[GitHub] [nifi-minifi-cpp] szaszm commented on a diff in pull request #1580: MINIFICPP-2126 Use sccache and Ninja in Windows CI

["szaszm (via GitHub)" <gi...@apache.org>]

szaszm commented on code in PR #1580:
URL: https://github.com/apache/nifi-minifi-cpp/pull/1580#discussion_r1212964250


##########
.github/workflows/ci.yml:
##########
@@ -73,8 +74,15 @@ jobs:
         run: git config --system core.longpaths true
       - id: checkout
         uses: actions/checkout@v3
-      - name: Set up MSBuild
-        uses: microsoft/setup-msbuild@v1.1
+      - name: Run sccache-cache
+        uses: mozilla-actions/sccache-action@v0.0.3
+      - name: sccache cache
+        uses: actions/cache@v3
+        with:
+          path: ~/AppData/Local/Mozilla/sccache/cache
+          key: ${{ runner.os }}-sccache

Review Comment:
   Done, but there is a problem with this approach: It requires multiple times as much storage as a single central cache: one for main, and one for each active PR. We're already hitting the 10 GB limit, caches are getting evicted, and we're on the way to cache trashing, but not quite there yet.
   
   We will need a custom action in the future, that drops older caches in a smart way, before GitHub drops the oldest, without considering the branch. https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflows#force-deleting-cache-entries
   
   Added the ref, but separating caches by SHA seems excessive: that's a separate cache key for each commit, raising the number of caches in storage from the number of active PRs * number of platforms to the number of commits * number of platforms.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@nifi.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org