How can I get any diagnostic information on why this one CI test seems to be hanging while all the others pass as expected?
in progress checks
CI / Julia nightly - macOS-latest - default - pull_request (pull_request)Started 39 minutes ago — This check has started...
successful checks
CI / Documentation (pull_request)Successful in 1m
CI / Julia 1 - macOS-latest - default - pull_request (pull_request)Successful in 3m
CI / Julia 1 - ubuntu-latest - default - pull_request (pull_request)Successful in 5m
CI / Julia 1 - windows-latest - default - pull_request (pull_request)Successful in 6m
CI / Julia 1.8 - macOS-latest - default - pull_request (pull_request)Successful in 3m
CI / Julia 1.8 - ubuntu-latest - default - pull_request (pull_request)Successful in 4m
CI / Julia 1.8 - windows-latest - default - pull_request (pull_request)Successful in 4m
CI / Julia 1.9 - macOS-latest - default - pull_request (pull_request)Successful in 3m
CI / Julia 1.9 - ubuntu-latest - default - pull_request (pull_request)Successful in 3m
CI / Julia 1.9 - windows-latest - default - pull_request (pull_request)Successful in 4m
CI / Julia lts - macOS-latest - default - pull_request (pull_request)Successful in 3m
CI / Julia lts - ubuntu-latest - default - pull_request (pull_request)Successful in 3m
CI / Julia lts - windows-latest - default - pull_request (pull_request)Successful in 4m
CI / Julia nightly - ubuntu-latest - default - pull_request (pull_request)Successful in 6m
CI / Julia nightly - windows-latest - default - pull_request (pull_request)Successful in 6m
CI / Julia pre - macOS-latest - default - pull_request (pull_request)Successful in 4m
CI / Julia pre - ubuntu-latest - default - pull_request (pull_request)Successful in 5m
CI / Julia pre - windows-latest - default - pull_request (pull_request)Successful in 6m
codecov/patch — 95.89% of diff hit (target 93.46%)
codecov/project — 94.37% (+0.91%) compared to 168f7af
Clicking on the 3 dots to the right hand side and asking for more information gives
Run julia-actions/julia-buildpkg@v1 40m 3s
Run echo "JULIA_PKG_SERVER_REGISTRY_PREFERENCE=${JULIA_PKG_SERVER_REGISTRY_PREFERENCE:-eager}" >> ${GITHUB_ENV}
Run if "false" == "false" && !isdir(DEPOT_PATH[1])
Also, how can I kill this - or will it time-out in the end?
I went out and it was still running 3 hours later when I came back.
I’ve manged to restart by adding a single blank line in the code of the PR, but the behaviour so far is the same.
The task that is hanging (julia-actions/julia-buildpkg@v1) ran in 3s on Julia Pre - MacOS latest and in 6s on Julia 1 - MacOS latest. It seems to be the task the builds the test environment, before even trying to run the package’s test suite.
I don’t have a Mac myself, so can’t explore this locally.
I don’t know how to kill the process, so it’s still running.
I will look at this but on first inspection I don’t really understand it. Let’s see how I get on!
The test itself was cancelled when it timed out after 6hours last night. The message was as follows:
Run julia-actions/julia-buildpkg@v1
Run echo "JULIA_PKG_SERVER_REGISTRY_PREFERENCE=${JULIA_PKG_SERVER_REGISTRY_PREFERENCE:-eager}" >> ${GITHUB_ENV}
Run if "false" == "false" && !isdir(DEPOT_PATH[1])
[1145] signal 2: Interrupt: 2
in expression starting at none:0
__psynch_cvwait at /usr/lib/system/libsystem_kernel.dylib (unknown line)
unknown function (ip: 0x0) at (unknown file)
__psynch_mutexwait at /usr/lib/system/libsystem_kernel.dylib (unknown line)
unknown function (ip: 0x0) at (unknown file)
Allocations: 1 (Pool: 1; Big: 0); GC: 0
Error: The operation was canceled.
For comparison, here is the output of the successful runs on MacOS (e.g. Julia 1)...
Just to clarify, I’m making a PR on XLSX.jl, which is a package I don’t own. More specifically, I’m making a minor update on an existing PR of mine which previously passed all tests. I changed a very few lines - mostly (but not all) comments - and now I get this failure.
I will also try to back out these changes one at a time to see if any of them specifically is triggering the failure but, as I said before, I think this process hangs before it even gets to my code.
I tried to explain this in my previous message, but perhaps it wasn’t clear enough: this issue has nothing to do with your code, it happens in all julia nightly jobs. Something changed over the last ~week in either the Julia nightly build, or the macOS environment on GitHub Actions (but somehow only affecting only the nightly version of Julia?), someone will have to investigate it.