[CI] Drop x86_64 macOS from JNI builds
Nessuno ha ancora preso questa issue.
Valutazione
- Difficoltà
- 3/5
- Tempo stimato
- 1-2 giorni
- Idoneità per principianti
- 72/100
- Tipo di issue
- Refactoring
- Chiarezza
- Specificata chiaramente
- Stato di attività
- Attiva
- Ambito
- build-system, ci-cd
Direzione di ricerca
Read rc.yml and test.yml first, focusing on the jni-macos, binaries, and macos matrix entries. Then inspect ci/scripts/jni_macos_build.sh and the Homebrew workaround sections. Done means x86_64 macOS is removed from the specified CI matrices and artifact checks, Intel-only workarounds are removed, and local x86_64 JNI builds remain supported.
Scritto dal modello di indicizzazione a partire dal testo della issue.
Descrizione
Describe the enhancement requested
Homebrew treats x86_64 macOS as a tier 3 configuration and rarely builds bottles for it. This has broken the JNI macos-15-intel x86_64 job in the RC workflow repeatedly, because brew bundle doesn't build from source and aborts with no bottle available! before anything is compiled:
- #1269 worked around it for
aws-sdk-cppand gRPC. - #1270 worked around it for Node.js.
Each of these was a formula-by-formula patch, and any new dependency added to apache/arrow's cpp/Brewfile can break the job again the same way. We can't keep an x86_64 macOS JNI build green on a configuration Homebrew doesn't support.
As noted in #1269, dropping the macos-15-intel entry needs its own discussion because it is user-facing, so this issue is that discussion.
Proposal
Drop x86_64 macOS from CI:
rc.yml: remove themacos-15-intel/x86_64entry from thejni-macosmatrix, and stop extracting and verifyingjni-macos-x86_64.tar.gzin thebinariesjob.test.yml: remove theAMD64/macos-15-intelentry from themacostest matrix.- Remove the Homebrew workarounds that only existed for the x86_64 runners (the
/usr/localpython@XXX --overwriteconflict loop, and theuname -mguard around thepkg-configuninstall).
User-facing impact
Release JARs would no longer bundle x86_64 macOS .dylib files for arrow_cdata_jni, arrow_dataset_jni, arrow_orc_jni and gandiva_jni. Intel Mac users of those modules would need to build the JNI libraries themselves. ci/scripts/jni_macos_build.sh still handles x86_64, so a local Intel build keeps working.
aarch64 macOS, x86_64 and aarch64 Linux, and x86_64 Windows are unaffected.
Component(s)
Java, CI
- Lingua principale
- Java
- Stelle
- 95
- Fork
- 154
- Merge medio
- 2g 10h
- PR unite (30g)
- 11
Guida per i contributori
Apri la guida per i contributori
Come iniziare
- Leggi tutta la issue e poi la guida ai contributi del progetto.
- Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
- Fai un fork del repository e lavora su un branch.
- Apri una pull request che faccia riferimento al numero della issue.
Altre issue di apache/arrow-java
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 74/100
apache/arrow-java#1261 ·
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 78/100
apache/arrow-java#1236 ·
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 78/100
apache/arrow-java#1230 ·
-
Type: bug
Difficoltà 2/5 1-3 ore Idoneità per principianti 85/100
apache/arrow-java#1205 ·
-
Type: bug
Difficoltà 2/5 1-3 ore Idoneità per principianti 68/100
apache/arrow-java#1196 · 1 commento ·
Tutte le issue di apache/arrow-java
Issue simili
-
bug untriaged
Difficoltà 2/5 1-3 ore Idoneità per principianti 84/100
opensearch-project/ml-commons#5094 ·
-
bug
Difficoltà 2/5 1-3 ore Idoneità per principianti 85/100
-
emitter:client:csharp feature
Difficoltà 2/5 1-3 ore Idoneità per principianti 72/100
-
affects/8.10 affects/8.9 component/clients kind/bug likelihood/mid severity/mid
Difficoltà 2/5 1-3 ore Idoneità per principianti 78/100
-
Two open-case totals on one screen: the Programs tile says 15,858 and the nav badge says 15,868 Apertabug frontend maui-pilot
Difficoltà 2/5 1-3 ore Idoneità per principianti 72/100