Create wait-for-as-server-to-start.bash

.github/workflows/wait-for-as-server-to-start.bash

@@ -0,0 +1,47 @@
+#!/bin/bash
+
+set -x
+# Makes sure that if the "docker exec" command fails, it is not ignored
+set -o pipefail
+
+container_name=$1
+is_security_enabled=$2
+
+if [[ $is_security_enabled == true ]]; then
+    # We need to pass credentials to asinfo if server requires it
+    # TODO: passing in credentials via command line flags since I can't figure out how to use --instance with global astools.conf
+    user_credentials="--user=admin --password=admin"
+fi
+
+while true; do
+    # An unset variable will have a default empty value
+    # Intermediate step is to print docker exec command's output in case it fails
+    # Sometimes, errors only appear in stdout and not stderr, like if asinfo throws an error because of no credentials
+    # (This is a bug in asinfo since all error messages should be sent to stderr)
+    # But piping and passing stdin to grep will hide the first command's stdout.
+    # grep doesn't have a way to print all lines passed as input.
+    # ack does have an option but it doesn't come installed by default
+    # shellcheck disable=SC2086 # The flags in user credentials should be separate anyways. Not one string
+    echo "Checking if we can reach the server via the service port..."
+    if docker exec "$container_name" asinfo $user_credentials -v status | tee >(cat) | grep -qE "^ok"; then
+        # Server is ready when asinfo returns ok
+        echo "Can reach server now."
+        # docker container inspect "$container_name"
+        break
+    fi
+
+    echo "Server didn't return ok via the service port. Polling again..."
+done
+
+# Although the server may be reachable via the service port, the cluster may not be fully initialized yet.
+# If we try to connect too soon (e.g right after "status" returns ok), the client may throw error code -1
+while true; do
+    echo "Waiting for server to stabilize (i.e return a cluster key)..."
+    # We assume that when an ERROR is returned, the cluster is not stable yet (i.e not fully initialized)
+    if docker exec "$container_name" asinfo $user_credentials -v cluster-stable 2>&1 | (! grep -qE "^ERROR"); then
+        echo "Server is in a stable state."
+        break
+    fi
+
+    echo "Server did not return a cluster key. Polling again..."
+done