[CASSANDRA-14874] Read repair can race with truncations

• Type: Bug
• Status: Open
• Priority: Low
• Resolution: Unresolved
• Component/s: Legacy/Local Write-Read Paths
• Severity: Low
• Assignee: Unassigned
• Reporter: Sylvain Lebresne
• Created: 08/Nov/18 10:00
• Updated: 16/Apr/19 09:29

Description

While hint and commit log replay handle truncation alright, we don't have anything to prevent a read/read-repair to race with TRUNCATE. In other words, you can have a read reading some pre-truncation data, some truncation running and removing that data, and then some read-repair mutation from that previous read that resurrects some data that should have bene truncated.

Probably not that common in practice, but can lead to seemingly random data surviving truncate.