Mountpoint crashes on reading 1 byte file #218
Labels
bug
Something isn't working
Comments
|
I suspect this is related to this recent issue in CRT: awslabs/aws-c-s3#272 |
jamesbornholt
added a commit
to jamesbornholt/mountpoint-s3
that referenced
this issue
Apr 28, 2023
This was helpful for debugging awslabs#218 and awslabs#226. Signed-off-by: James Bornholt <bornholt@amazon.com>
jamesbornholt
added a commit
that referenced
this issue
Apr 28, 2023
Merged
jamesbornholt
added a commit
to jamesbornholt/mountpoint-s3
that referenced
this issue
May 1, 2023
This picks up the fix for awslabs#218. Signed-off-by: James Bornholt <bornholt@amazon.com>
jamesbornholt
added a commit
that referenced
this issue
May 2, 2023
This picks up the fix for #218. Signed-off-by: James Bornholt <bornholt@amazon.com>
This was referenced May 3, 2023
passaro
added a commit
that referenced
this issue
May 4, 2023
Improve PartQueue to report an error (rather than blocking indeterminately) when trying to read beyond the body returned by the GetObject request. The previous behavior, compounded with a bug in the CRT (now fixed: #228) where GetObject unexpectedly returned an empty body, manifested in #218. --------- Signed-off-by: Alessandro Passaro <alexpax@amazon.co.uk>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Mountpoint for Amazon S3 version
mountpoint-s3 0.2.0-70b2b68
AWS Region
eu-west-1
Describe the running environment
Running in EC2 on Amazon Linux 2 using instance profile credentials against an S3 bucket in the same account.
Also same happened on on my Mac using instance profile credentials against an S3 bucket.
What happened?
Mountpoint is constantly crashing when trying to read a 1 byte file. Tried it on several such objects. When they are wriiten to a file using getObject s3api CLI command, output the following result -
.%(where % is highlighted).I am attaching the trace logs. Prefetcher is able to find the correct range and also finishing the rpefetching task. But it stops working after that.
Relevant log output
2023-04-21T16:40:51.019337Z TRACE getattr{req=5 ino=5}: mountpoint_s3::fs: fs:getattr with ino 5 2023-04-21T16:40:51.019471Z TRACE open{req=2 ino=5}: mountpoint_s3::fs: fs:open with ino 5 flags 0 2023-04-21T16:40:51.019742Z TRACE read{req=6 ino=5 fh=1 offset=0 size=1}: mountpoint_s3::fs: fs:read with ino 5 fh 1 offset 0 size 1 2023-04-21T16:40:51.019770Z TRACE read{req=6 ino=5 fh=1 offset=0 size=1}: mountpoint_s3::prefetch: read offset=0 length=1 next_seq_offset=0 2023-04-21T16:40:51.019789Z TRACE read{req=6 ino=5 fh=1 offset=0 size=1}: mountpoint_s3::prefetch: to_read=1 2023-04-21T16:40:51.019812Z TRACE read{req=6 ino=5 fh=1 offset=0 size=1}: mountpoint_s3::prefetch: spawning request range=0..1 size=1 2023-04-21T16:40:51.020002Z DEBUG read{req=6 ino=5 fh=1 offset=0 size=1}:prefetch{range=0..1}:get_object{id=21}: mountpoint_s3_client::s3_crt_client::get_object: new request bucket="plutotestankit" key="test/test_sdk_create_object/14899866888963238197/hello" range=Some(0..1) if_match=Some(ETag { etag: "\"5058f1af8388633f609cadb75a75dc9d\"" }) size=Some(1) 2023-04-21T16:40:51.250506Z DEBUG read{req=6 ino=5 fh=1 offset=0 size=1}:prefetch{range=0..1}:get_object{id=21}: mountpoint_s3_client::s3_crt_client: request finished request_id="GQNFMC3TRQWP8DJS" duration_us=230374 2023-04-21T16:40:51.250735Z TRACE read{req=6 ino=5 fh=1 offset=0 size=1}:prefetch{range=0..1}: mountpoint_s3::prefetch: finished (It stops after this and crashes)The text was updated successfully, but these errors were encountered: