[
https://issues.apache.org/jira/browse/IMPALA-8521?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16835635#comment-16835635
]
Sahil Takiar commented on IMPALA-8521:
--------------------------------------
[~tarmstrong] do you have an easy way to re-produce this? Do you have the
stderr / stdout of the impalads? libhdfs prints directly to stderr / stdout so
it may have printed out some error messages there. I'm not sure what is causing
the cache read to fail, hoping the stderr / stdout shows something.
{{but there are still unreleased ByteBuffers allocated by read()}} is printed
by the hdfs-client when we do a zero-copy read (HDFS-4953), but don't call
{{releaseBuffer}} on the returned {{ByteBuffer}}.
Looks like there are two issues (1) cache reads are failing for some unknown
reason, and (2) it seems like {{hadoopReadZero}} in libhdfs can succeed in
calling {{HasEnhancedByteBufferAccess#read}} but fail afterwards, and forget to
call {{releaseBuffer}} on the buffer returned by {{#read}}.
> Lots of "unreleased ByteBuffers allocated by read()" errors from HDFS client
> ----------------------------------------------------------------------------
>
> Key: IMPALA-8521
> URL: https://issues.apache.org/jira/browse/IMPALA-8521
> Project: IMPALA
> Issue Type: Bug
> Components: Backend
> Reporter: Tim Armstrong
> Assignee: Sahil Takiar
> Priority: Critical
>
> I'm looking at some job logs and seeing a bunch of errors like this. I don't
> know if it's benign or if it's something more serious.
> {noformat}
> I0507 07:34:53.934693 20195 scan-range.cc:607]
> dd4d6eb8d2ad9587:6b44fe1b00000002] Cache read failed for scan range:
> file=hdfs://localhost:20500/test-warehouse/f861f1a3/nation.tbl disk_id=0
> offset=1024 exclusive_hdfs_fh=0xec09220 num_remote_bytes=0 cancel_status=
> buffer_queue=0 num_buffers_in_readers=0 unused_iomgr_buffers=0
> unused_iomgr_buffer_bytes=0 blocked_on_buffer=0. Switching to disk read path.
> W0507 07:34:53.934787 20195 DFSInputStream.java:668]
> dd4d6eb8d2ad9587:6b44fe1b00000002] closing file
> /test-warehouse/f861f1a3/nation.tbl, but there are still unreleased
> ByteBuffers allocated by read(). Please release
> java.nio.DirectByteBufferR[pos=1024 lim=2048 cap=2199].
> {noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
