[KAFKA-4873] Investigate issues uncovered by CORDS

• Type: Bug
• Status: Open
• Priority: Critical
• Resolution: Unresolved
• Labels: reliability
• Assignee: Unassigned
• Reporter: Ismael Juma
• Created: 09/Mar/17 10:09
• Updated: 28/Mar/17 14:48

Description

"CORDS is a fault-injection system consisting of errfs, a FUSE file system, and errbench, a set of workloads and a behaviour inference script for each system under test."

Kafka seemed to have issues with single bit errors, see the following for more details:

https://blog.acolyer.org/2017/03/08/redundancy-does-not-imply-fault-tolerance-analysis-of-distributed-storage-reactions-to-single-errors-and-corruptions/

All the code seems to be available:

http://research.cs.wisc.edu/adsl/Software/cords/

Issue Links

• is related to: KAFKA-4009 Data corruption or EIO leads to data loss