Fixing wsrelay connection loop (#14692)

* Fixing wsrelay connection loop

* The loop was being interrupted when reaching the return statements, causing a race condition that would make nodes remain disconnected from their websockets
* Added log messages for the previous return state to improve the logging from this state.

* Added logging for malformed payload

* Update awx/main/wsrelay.py

Co-authored-by: Rick Elrod <rick@elrod.me>

* Moved logmsg outside condition

---------

Co-authored-by: Lucas Benedito <lbenedit@redhat.com>
Co-authored-by: Rick Elrod <rick@elrod.me>

.github/workflows/promote.yml

@@ -13,7 +13,7 @@ permissions:
 
 jobs:
   promote:
-    if: endsWith(github.repository, '/awx') 
+    if: endsWith(github.repository, '/awx')
     runs-on: ubuntu-latest
     steps:
       - name: Checkout awx
@@ -46,7 +46,7 @@ jobs:
           COLLECTION_TEMPLATE_VERSION: true
         run: |
           make build_collection
-          if [ "$(curl --head -sw '%{http_code}' https://galaxy.ansible.com/download/${{ env.collection_namespace }}-awx-${{ github.event.release.tag_name }}.tar.gz | tail -1)" == "302" ] ; then \
+          if [ "$(curl -L --head -sw '%{http_code}' https://galaxy.ansible.com/download/${{ env.collection_namespace }}-awx-${{ github.event.release.tag_name }}.tar.gz | tail -1)" == "302" ] ; then \
               echo "Galaxy release already done"; \
           else \
               ansible-galaxy collection publish \

awx/main/wsrelay.py

@@ -227,7 +227,8 @@ class WebSocketRelayManager(object):
                 continue
             try:
                 if not notif.payload or notif.channel != "web_ws_heartbeat":
-                    return
+                    logger.warning(f"Unexpected channel or missing payload. {notif.channel}, {notif.payload}")
+                    continue
 
                 try:
                     payload = json.loads(notif.payload)
@@ -235,13 +236,15 @@ class WebSocketRelayManager(object):
                     logmsg = "Failed to decode message from pg_notify channel `web_ws_heartbeat`"
                     if logger.isEnabledFor(logging.DEBUG):
                         logmsg = "{} {}".format(logmsg, payload)
-                        logger.warning(logmsg)
-                    return
+                    logger.warning(logmsg)
+                    continue
 
                 # Skip if the message comes from the same host we are running on
                 # In this case, we'll be sharing a redis, no need to relay.
                 if payload.get("hostname") == self.local_hostname:
-                    return
+                    hostname = payload.get("hostname")
+                    logger.debug("Received a heartbeat request for {hostname}. Skipping as we use redis for local host.")
+                    continue
 
                 action = payload.get("action")
 
@@ -250,7 +253,7 @@ class WebSocketRelayManager(object):
                     ip = payload.get("ip") or hostname  # try back to hostname if ip isn't supplied
                     if ip is None:
                         logger.warning(f"Received invalid {action} ws_heartbeat, missing hostname and ip: {payload}")
-                        return
+                        continue
                     logger.debug(f"Web host {hostname} ({ip}) {action} heartbeat received.")
 
                 if action == "online":