[CASSANDRA-9032] Reduce logging level for MigrationTask abort due to down node from ERROR to INFO

• Type: Improvement
• Status: Resolved
• Priority: Low
• Resolution: Fixed
• Fix Version/s: 2.0.15, 2.1.5
• Assignee: Tom Hobbs
• Reporter: Tom Hobbs

Authors

Tom Hobbs

Reviewers

Aleksey Yeschenko

• Created: 24/Mar/15 22:19
• Updated: 16/Apr/19 09:31
• Resolved: 25/Mar/15 15:43

Description

A lot of the dtests are failing during Jenkins runs due to the following error message in the logs:

"ERROR [MigrationStage:1] 2015-03-24 20:02:03,464 MigrationTask.java:62 - Can't send migration request: node /127.0.0.3 is down.\n"]

This log message happens when a schema pull is scheduled, but the target endpoint is down when the scheduled task actually runs. The failing dtests generally stop a node as part of the test, which results in this.

I believe the log message should be moved from ERROR to INFO (or perhaps even DEBUG). This isn't an unexpected type of problem (nodes go down all the time), and it's not actionable by the user. This would also have the nice side effect of fixing the dtests.

Issue Links

• is part of: CASSANDRA-8996 dtests should pass on trunk
• relates to: CASSANDRA-12653 In-flight shadow round requests