Merge branch 'main' into improve-example-docs-script

.github/workflows/e2e.yml

@@ -164,6 +164,9 @@ jobs:
       - name: Run driver test with client authentication
         if: ${{ matrix.directory == 'bare-client-auth' }}
         run: ./../test_driver.sh bare client-auth
+      - name: Run reconnection test with SQLite database
+        if: ${{ matrix.directory == 'bare' }}
+        run: ./../test_reconnection.sh sqlite
       - name: Cache save Python location
         id: cache-save-python
         uses: actions/cache/save@v4

e2e/test_reconnection.sh

@@ -0,0 +1,88 @@
+#!/bin/bash
+set -e
+
+case "$1" in
+  rest)
+    rest_arg="--rest"
+    server_app_address="http://localhost:9091"
+    server_address="http://localhost:9093"
+    db_arg="--database :flwr-in-memory-state:"
+    ;;
+  sqlite)
+    rest_arg=""
+    server_address="127.0.0.1:9092"
+    server_app_address="127.0.0.1:9091"
+    db_arg="--database $(date +%s).db"
+    ;;
+  *)
+    rest_arg=""
+    server_address="127.0.0.1:9092"
+    server_app_address="127.0.0.1:9091"
+    db_arg="--database :flwr-in-memory-state:"
+    ;;
+esac
+
+dir_arg="--dir ./.."
+
+timeout 2m flower-superlink --insecure $db_arg $rest_arg &
+sl_pid=$!
+echo "Starting SuperLink"
+sleep 3
+
+timeout 2m flower-client-app client:app --insecure $rest_arg --server $server_address &
+cl1_pid=$!
+echo "Starting first client"
+sleep 3
+
+timeout 2m flower-client-app client:app --insecure $rest_arg --server $server_address &
+cl2_pid=$!
+echo "Starting second client"
+sleep 3
+
+# Kill superlink, this should send the clients into their retry loops
+kill $sl_pid
+echo "Killing Superlink"
+sleep 3
+
+# Restart superlink, the clients should now be able to reconnect to it
+timeout 2m flower-superlink --insecure $db_arg $rest_arg &
+sl_pid=$!
+echo "Restarting Superlink"
+sleep 20
+
+# Kill first client, this should send a DeleteNode message to the Superlink
+kill $cl1_pid
+echo "Killing first client"
+sleep 3
+
+# Starting new client, this is so we have enough clients to start the server-app
+timeout 2m flower-client-app client:app --insecure $rest_arg --server $server_address &
+cl1_pid=$!
+echo "Starting new client"
+sleep 5
+
+# We start the server-app to begining the training
+timeout 2m flower-server-app server:app --insecure $dir_arg $rest_arg --server $server_app_address &
+pid=$!
+echo "Starting server-app to start training"
+
+# Kill first client as soon as the training starts,
+# the server-app should just receive a failure in this case and continue the rounds
+# when enough clients are connected
+kill $cl1_pid
+echo "Killing first client"
+sleep 1
+
+# Restart first client so enough clients are connected to continue the FL rounds
+timeout 2m flower-client-app client:app --insecure $rest_arg --server $server_address &
+cl1_pid=$!
+echo "Starting new client"
+
+wait $pid
+res=$?
+
+if [[ "$res" = "0" ]];
+  then echo "Training worked correctly"; kill $cl1_pid; kill $cl2_pid; kill $sl_pid;
+  else echo "Training had an issue" && exit 1;
+fi
+