retry on cluster startup failure

Author: rashtao

.github/workflows/maven.yml

@@ -49,7 +49,7 @@ jobs:
         with:
           java-version: 1.8
       - name: Start Database
-        run: ./docker/start_db_${{ matrix.topology }}.sh ${{ matrix.docker-img }}
+        run: ./docker/start_db_${{ matrix.topology }}_retry_fail.sh ${{ matrix.docker-img }}
         env:
           ARANGO_LICENSE_KEY: ${{ secrets.ARANGO_LICENSE_KEY }}
       - name: Test

docker/clean_cluster.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+
+for c in agent1 \
+  agent2 \
+  agent3 \
+  dbserver1 \
+  dbserver2 \
+  coordinator1 \
+  coordinator2; do
+  docker rm -f $c
+done

docker/clean_single.sh

@@ -0,0 +1,3 @@
+#!/bin/bash
+
+docker rm -f arangodb

docker/start_db_cluster_retry_fail.sh

@@ -0,0 +1,16 @@
+#!/bin/bash
+
+# USAGE:
+#   export ARANGO_LICENSE_KEY=<arangodb-enterprise-license>
+#   ./docker/start_db_cluster_retry_fail.sh <dockerImage>
+
+# EXAMPLE:
+#   ./docker/start_db_cluster_retry_fail.sh docker.io/arangodb/arangodb:3.7.1
+
+./docker/start_db_cluster.sh "$1"
+while [ $? -ne 0 ]; do
+  echo "=== === ==="
+  echo "cluster startup failed, retrying ..."
+  ./docker/clean_cluster.sh
+  ./docker/start_db_cluster.sh "$1"
+done

docker/start_db_single_retry_fail.sh

@@ -0,0 +1,16 @@
+#!/bin/bash
+
+# USAGE:
+#   export ARANGO_LICENSE_KEY=<arangodb-enterprise-license>
+#   ./docker/start_db_single_retry_fail.sh <dockerImage>
+
+# EXAMPLE:
+#   ./docker/start_db_single_retry_fail.sh docker.io/arangodb/arangodb:3.7.1
+
+./docker/start_db_single.sh "$1"
+while [ $? -ne 0 ]; do
+  echo "=== === ==="
+  echo "single startup failed, retrying ..."
+  ./docker/clean_single.sh
+  ./docker/start_db_single.sh "$1"
+done