 34 already precompiled.
  3 dependencies had output during precompilation:
┌ LinearSolve
│   Downloading artifact: MKL
└
┌ SciMLBase
│  ┌ Warning: Replacing docs for `SciMLBase.AlgorithmInterpretation :: Union{}` in module `SciMLBase.AlgorithmInterpretation`
│  └ @ Base.Docs docs\Docs.jl:249
└
┌ MKL_jll
│   Downloading artifact: IntelOpenMP
│   Downloading artifact: oneTBB
└
  3 dependencies errored.
  For a report of the errors see `julia> err`. To retry use `pkg> precompile`

The err() function outputs:

ERROR: MethodError: objects of type Base.Precompilation.PkgPrecompileError are not callable
The object of type `Base.Precompilation.PkgPrecompileError` exists, but no method is defined for this combination of argument types when trying to treat it as a callable object.
Stacktrace:
 [1] top-level scope
   @ REPL[2]:1

Can anyone confirm this?

2

Can’t reproduce this on Julia 1.12 on Windows - what’s your versioninfo()?

Also note that err is not a function - you need to call err as the message says, not err(), the error you’re showing just comes from the fact that err is not callable, it’s the same as:

ulia> "a"()
ERROR: MethodError: objects of type String are not callable
3

Hi, the problem solved itself for me. I tried the same steps as above again and now precompilation works. It seems to have been just a temporary hickup.

4

Another thread mentioning the same issue does not seem to be resolved yet, though:

https://discourse.julialang.org/t/cant-install-differentialequations/138037/9
5

Well, there are two possible reasons for this issue:

  1. Something is wrong on the local computer
  2. Something is wrong with the package ecosystem, which usually means some packages were updated and others not yet, which can lead to inconsistencies.

The first issue usually can be fixed by creating a local project as described here

The second issue usually disappears if you try again later. Reporting it is nevertheless useful. Sometimes this line of code before launching Julia helps on Linux:

export JULIA_PKG_SERVER_REGISTRY_PREFERENCE=eager