FrankYang0529 commented on code in PR #18735:
URL: https://github.com/apache/kafka/pull/18735#discussion_r1934272060
##########
clients/src/test/java/org/apache/kafka/clients/admin/KafkaAdminClientTest.java:
##########
@@ -1783,12 +1782,14 @@ public void
testDescribeTopicsWithDescribeTopicPartitionsApiErrorHandling() thro
}
}
- @Flaky("KAFKA-18441")
@Test
public void testAdminClientApisAuthenticationFailure() {
Cluster cluster = mockBootstrapCluster();
try (final AdminClientUnitTestEnv env = new
AdminClientUnitTestEnv(Time.SYSTEM, cluster,
- newStrMap(AdminClientConfig.REQUEST_TIMEOUT_MS_CONFIG,
"1000"))) {
+ newStrMap(AdminClientConfig.REQUEST_TIMEOUT_MS_CONFIG, "1000",
+ // The default "retry.backoff.ms" is 100. if following
assertion can't finish in 100L,
+ // the test will fail. Set it to 5000 to make sure the
test can finish in time.
+ AdminClientConfig.RETRY_BACKOFF_MS_CONFIG, "5000"))) {
env.kafkaClient().setNodeApiVersions(NodeApiVersions.create());
env.kafkaClient().createPendingAuthenticationError(cluster.nodes().get(0),
TimeUnit.DAYS.toMillis(1));
Review Comment:
Hi @AndrewJSchofield and @lianetm, thanks for the review and information.
After checking the code again, my PR description is unclear.
With default "retry.backoff.ms", if all assertion can't finish in `100L`,
the `KafkaAdminClient` make a metadata call again.
https://github.com/apache/kafka/blob/048dfeffd0b8dcda01ec8c4c6f110b9d60994b84/clients/src/main/java/org/apache/kafka/clients/admin/KafkaAdminClient.java#L1541-L1550
In the `fetchMetadata` request, it uses `MetadataUpdateNodeIdProvider`.
https://github.com/apache/kafka/blob/048dfeffd0b8dcda01ec8c4c6f110b9d60994b84/clients/src/main/java/org/apache/kafka/clients/admin/KafkaAdminClient.java#L1686-L1690
In `MetadataUpdateNodeIdProvider`, it triggers
`AdminMetadataManager#rebootstrap`.
https://github.com/apache/kafka/blob/048dfeffd0b8dcda01ec8c4c6f110b9d60994b84/clients/src/main/java/org/apache/kafka/clients/admin/KafkaAdminClient.java#L722-L730
Finally, the `rebootstrap` function makes `fatalException` be null again in
`AdminMetadataManager#update`.
https://github.com/apache/kafka/blob/632aedcf4ff7e718bf1295fe5132ebaf1d3f92a0/clients/src/main/java/org/apache/kafka/clients/admin/internals/AdminMetadataManager.java#L303-L313
This case relies on `AdminMetadataManager#isReady` to throw error. If
`fatalException` is null, the assertion can't be fulfilled.
https://github.com/apache/kafka/blob/632aedcf4ff7e718bf1295fe5132ebaf1d3f92a0/clients/src/main/java/org/apache/kafka/clients/admin/internals/AdminMetadataManager.java#L186-L190
--
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: jira-unsubscr...@kafka.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
