[CASSANDRA-17754] When bootstrap fails, CassandraRoleManager may attempt to do read queries that fail with "Cannot read from a bootstrapping node", and increments unavailables counters

• Type: Bug
• Status: Resolved
• Priority: Normal
• Resolution: Fixed
• Fix Version/s: 5.0-alpha1, 5.0
• Component/s: Consistency/Bootstrap and Decommission, Feature/Authorization
• Bug Category: Correctness - Recoverable Corruption / Loss
• Severity: Normal
• Complexity: Low Hanging Fruit
• Discovered By: User Report
• Platform: All
• Since Version: 2.0.0

Source Control Link

https://github.com/apache/cassandra/commit/db3a29428df5f4e17791ce8a4e075532abb36667

Test and Documentation Plan

added tests

• Assignee: David Capwell
• Reporter: David Capwell

Authors

David Capwell

Reviewers

Sam Tunnicliffe

• Created: 14/Jul/22 22:58
• Updated: 12/Sep/23 13:02
• Resolved: 25/Jul/22 21:35

Description

If bootstrap fails we stay alive but avoid joining the ring. CassandraRoleManager attempts to schedule its setup every 10 seconds, but this will fail due to Bootstrap being pending; the side effect to this is that readMetrics.unavailables gets incremented.

We should defer attempting to setup only after we know its safe to perform such as read

Issue Links

• links to: CI: trunk
• links to: PR: trunk