The Julia docker image for 1.6 is available at
https://hub.docker.com/_/julia
2 Likes
I also updated my docker image that’s based on the Julia docker image. This image contains JupyterLab and Pluto already installed, and comes with instructions.
https://hub.docker.com/repository/docker/statisticalmice/julia-jupyter
6 Likes
Any chance to make the Portable version portable for real?
Namely it will use the installation folder for .julia and not USERDATA?
2 Likes
And you have no other Julia version in the search path? And your username does not contain spaces or non-ascii characters?
I’m a simple user, with little computer science knowledge. I typically uninstall Julia, delete all directories and files I find (including jupyter and matplotlib), and install a fresh version.
I know this is probably not what experienced users would recommend, but it works for me. I use “include()” to run a script file where I have listed all packages I use, so I just run a script to re-install everything.
1 Like
For just running Julia, the same instructions of a fresh new installation should work. To transfer the previously installed packages:
2 Likes
Help! How do I find out what the previous error is when I perform an update as seen below
(@v1.6) pkg> up
Updating registry at `~/.julia/registries/General`
Updating git-repo `https://github.com/JuliaRegistries/General.git`
Installed StatsFuns ─ v0.9.7
Installed Rmath ───── v0.7.0
Installed Rmath_jll ─ v0.3.0+0
Downloaded artifact: Rmath
No Changes to `~/.julia/environments/v1.6/Project.toml`
Updating `~/.julia/environments/v1.6/Manifest.toml`
[79098fc4] ↑ Rmath v0.6.1 ⇒ v0.7.0
[4c63d2b9] ↑ StatsFuns v0.9.6 ⇒ v0.9.7
[f50d1b31] ↑ Rmath_jll v0.2.2+2 ⇒ v0.3.0+0
Progress [========================================>] 13/13
12 dependencies successfully precompiled in 56 seconds (280 already precompiled,
1 skipped during auto due to previous errors)
Is there a logfile I can look at to find the so called previous error?
There’s no log file, but you can call pkg> precompile explicitly, which will retry everything and throw errors fully, unlike the auto-precompilation like in your example which doesn’t throw and doesn’t retry package versions that have errored before.
Or you can always do the old way and just go and using/import the packages in your project and get the usual code load outputs.
It was a trifle more complicated than that for me. When I did
[ build IJulia
the build failed with this message
(@v1.6) pkg> build IJulia
Building Conda ─→ `~/.julia/scratchspaces/44cfe95a-1eb2-52ea-b672-e2afdf69b78f/c0647249d785f1d5139c0cc96db8f6b32f7ec416/build.log`
Building IJulia → `~/.julia/scratchspaces/44cfe95a-1eb2-52ea-b672-e2afdf69b78f/0862f73c51b49d80168e75b141a26d1cbb9a7295/build.log`
┌ Warning: Could not use exact versions of packages in manifest, re-resolving
└ @ Pkg.Operations /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Pkg/src/Operations.jl:1524
Downloaded artifact: OpenSpecFun
Precompiling project...
Progress [========================================>] 43/43
✗ OpenSpecFun_jll
✗ SpecialFunctions
? SIAMFANLEquations
✗ FastGaussQuadrature
39 dependencies successfully precompiled in 10 seconds (31 already precompiled)
1 dependency failed but may be precompilable after restarting julia
3 dependencies errored
On one of my three Macs a couple retrys did the job. On the other two a had to do [ rm Ijulia and delete .julia/packages/IJulia to get things to work. This was a real pain for me prior to Julia 1.5, where [ build IJulia worked fine. It’s back to pain with 1.6. I have more experience that I did the first time this happened and know what to do, but it is still annoying.
Hi @giordano , thanks for this info. This is new to me 
Sadly, putting square braces in wget command also fails to download.
$ wget https://[2a04:4e42:2::561]/bin/linux/x64/1.6/julia-1.6.0-linux-x86_64.tar.gz
--2021-03-29 20:26:34-- https://[2a04:4e42:2::561]/bin/linux/x64/1.6/julia-1.6.0-linux-x86_64.tar.gz
Connecting to [2a04:4e42:2::561]:443... failed: No route to host.
Seems like problem of my institution’s firewall.
For now, I downloaded a copy locally and scp’d it to the server.
Just wanted to chime in to say, “Bravo!”
It is a joy to watch the parallel precompilation after pkg> up.
12 Likes
Yes I also love the automatic precompilation in 1.6. This is a fantastic release!
5 Likes
2 posts were split to a new topic: Trouble adding packages: “The connection with the server was terminated abnormally”
Anybody know someone I could reach out to on the Debian Julia team and offer to help get 1.6 packaged?
I believe that you can reach the Debian Julia team at Debian Julia Team <pkg-julia-devel@lists.alioth.debian.org> However, Debian is getting ready for a release before long too long, so getting package updates into unstable may be a low priority until the release happens.
Norbert Preining, the maintainer of the Julia package in Debian, is working on it:
7 Likes