Nic (Mar 10 2026 at 10:48): Nic (Mar 10 2026 at 10:48):Hi together,
I am new to Julia and would like to use it via JuliaCall / the diffeqr package. I got JuliaCall working in R on a Linux server after some environment issues. Initially package installations were segfaulting. After reinstalling Julia via juliaup and unsetting LD_LIBRARY_PATH / LD_PRELOAD / LIBRARY_PATH, JuliaCall now works fine (julia_setup() and julia_eval("1+1") both work).
My current problem is specifically with diffeqr. When I run
de <- diffeqr::diffeq_setup()
(after library(diffeqr)) in R / RStudio, it hangs for a while and then the R session terminates / crashes.
Is this a known issue? Is there a recommended Julia / package version combination that currently works with diffeqr?
Also: as a next step, I would like to use the diffeqr to solve Stochastic Delay Differential Equations (see StochasticDelayDiffEq.jl). In the original Julia package this is possible, but can I also access the same features via diffeqr (since the Github description only mentions "ordinary differential equations (ODEs), stochastic differential equations (SDEs), delay differential equations (DDEs), and differential-algebraic equations (DAEs)")?
Thank you very much in advance and have a great day! :)
EDIT:
Trying to directly use the functonalities in julia, loading/compiling the package StochasticDelayDiffEq seems to make some troubles:
user@server:~$ unset LD_LIBRARY_PATH LD_PRELOAD LIBRARY_PATH
user@server:~$ julia
julia> import Pkg
julia> Pkg.add("DifferentialEquations")
...
✗ StochasticDelayDiffEq
Precompiling packages finished.
1 dependency errored.
For a report of the errors see julia> err
julia> err
PkgPrecompileError: The following 1 direct dependency failed to precompile:
StochasticDelayDiffEq
Failed to precompile StochasticDelayDiffEq to "<path>".
ERROR: LoadError: MethodError: no method matching
OrdinaryDiffEqCore.DEOptions(...)
The type OrdinaryDiffEqCore.DEOptions exists, but no method is defined
for this combination of argument types when trying to construct it.
Stacktrace:
[1] __init(..., alg::StochasticDiffEq.EM{true}, ...)
@ StochasticDelayDiffEq <path>/src/solve.jl:447
[2] __solve(...)
@ StochasticDelayDiffEq <path>/src/solve.jl:6
[3] solve_call(...)
@ DiffEqBase <path>/src/solve.jl:172
...
in expression starting at <path>/StochasticDelayDiffEq/src/precompile.jl
julia> Pkg.precompile("StochasticDelayDiffEq")
[similar error as above]
Christopher Rackauckas (Mar 10 2026 at 14:35):StochasticDelayDiffEq.jl isn't bundled with diffeqr.
Christopher Rackauckas (Mar 10 2026 at 14:35):We could set that up, but it's not curently bundled at least
Nic (Mar 16 2026 at 14:15):Thank you for your fast reply!
But in this case the direct implementation in julia should still work, correct?
Unfortunately I also have problems with the normal import of the DifferentialEquations package in julia, which gives the following error (for a more detailed error description, see the EDIT in my above message):
julia> Pkg.add("DifferentialEquations")
...
✗ StochasticDelayDiffEq
Precompiling packages finished.
1 dependency errored.
(The following error seems to hinder the precompilation:
ERROR: LoadError: MethodError: no method matching OrdinaryDiffEqCore.DEOptions(...)
The type OrdinaryDiffEqCore.DEOptions exists, but no method is defined for this combination of argument types when trying to construct it.)
Nic (Mar 17 2026 at 08:37):Okay, this seems to be a general issue:
https://github.com/SciML/StochasticDelayDiffEq.jl/issues/106
Christopher Rackauckas (Mar 19 2026 at 19:19):Oh yes we're in the middle of restructuring that whole library. It should be fixed by tomorrow and that git repo will be archived. Unfortunate timing but it is being changed to reuse a lot more code which will ultimately keep it in sync with everything else, so we need the temporary break to make it saner
Last updated: Mar 29 2026 at 05:56 UTC