This is an automated email from the ASF dual-hosted git repository.

zeroshade pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow-go.git


The following commit(s) were added to refs/heads/main by this push:
     new 10b10e66 ci: run benchmark metadata steps on Python 3.12 (#1427)
10b10e66 is described below

commit 10b10e66e4e695baac5ee15b367ba221e6e0223e
Author: Matt Topol <[email protected]>
AuthorDate: Tue Oct 27 15:07:10 2026 +0000

    ci: run benchmark metadata steps on Python 3.12 (#1427)
    
    ### Rationale for this change
    
    Every `benchmark` shard is failing at "Set up Python for benchmark
    metadata":
    
    ```
    ##[error]The version '3.9' with architecture 'x64' was not found for Ubuntu 
26.04.
    ```
    
    `benchmark.yml` requests Python 3.9 on `ubuntu-latest`, which is moving
    to Ubuntu 26.04; `actions/setup-python` has no 3.9 build for that image,
    and 3.9 is end of life. This fails on `main` (e.g. run 40373053653) and
    on any PR that runs the benchmark workflow, such as #1423.
    
    ### What changes are included in this PR?
    
    Use Python 3.12 for both setup-python steps in `benchmark.yml`, matching
    `lint.yml` and `test.yml`. Python is only used to install and run the
    conbench `benchadapt` collector (`ci/scripts/bench_adapt.py`).
    
    ### Are these changes tested?
    
    Locally on Python 3.12.13: the pinned `benchadapt` commit installs, and
    `bench_adapt.py --capture-machine-info` writes the machine-info JSON for
    a sample result file. `actionlint` accepts the workflow. This PR's own
    benchmark run exercises the change on CI.
    
    ### Are there any user-facing changes?
    
    No.
---
 .github/workflows/benchmark.yml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/benchmark.yml b/.github/workflows/benchmark.yml
index bbf14057..bda54bb9 100644
--- a/.github/workflows/benchmark.yml
+++ b/.github/workflows/benchmark.yml
@@ -66,7 +66,7 @@ jobs:
       - name: Set up Python for benchmark metadata
         uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # 
v7.0.0
         with:
-          python-version: '3.9'
+          python-version: '3.12'
       - name: Run benchmark shard ${{ matrix.id }}
         run: |
           bash ci/scripts/bench.sh "$(pwd)" --run \
@@ -118,7 +118,7 @@ jobs:
         if: github.event_name == 'push' && github.repository == 
'apache/arrow-go' && github.ref_name == 'main'
         uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # 
v7.0.0
         with:
-          python-version: '3.9'
+          python-version: '3.12'
       - name: Upload results
         if: github.event_name == 'push' && github.repository == 
'apache/arrow-go' && github.ref_name == 'main'
         env:

Reply via email to