HDFS-17293. First packet data + checksum size will be set to 516 bytes when writing to a new block. #6368
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…s when writing to a new block.
Contributor
Author
|
@Hexiaoqiao @tomscut @ayushtkn @zhangshuyan0 Sir, could you please help me review this code when you have free time？Thanks ahead. |
|
💔 -1 overall
This message was automatically generated. |
Hexiaoqiao
reviewed
Dec 24, 2023
...op-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/DFSOutputStream.java
Show resolved
Hide resolved
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
|
🎊 +1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
Contributor
Author
|
@Hexiaoqiao @zhangshuyan0 @slfan1989 Sir, have updated this PR, added an unit test. could you please help me review this pr when you are free? Thanks~ |
zhangshuyan0
requested changes
Jan 19, 2024
Contributor
zhangshuyan0
left a comment
zhangshuyan0
left a comment
There was a problem hiding this comment.
Great catch! Leave some small suggestions.
| fos.write(buf); | ||
| fos.hflush(); | ||
| loop++; | ||
| Assert.assertNotEquals(crc32c.getBytesPerChecksum() + crc32c.getChecksumSize(), |
Contributor
There was a problem hiding this comment.
It is more appropriate to precisely specify the expected packetSize here.
Outside the while loop:
int chunkSize = crc32c.getBytesPerChecksum() + crc32c.getChecksumSize();
int packetContentSize = (dfsConf.getInt(DFS_CLIENT_WRITE_PACKET_SIZE_KEY, DFS_CLIENT_WRITE_PACKET_SIZE_DEFAULT) - PacketHeader.PKT_MAX_HEADER_LEN)/chunkSize*chunkSize;
And here:
Assert.assertEquals(((DFSOutputStream) fos.getWrappedStream()).packetSize, packetContentSize);
Contributor
Author
There was a problem hiding this comment.
Sir, thanks for this valuable suggestion. Will fix it soon.
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestDFSOutputStream.java
Outdated
Show resolved
Hide resolved
Contributor
|
This PR has corrected the size of the first packet in a new block, which is great. However, due to the original logical problem in Line540, when we pass blockSize - getStreamer().getBytesCurBlock() to computePacketChunkSize as the first parameter, computePacketChunkSize is likely to cause the data that could have been sent in one data packet to be split into two data packets and sent.
|
hfutatzhanghb
force-pushed
the
HDFS-17293
branch
4 times, most recently
from
0a43db0 to
af8a23b
Compare
Contributor
Author
|
@zhangshuyan0 Sir, Have modified according to your review opinions. Please take a look when you have free time. Thanks a lot~ |
|
💔 -1 overall
This message was automatically generated. |
hfutatzhanghb
force-pushed
the
HDFS-17293
branch
from
af8a23b to
da1ea49
Compare
hfutatzhanghb
force-pushed
the
HDFS-17293
branch
from
da1ea49 to
68292c7
Compare
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
Contributor
Author
Sir, very nice catch. I think below code may resolve the problem you found. Please take a look when you are free. if (!getStreamer().getAppendChunk()) {
int psize = 0;
if (blockSize == getStreamer().getBytesCurBlock()) {
psize = writePacketSize;
} else if (blockSize - getStreamer().getBytesCurBlock() + PacketHeader.PKT_MAX_HEADER_LEN
< writePacketSize ) {
psize = (int)(blockSize - getStreamer().getBytesCurBlock()) + PacketHeader.PKT_MAX_HEADER_LEN;
} else {
psize = (int) Math
.min(blockSize - getStreamer().getBytesCurBlock(), writePacketSize);
}
computePacketChunkSize(psize, bytesPerChecksum);
} |
hfutatzhanghb
force-pushed
the
HDFS-17293
branch
from
68292c7 to
96da2c8
Compare
Contributor
Thank you very much for investing your time in fixing these bugs. The above fixes did not take |
Contributor
Author
@zhangshuyan0 Sir, Agree with you, let's discuss this issue in the new PR. The failed tests were all passed in my local. |
Contributor
|
Committed to trunk. Thanks for your contribution @hfutatzhanghb . |
Contributor
Author
@zhangshuyan0 Sir, Thanks a lot for your reviewing. |
|
🎊 +1 overall
This message was automatically generated. |
jiajunmao
pushed a commit
to jiajunmao/hadoop-MLEC
that referenced
this pull request
Feb 6, 2024
…s when writing to a new block. (apache#6368). Contributed by farmmamba. Reviewed-by: He Xiaoqiao <hexiaoqiao@apache.org> Signed-off-by: Shuyan Zhang <zhangshuyan@apache.org>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description of PR
First packet size(data length + checksum length) will be set to 516 bytes when writing to a new block. It is an unnecessary and wrong behavior.
In method computePacketChunkSize, the parameters psize and csize would be （0, 512）when writting to a new block.
It should better use writePacketSize to improve the speed of sending data.