You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Abhijit Sarkar (JIRA)" <ji...@apache.org> on 2019/06/12 03:59:00 UTC
[jira] [Comment Edited] (CASSANDRA-15154) Add console log to
indicate the node is ready to accept requests
[ https://issues.apache.org/jira/browse/CASSANDRA-15154?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16861734#comment-16861734 ]
Abhijit Sarkar edited comment on CASSANDRA-15154 at 6/12/19 3:58 AM:
---------------------------------------------------------------------
[~mshuler] I'm sure something in the logs can be dug out that'd tell an experienced Cassandra user that the node is ready; however, the point of this ticket is to make that very obvious. Even in the snippet you posted, "Starting listening for CQL clients" isn't necessarily the same as "I'm ready". It can easily be presumed that the node might have more to do after it had started listening for CQL clients.
BTW, this ticket is filed as an improvement, not a bug. "Not a bug" doesn't apply; "Won't do because we don't feel like it" might, but isn't one of the goals of good software is to be more intuitive to use?
was (Author: socialguy):
[~mshuler] I'm sure something in the logs can be dug out that'd tell an experienced Cassandra user that the node is ready; however, the point of this ticket is to make that very obvious. Even in the snippet you posted, "Starting listening for CQL clients" isn't necessarily the same as "I'm ready". It can easily be presumed that the node might have more to do after it had started listening for CQL clients.
> Add console log to indicate the node is ready to accept requests
> ----------------------------------------------------------------
>
> Key: CASSANDRA-15154
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15154
> Project: Cassandra
> Issue Type: Improvement
> Components: Cluster/Membership, Local/Startup and Shutdown, Observability/Logging
> Reporter: Abhijit Sarkar
> Priority: Normal
>
> Depending on whether a cluster is initialized the first time, or a node is restarted, the last message on the console varies. In either case, there's no indication that the cluster/node is ready to accept requests.
> For example, when I create a new Cassandra Docker container locally:
> {code}
> $ docker run --name cas -p 9042:9042 -p 9091:9091 -e CASSANDRA_DC=dev cassandra
> ...
> INFO [OptionalTasks:1] 2019-06-11 23:31:35,527 CassandraRoleManager.java:356 - Created default superuser role 'cassandra'
> {code}
> After shutting it down (CTRL + C), and restarting:
> {code}
> $ docker start cas
> ...
> INFO [main] 2019-06-11 23:32:57,980 CassandraDaemon.java:556 - Not starting RPC server as requested. Use JMX (StorageService->startRPCServer()) or nodetool (enablethrift) to start it
> {code}
> In either of the above cases, how is a regular user, whose full time job is not working with Cassandra, expected to know whether the server is ready? We have a new member in the team who previously was an iOS developer. He left the server running overnight, assuming the node hadn't finished initialization; the next morning, the last message was still "Created default superuser role 'cassandra'".
> Please add a simple log statement with basic information like node IPs in the cluster indicating the node is ready. For example, this is what Spring Boot does:
> {code}
> 2019-06-11 16:37:28.295 INFO [my-app,,,] 17392 --- [ main] o.s.b.w.embedded.tomcat.TomcatWebServer : Tomcat started on port(s): 11900 (http) with context path ''
> 2019-06-11 16:37:28.299 INFO [my-app,,,] 17392 --- [ main] mypackage.MyApp : Started MyApp in 5.279 seconds (JVM running for 5.916)
> {code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org