[
https://issues.apache.org/jira/browse/HDDS-419?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16609619#comment-16609619
]
Xiaoyu Yao edited comment on HDDS-419 at 9/10/18 6:08 PM:
----------------------------------------------------------
Thanks [~msingh] for reporting the issue and posting the fix. The fix looks
good with the new test added but it hides a latent bug in
ChunkGroupInputStream#read(). I propose to fix it there like below without
changing OzoneInputStream, let me know your thoughts.
{code:java}
*---
a/hadoop-ozone/client/src/main/java/org/apache/hadoop/ozone/client/io/ChunkGroupInputStream.java*
*+++
b/hadoop-ozone/client/src/main/java/org/apache/hadoop/ozone/client/io/ChunkGroupInputStream.java*
@@ -122,12 +122,8 @@ public synchronized int read(byte[] b, int off, int len)
throws IOException {
return totalReadLen > 0 ? totalReadLen : EOF;
}
totalReadLen += actualLen;
- // this means there is no more data to read beyond this point, return
- if (actualLen != readLen) {
- return totalReadLen;
- }
- off += readLen;
- len -= readLen;
+ off += actualLen;
+ len -= actualLen;
if (current.getRemaining() <= 0) {
currentStreamIndex += 1;
}
{code}
was (Author: xyao):
Thanks [~msingh] for reporting the issue and posting the fix. The fix looks
good with the new test added but it hides a latent bug in
ChunkGroupInputStream#read(). I propose to fix it there like below, let me know
your thoughts.
{code}
*---
a/hadoop-ozone/client/src/main/java/org/apache/hadoop/ozone/client/io/ChunkGroupInputStream.java*
*+++
b/hadoop-ozone/client/src/main/java/org/apache/hadoop/ozone/client/io/ChunkGroupInputStream.java*
@@ -122,12 +122,8 @@ public synchronized int read(byte[] b, int off, int len)
throws IOException {
return totalReadLen > 0 ? totalReadLen : EOF;
}
totalReadLen += actualLen;
- // this means there is no more data to read beyond this point, return
- if (actualLen != readLen) {
- return totalReadLen;
- }
- off += readLen;
- len -= readLen;
+ off += actualLen;
+ len -= actualLen;
if (current.getRemaining() <= 0) {
currentStreamIndex += 1;
}
{code}
> ChunkInputStream bulk read api does not read from all the chunks
> ----------------------------------------------------------------
>
> Key: HDDS-419
> URL: https://issues.apache.org/jira/browse/HDDS-419
> Project: Hadoop Distributed Data Store
> Issue Type: Bug
> Components: Ozone Client
> Affects Versions: 0.2.1
> Reporter: Mukul Kumar Singh
> Assignee: Mukul Kumar Singh
> Priority: Blocker
> Fix For: 0.2.1
>
> Attachments: HDDS-419.001.patch
>
>
> After enabling of bulk reads with HDDS-408, testDataValidate started failing
> because the bulk read api does not read all the chunks from the block.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
