Sundials.jl errored during testing.

[iajzenszmi]

[IDAS ERROR] IDACalcIC
Newton/Linesearch algorithm failed to converge.

Test Summary: | Pass Error Total Time
Common Interface | 72 1 73 1m56.0s
CVODE | 30 30 50.8s
ARKODE | 4 4 6.1s
IDA | 14 14 21.4s
Jacobians | 9 9 10.1s
Callbacks | 9 9 10.2s
Iterator | 2 1 3 1.7s
Errors | 1 1 2.1s
Mass Matrix | 1 1 2.8s
Preconditioners | 2 2 10.7s
ERROR: LoadError: Some tests did not pass: 72 passed, 0 failed, 1 errored, 0 broken.
in expression starting at /home/ian/.julia/packages/Sundials/UdHby/test/runtests.jl:60
ERROR: Package Sundials errored during testing
Describe the bug 🐞

A clear and concise description of what the bug is.

Expected behavior

A clear and concise description of what you expected to happen.

Minimal Reproducible Example 👇

Without MRE, we would only be able to help you to a limited extent, and attention to the issue would be limited. to know more about MRE refer to wikipedia and stackoverflow.

Error & Stacktrace ⚠️

Environment (please complete the following information):

• Output of using Pkg; Pkg.status()

• Output of using Pkg; Pkg.status(; mode = PKGMODE_MANIFEST)

• Output of versioninfo()

Additional context

Add any other context about the problem here.

[iajzenszmi]

/'GPT

It appears you've encountered a complex issue while working with Julia and the Sundials package. The errors you're facing seem to be related to package dependencies and specific functionalities within the Sundials package. Let's break down the main issues and potential solutions:

Package Dependency and Compatibility Issue:
The error log starts with a dependency conflict involving the Compat package. This conflict arises because different packages you are trying to use (like FortranFiles, ClimateTasks, and DataStructures) require different versions of Compat, which are not compatible with each other.

To resolve this, you might need to downgrade or upgrade certain packages to ensure they all depend on a compatible version of Compat. This can be tricky, as it often involves a bit of trial and error to find a set of package versions that work well together.

You can try updating all packages with ] update in the Julia REPL, and then try adding Sundials again. If that doesn't work, you might have to manually adjust the versions of the conflicting packages.

Sundials Testing and Functionality Issues:
Later in the log, there are errors related to specific functionalities within the Sundials package. These seem to be arising from internal operations of the package, such as during the solve function call, jacobian matrix calculations, etc.

These errors might be due to bugs in the Sundials package or incompatibilities between Sundials and other packages or Julia versions. You can consider the following steps:
    Check if you are using the latest version of Sundials that is compatible with your Julia version.
    Search for or report these issues on the Sundials Julia package GitHub repository. The maintainers or community might have fixes or workarounds.
    Ensure that your code is using the Sundials package correctly. Reviewing documentation or examples might help identify any misuse.

Julia and Package Versions:
Make sure you are using a Julia version that is compatible with the Sundials package version you're trying to use. Sometimes, using the latest version of Julia can cause compatibility issues with packages that have not yet been updated.

Specific Errors in Sundials Tests:
The errors occurring during the test phase of Sundials (like CVODE and ARKode errors) might be specific to the package's internal test suite. If you are just trying to use Sundials for your projects, these errors might not affect you unless you are using the specific features that are failing.

In summary, resolving this issue will likely involve a combination of updating packages, ensuring compatibility, and possibly seeking assistance from the Sundials user community if the problems are due to bugs or undocumented features.`/

[ChrisRackauckas]

]st?

[iajzenszmi]

[IDAS ERROR] IDACalcIC
Newton/Linesearch algorithm failed to converge.

Test Summary: | Pass Error Total Time
Common Interface | 72 1 73 1m56.0s
CVODE | 30 30 50.8s
ARKODE | 4 4 6.1s
IDA | 14 14 21.4s
Jacobians | 9 9 10.1s
Callbacks | 9 9 10.2s
Iterator | 2 1 3 1.7s
Errors | 1 1 2.1s
Mass Matrix | 1 1 2.8s
Preconditioners | 2 2 10.7s
ERROR: LoadError: Some tests did not pass: 72 passed, 0 failed, 1 errored, 0 broken.
in expression starting at /home/ian/.julia/packages/Sundials/UdHby/test/runtests.jl:60
ERROR: Package Sundials errored during testing

(@v1.10) pkg>

[ChrisRackauckas]

Please show ]st. Is this a clean environment with nothing else?

