moving jupyter (ihaskell) folder off of dropbox

Question

I have been working with Jupyter notebooks in a directory inside of my dropbox. (This work is actually in IHaskell, but I am not sure if this is relevant.)

The folder got too big for dropbox, and so I moved it out. After that, I can't get the kernel to start on any notebooks. It dies and dies.

It seems that moving the folder that contained IHaskell has messed up my installation.

Do I have to re-install everything, or is there a fix?

Answer 1

It dies and dies

If the IHaskell kernel keeps dying, I would follow the suggestion mentioned on the IHaskell Troubleshooting guide (last paragraph of the homepage), especially the "kernel keeps dying" paragraph:

If you've e.g. installed an lts-10 IHaskell and are using it with an lts-9 project the mismatch between GHC 8.2 and GHC 8.0 will cause this error.

If this is your case, I would:

Verify that the haskell kernel is available

$ jupyter kernelspec list
Available kernels:
  haskell    [...]/kernels/haskell      <-- OK
  scala      [...]/kernels/scala
  python3    [...]/kernels/python3

Verify that the ghc version is matching

$ cat [...]/kernels/haskell/kernel.json
.../.stack/programs/.../ghc-X.Y.Z/lib/ghc-X.Y.Z"...

$ stack ghc -- --version
The Glorious Glasgow Haskell Compilation System, version X.Y.Z

If the versions mismatch

Stack also has the notion of a 'global project' located at ~/.stack/global-project/ and the stack.yaml for that project should be on the same LTS as the version of IHaskell installed to avoid this issue.

I would keep the lts-A.B consistent:

$ cat #HOME/.stack/global-project/stack.yaml
...
resolver: lts-A.B

$ stack install ihaskell --resolver lts-A.B

To choose the lts-A.B / ghc-X.Y.Z combinations right for you, you can simply use https://www.stackage.org/lts-A.B

---

Today, for instance, you can simply use the lts-11.4