mirror of
https://github.com/s3fs-fuse/s3fs-fuse.git
synced 2026-04-25 13:26:00 +03:00
[GH-ISSUE #264] Intermittent test failures testing against Amazon S3 servers #137
Labels
No labels
bug
bug
dataloss
duplicate
enhancement
feature request
help wanted
invalid
need info
performance
pull-request
question
question
testing
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
starred/s3fs-fuse#137
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @RobbKistler on GitHub (Sep 11, 2015).
Original GitHub issue: https://github.com/s3fs-fuse/s3fs-fuse/issues/264
I opened PR #263 that includes the ability to optionally run the s3fs-fuse integration tests against an Amazon S3 server instead of s3proxy.
I see intermittent failures for the integration-test-main.sh/test_append_file test. It happens more frequently on a multicore test machine. I assume this is an S3 object coherency issue.
When I modify the test to append a sequence number, the dropped lines can occur anywhere:
Is this the expected behavior (and a bad test)? Unless s3fs-fuse does something heroic like poll HEAD the object to wait for the Etag to update -- I don't see that happening currently.
@gaul commented on GitHub (Sep 11, 2015):
Integration tests against AWS fail due to eventual consistency of read-after-overwrite. If you use a strongly consistent S3 provider like Ceph/DreamObjects or the default S3Proxy harness you will not experience this. Some background:
https://github.com/andrewgaul/are-we-consistent-yet
We should modify the test to avoid overwrites.
@RobbKistler commented on GitHub (Sep 11, 2015):
Great writeup, thanks!
Will try to address with PR #264
@ggtakec commented on GitHub (Sep 13, 2015):
I merged #263, if you have more problem, please reopen this issue.