[ChrisRackauckas]

Looks like master is reproducing it. #451

[iajzenszmi]

[IDAS ERROR] IDACalcIC
Newton/Linesearch algorithm failed to converge.

Test Summary: | Pass Error Total Time
Common Interface | 72 1 73 1m56.0s
CVODE | 30 30 50.8s
ARKODE | 4 4 6.1s
IDA | 14 14 21.4s
Jacobians | 9 9 10.1s
Callbacks | 9 9 10.2s
Iterator | 2 1 3 1.7s
Errors | 1 1 2.1s
Mass Matrix | 1 1 2.8s
Preconditioners | 2 2 10.7s
ERROR: LoadError: Some tests did not pass: 72 passed, 0 failed, 1 errored, 0 broken.
in expression starting at /home/ian/.julia/packages/Sundials/UdHby/test/runtests.jl:60
ERROR: Package Sundials errored during testing

(@v1.10) pkg> st
Status ~/.julia/environments/v1.10/Project.toml
[c3572dad] Sundials v4.23.2

(@v1.10) pkg>

[ChrisRackauckas]

Handled by SciML/DiffEqBase.jl#1007

[iajzenszmi]

│ 1. f(du, u) or f(u) is defined, in-which case p and/or t should not
│ be supplied.
│ 2. f(args...) is defined, in which case hasmethod can be spurious.
│
│ Currently, we perform the check for f(du, u) and f(u) first, but in
│ future breaking releases, this check will be performed last, which means
│ that if t is provided f(du, u, p, t)/f(u, p, t) will be given
│ precedence, similarly if p is provided f(du, u, p)/f(u, p) will be
│ given precedence.
│ caller = JacVec(f::Function, u::Vector{Float64}, p::SciMLBase.NullParameters, t::Nothing; fu::Nothing, autodiff::AutoForwardDiff{nothing, Nothing}, tag::SparseDiffTools.DeivVecTag, use_deprecated_ordering::Val{true}, kwargs::Base.Pairs{Symbol, Union{}, Tuple{}, @NamedTuple{}}) at jaches_products.jl:265
└ @ SparseDiffTools ~/.julia/packages/SparseDiffTools/zWIg7/src/differentiation/jaches_products.jl:265
Jacobian vs no Jacobian difference:
7.904787935331115e-14
WARNING: Method definition testjac(Any, Any, Any, Any, Any) in module Main at /home/ian/.julia/packages/Sundials/jrHAF/test/common_interface/jacobians.jl:67 overwritten at /home/ian/.julia/packages/Sundials/jrHAF/test/common_interface/jacobians.jl:103.
WARNING: Method definition testjac_jac(Any, Any, Any, Any, Any, Any) in module Main at /home/ian/.julia/packages/Sundials/jrHAF/test/common_interface/jacobians.jl:73 overwritten at /home/ian/.julia/packages/Sundials/jrHAF/test/common_interface/jacobians.jl:109.
[ Info: 0.044464449958102746
[ Info: 0.1199702927305579
[ Info: 0.19547613550301304
[ Info: 0.2709819782754682
[ Info: 0.34648782104792336
[ Info: 0.4219936638203785
[ Info: 0.5889202251875935
[ Info: 0.7558467865548084
[ Info: 0.9227733479220234
[ Info: 1.0
Test error handling

[CVODES WARNING] CVode
Internal t = 0 and h = 0 are such that t + h = t on the next step. The solver will continue anyway.

[CVODES ERROR] CVode
At t = 0 and h = 0, the corrector convergence test failed repeatedly or with |h| = hmin.

[IDAS ERROR] IDACalcIC
Newton/Linesearch algorithm failed to converge.

Test Summary: | Pass Total Time
Common Interface | 77 77 1m51.4s
Test Summary: | Pass Total Time
Interpolation | 102 102 3.2s
Testing Sundials tests passed

[ChrisRackauckas]

Yup passes now and errors in there are expected error tests. Let me know if you have any issues.