Re: [PATCH] ci: use upload-artifacts v1 for dockerized jobs

Carlo Marcelo Arenas Belón  [off-list ref] writes:

e9f79acb28 (ci: upgrade to using actions/{up,down}load-artifacts v2,
2021-06-23) changed all calls to that action from v1 to v2, but there
is still an open bug[1] that affects all nodejs actions and prevents
its use in 32-bit linux (as used by the Linux32 container)

move all dockerized jobs to use v1 that was build in C# and therefore

"was build -> was built", I would think.

doesn't have this problem which will otherwise manifest with confusing
messages like:

  /usr/bin/docker exec  0285adacc4536b7cd962079c46f85fa05a71e66d7905b5e4b9b1a0e8b305722a sh -c "cat /etc/*release | grep ^ID"
  OCI runtime exec failed: exec failed: container_linux.go:380: starting container process caused: no such file or directory: unknown

[1] https://github.com/actions/runner/issues/1011
---

Not signed off.

Intended or accidental?  At least it is not explained where this
change comes from in the proposed log message.

This one matches what the proposed log message says, switching from
@v2 to @v1.

       with:
         name: failed-tests-${{matrix.vector.jobname}}
         path: ${{env.FAILED_TEST_ARTIFACTS}}

On Thu, Aug 12, 2021 at 12:59 PM Junio C Hamano [off-list ref] wrote:

Carlo Marcelo Arenas Belón  [off-list ref] writes:

quoted

diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml
index 47876a4f02..9dd41a1742 100644
--- a/.github/workflows/main.yml
+++ b/.github/workflows/main.yml

@@ -258,7 +258,7 @@ jobs:
         - jobname: linux-musl
           image: alpine
         - jobname: Linux32
-          image: daald/ubuntu32:xenial
+          image: i386/ubuntu:xenial

Intended or accidental?

yes!.

It was intended but more as an RFC, because I couldn't find an
explanation on what was the advantage of using this 5 year old
unofficial image when there was an equivalent official one that had
been updated just last month.

It was accidental, because I didn't realize I hadn't updated the log
message and it isn't needed to fix the regression which was the main
focus anyway.

 At least it is not explained where this
change comes from in the proposed log message.

The advantages of using an official image will be :

* more likely to be cached and therefore less likely to fail the job
* easier to move to a newer version (specially considering 16.04 is EOL)

Would it be better to drop or document it for v2?

Carlo