Pkg.precompile error when adding packages in project virtual environment (Julia) - julia

Question

Coming from a Python background, I am now trying to learn Julia, so my problem may be related to my confusion between virtual environments between Python and Julia. In Julia, I experience difficulties adding packages into a project virtual environment. BUT I have no problems adding packages when I am not using a project environment, such as within REPL:
(#v1.6) pkg>add CSV
successfully adds the CSV package. However, when I activate my project environment (called self_learn) and try to add this same package into this environment, I see these error messages within REPL:
(#1.6) pkg> activate .
(self_learn) pkg> st
Project self_learn v0.1.0
Status `D:\Dropbox\Julia\self_learn\Project.toml` (empty project)
(self_learn) pkg> add CSV
Resolving package versions...
Updating `D:\Dropbox\Julia\self_learn\Project.toml`
[336ed68f] + CSV v0.8.4
Updating `D:\Dropbox\Julia\self_learn\Manifest.toml`
[336ed68f] + CSV v0.8.4
[9a962f9c] + DataAPI v1.6.0
[e2d170a0] + DataValueInterfaces v1.0.0
[82899510] + IteratorInterfaceExtensions v1.0.0
[69de0a69] + Parsers v1.1.0
[2dfb63ee] + PooledArrays v1.2.1
[91c51154] + SentinelArrays v1.2.16
[3783bdb8] + TableTraits v1.0.1
[bd369af6] + Tables v1.4.2
[2a0f44e3] + Base64
[ade2ca70] + Dates
[9fa8497b] + Future
[b77e0a4c] + InteractiveUtils
[8f399da3] + Libdl
[37e2e46d] + LinearAlgebra
[56ddb016] + Logging
[d6f4376e] + Markdown
[a63ad114] + Mmap
[de0858da] + Printf
[9a3f8284] + Random
[9e88b42a] + Serialization
[8dfed614] + Test
[4ec0a83e] + Unicode
Precompiling project...
Progress [> ] 0/1
◑ self_learn
┌ Error: Pkg.precompile error
│ exception =
│ ArgumentError: Invalid header in cache file C:\Users\Admin\.julia\compiled\v1.6\self_learn\jl_A3B4.tmp.
│ Stacktrace:
│ [1] preferences_hash(cachefile::String)
│ # Base .\loading.jl:1478
│ [2] compilecache(pkg::Base.PkgId, path::String, internal_stderr::IOBuffer, internal_stdout::Base.DevNull)
│ # Base .\loading.jl:1337
│ [3] (::Pkg.API.var"#215#242"{IOBuffer, String, Base.PkgId})()
│ # Pkg.API C:\buildbot\worker\package_win64\build\usr\share\julia\stdlib\v1.6\Pkg\src\API.jl:1144
│ [4] with_logstate(f::Function, logstate::Any)
│ # Base.CoreLogging .\logging.jl:491
│ [5] with_logger
│ # .\logging.jl:603 [inlined]
│ [6] macro expansion
│ # C:\buildbot\worker\package_win64\build\usr\share\julia\stdlib\v1.6\Pkg\src\API.jl:1143 [inlined]
│ [7] (::Pkg.API.var"#212#239"{Bool, Vector{Task}, Pkg.API.var"#handle_interrupt#231"{Base.Event, ReentrantLock, Base.TTY}, Pkg.API.var"#color_string#229", Base.Event, Base.Event, ReentrantLock, Vector{Base.PkgId}, Vector{Base.PkgId}, Dict{Base.PkgId, String}, Vector{Base.PkgId}, Vector{Base.PkgId}, Dict{Base.PkgId, Bool}, Dict{Base.PkgId, Base.Event}, Dict{Base.PkgId, Bool}, Dict{Base.UUID, Pkg.Types.PackageEntry}, Vector{Base.PkgId}, Bool, Base.TTY, Base.Semaphore, String, Vector{String}, Vector{Base.PkgId}, Base.PkgId})()
◐ self_learn
(self_learn) pkg>
Adding other packages such as DataFrames and Pipe result in the same precompile errors. I have tried uninstalling and reinstalling Julia 1.6.1, but this issue persists. I am using Windows 10.
Any help will be appreciated.

Answer 1

The package cache for your module seems to be stalled.
This hast most likely happened when you were installing packages for your module and hit Ctrl+Enter - the package repo usually does not survive situations too well.
Uninstalling Julia did not help because the package repo is held in a ~/.julia (or %HOME%\.julia on Windows) folder that lives independently of Julia installation.
What you need to do is just to delete the folder: C:\Users\Admin\.julia\compiled\v1.6\self_learn\
If this does not help you might need to install the entire C:\Users\Admin\ but this would require re-installation of all packages.