When a file is deleted in S3 using S3Guard a tombstone marker will be added for that file in the MetadataStore. If another process creates the file without using S3Guard (as an out of band operation - OOB) the file still not be visible for the client using S3Guard because of the deletion tombstone.
The whole of S3Guard is potentially brittle to
- OOB deletions: we skip it in
HADOOP-15999, so no worse, but because the S3AInputStream retries on FNFE, so as to "debounce" cached 404s, it's potentially going to retry forever.
- OOB creation of a file which has a deletion tombstone marker.
The things this issue covers:
- Write a test to simulate that deletion problem, to see what happens. We ought to have the S3AInputStream retry briefly on that initial GET failing, but only on that initial one. (after setting "fs.s3a.retry.limit" to something low & the interval down to 10ms or so to fail fast)
The StoreStatistics of the filesystem's IGNORED_ERRORS stat will be increased on the ignored error, so on sequence 1 will have increased, whereas on sequence 2 it will not have. If either of these tests don't quite fail as expected, we can disable the tests and continue, at least now with some tests to simulate a condition we don't have a fix for.
- For both, we just need to have some model of how long it takes for debouncing to stabilize. Then in this new check, if an FNFE is raised and the check is happening > (modtime+ debounce-delay) then it's a real FNFE.