cockroachdb / cockroachdb/cockroach
roachprod/failure-injection: purposely undersize certain disks
Open
A-testeng-foundations
C-enhancement
O-postmortem
P-3
T-testeng
- Dominant language
- Go
- Stars
- 32.5k
- Forks
- 4.1k
- PR merge metrics
- PR metrics pending
Description
The bulk ingestion corruption described in #144650 could be induced by undersized disks. AFAICT, we do not test the following:
- begin bulk ingestion where a node contains an undersized disk
- observe the job pause
- resize the disk
- resume the job
Perhaps testing this flow in the new failure injection framework could provide additional coverage.
Jira issue: CRDB-50522
Contributor guide
Assessment
This issue has not been assessed yet.