abstractdog commented on code in PR #6591:
URL: https://github.com/apache/hive/pull/6591#discussion_r3577596830


##########
packaging/src/docker/docker-compose.yml:
##########
@@ -141,6 +160,11 @@ services:
       - zookeeper_data:/data
       - zookeeper_datalog:/datalog
       - zookeeper_logs:/logs
+    healthcheck:
+      test: ["CMD-SHELL", "bash -c 'echo ruok > /dev/tcp/localhost/2181'"]
+      interval: 5s
+      timeout: 3s
+      retries: 30
 
   tezam:

Review Comment:
   is there a chance to have healthcheck for tezams?
   `tez.am.rpc.port` could be set to a static value, and multiple tez AM's 
won't collide given they are on different hosts like llap daemons
   
   
https://github.com/apache/tez/blob/4bccf3a2339c913ade0bc5447bdfc505c5536128/tez-dag/src/main/java/org/apache/tez/dag/api/client/DAGClientServer.java#L69



##########
packaging/src/docker/docker-compose.yml:
##########
@@ -141,6 +160,11 @@ services:
       - zookeeper_data:/data
       - zookeeper_datalog:/datalog
       - zookeeper_logs:/logs
+    healthcheck:
+      test: ["CMD-SHELL", "bash -c 'echo ruok > /dev/tcp/localhost/2181'"]

Review Comment:
   for clarity's sake, zookeeper is supposed to return "imok", however this 
happens after start:
   ```
   docker compose exec zookeeper sh -c 'printf ruok | nc -w 2 127.0.0.1 2181'
   ...
   ruok is not executed because it is not in the whitelist.
   ```
   maybe     
   ```
   environment:
         ZOO_4LW_COMMANDS_WHITELIST: ruok
   ```
   I think if the zookeeper is unable to start, then a connection refused is 
already enough to ensure the unhealthy state, however, it would be good to 
somehow have "imok" in the response check



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to