[
https://issues.apache.org/jira/browse/CASSANDRA-14858?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
C. Scott Andreas updated CASSANDRA-14858:
-----------------------------------------
Component/s: Local Write-Read Paths
> Validate range tombstones and clustering order on the write path
> ----------------------------------------------------------------
>
> Key: CASSANDRA-14858
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14858
> Project: Cassandra
> Issue Type: Bug
> Components: Local Write-Read Paths
> Reporter: Blake Eggleston
> Assignee: Blake Eggleston
> Priority: Major
>
> Some fuzz testing I've been doing uncovered some situations where invalid
> tombstones could be written and/or unfiltereds could be written to sstables
> out of clustering order. Both of these caused pretty severe database
> breakage. Although cql statement logic should (in theory) prevent users from
> triggering these bugs in normal use, it's possible there's an edge case
> somewhere that could write data like this. We should add some validation in
> the lower levels of the storage layer to protect it from bugs / malicious
> input from the higher levels.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
