Copy tune-runner-vm and clean-disk from Pulsar CI and clean disk

.github/actions/clean-disk/action.yml

@@ -0,0 +1,57 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+#
+
+name: clean disk
+description: makes some more space available on the disk by removing files
+inputs:
+  mode:
+    description: "Use 'full' to clean as much as possible"
+    required: false
+runs:
+  using: composite
+  steps:
+    - run: |
+        if [[ "$OSTYPE" == "linux-gnu"* ]]; then
+          directories=(/usr/local/lib/android /opt/ghc)
+          if [[ "${{ inputs.mode }}" == "full" ]]; then
+            # remove these directories only when mode is 'full'
+            directories+=(/usr/share/dotnet /opt/hostedtoolcache/CodeQL)
+          fi
+          emptydir=/tmp/empty$$/
+          mkdir $emptydir
+          echo "::group::Available diskspace"
+          time df -BM / /mnt
+          echo "::endgroup::"
+          for directory in "${directories[@]}"; do
+            echo "::group::Removing $directory"
+            # fast way to delete a lot of files on linux
+            time sudo eatmydata rsync -a --delete $emptydir ${directory}/
+            time sudo eatmydata rm -rf ${directory}
+            time df -BM / /mnt
+            echo "::endgroup::"
+          done
+          echo "::group::Cleaning apt state"
+          time sudo bash -c "apt-get clean; apt-get autoclean; apt-get -y --purge autoremove"
+          time df -BM / /mnt
+          echo "::endgroup::"
+        fi
+        echo "::group::Available diskspace"
+        time df -BM / /mnt
+        echo "::endgroup::"
+      shell: bash

.github/actions/tune-runner-vm/action.yml

@@ -24,8 +24,79 @@ runs:
   steps:
     - run: |
         if [[ "$OSTYPE" == "linux-gnu"* ]]; then
+            echo "::group::Configure and tune OS"
             # Ensure that reverse lookups for current hostname are handled properly
             # Add the current IP address, long hostname and short hostname record to /etc/hosts file
             echo -e "$(ip addr show eth0 | grep "inet\b" | awk '{print $2}' | cut -d/ -f1)\t$(hostname -f) $(hostname -s)" | sudo tee -a /etc/hosts
+
+            # The default vm.swappiness setting is 60 which has a tendency to start swapping when memory
+            # consumption is high.
+            # Set vm.swappiness=1 to avoid swapping and allow high RAM usage
+            echo 1 | sudo tee /proc/sys/vm/swappiness
+            (
+              shopt -s nullglob
+              # Set swappiness to 1 for all cgroups and sub-groups
+              for swappiness_file in /sys/fs/cgroup/memory/*/memory.swappiness /sys/fs/cgroup/memory/*/*/memory.swappiness; do
+                echo 1 | sudo tee $swappiness_file > /dev/null
+              done
+            ) || true
+
+            # use "madvise" Linux Transparent HugePages (THP) setting
+            # https://www.kernel.org/doc/html/latest/admin-guide/mm/transhuge.html
+            # "madvise" is generally a better option than the default "always" setting
+            # Based on Azul instructions from https://docs.azul.com/prime/Enable-Huge-Pages#transparent-huge-pages-thp
+            echo madvise | sudo tee /sys/kernel/mm/transparent_hugepage/enabled
+            echo advise | sudo tee /sys/kernel/mm/transparent_hugepage/shmem_enabled
+            echo defer+madvise | sudo tee /sys/kernel/mm/transparent_hugepage/defrag
+            echo 1 | sudo tee /sys/kernel/mm/transparent_hugepage/khugepaged/defrag
+    
+            # tune filesystem mount options, https://www.kernel.org/doc/Documentation/filesystems/ext4.txt
+            # commit=999999, effectively disables automatic syncing to disk (default is every 5 seconds)
+            # nobarrier/barrier=0, loosen data consistency on system crash (no negative impact to empheral CI nodes)
+            sudo mount -o remount,nodiscard,commit=999999,barrier=0 /
+            sudo mount -o remount,nodiscard,commit=999999,barrier=0 /mnt
+            # disable discard/trim at device level since remount with nodiscard doesn't seem to be effective
+            # https://www.spinics.net/lists/linux-ide/msg52562.html
+            for i in /sys/block/sd*/queue/discard_max_bytes; do
+              echo 0 | sudo tee $i
+            done
+            # disable any background jobs that run SSD discard/trim
+            sudo systemctl disable fstrim.timer || true
+            sudo systemctl stop fstrim.timer || true
+            sudo systemctl disable fstrim.service || true
+            sudo systemctl stop fstrim.service || true
+
+            # stop php-fpm
+            sudo systemctl stop php8.0-fpm.service || true
+            sudo systemctl stop php7.4-fpm.service || true
+            # stop mono-xsp4
+            sudo systemctl disable mono-xsp4.service || true
+            sudo systemctl stop mono-xsp4.service || true
+            sudo killall mono || true
+
+            # stop Azure Linux agent to save RAM
+            sudo systemctl stop walinuxagent.service || true
+          
+            # enable docker experimental mode which is
+            # required for using "docker build --squash" / "-Ddocker.squash=true"
+            daemon_json="$(sudo cat /etc/docker/daemon.json  | jq '.experimental = true')"
+            echo "$daemon_json" | sudo tee /etc/docker/daemon.json
+            # restart docker daemon
+            sudo systemctl restart docker
+            echo '::endgroup::'
+
+            # show memory
+            echo "::group::Available Memory"
+            free -m
+            echo '::endgroup::'
+            # show disk
+            echo "::group::Available diskspace"
+            df -BM
+            echo "::endgroup::"
+            # show cggroup
+            echo "::group::Cgroup settings for current cgroup $CURRENT_CGGROUP"
+            CURRENT_CGGROUP=$(cat /proc/self/cgroup | grep '0::' | awk -F: '{ print $3 }')
+            sudo cgget -a $CURRENT_CGGROUP || true
+            echo '::endgroup::'
         fi
-      shell: bash
+      shell: bash

.github/workflows/bk-ci.yml

@@ -218,6 +218,11 @@ jobs:
       - name: Tune Runner VM
         uses: ./.github/actions/tune-runner-vm
 
+      - name: Clean Disk
+        uses: ./.github/actions/clean-disk
+        with:
+          mode: full
+
       - name: Cache local Maven repository
         id: cache
         uses: actions/cache@v4
@@ -282,6 +287,11 @@ jobs:
       - name: Tune Runner VM
         uses: ./.github/actions/tune-runner-vm
 
+      - name: Clean Disk
+        uses: ./.github/actions/clean-disk
+        with:
+          mode: full
+
       - name: Cache local Maven repository
         id: cache
         uses: actions/cache@v4