Toshihiko Uchida created HDFS-15759:
---------------------------------------
Summary: EC: Verify EC reconstruction correctness on DataNode
Key: HDFS-15759
URL: https://issues.apache.org/jira/browse/HDFS-15759
Project: Hadoop HDFS
Issue Type: New Feature
Components: datanode, ec, erasure-coding
Affects Versions: 3.4.0
Reporter: Toshihiko Uchida
EC reconstruction on DataNode has caused data corruption: HDFS-14768,
HDFS-15186 and HDFS-15240. Those issues occur under specific conditions and the
corruption is neither detected nor auto-healed by HDFS. It is obviously hard
for users to monitor data integrity by themselves, and even if they find
corrupted data, it is difficult or sometimes impossible to recover them.
To prevent further data corruption issues, this feature proposes a simple and
effective way to verify EC reconstruction correctness on DataNode at each
reconstruction process.
It verifies correctness of outputs decoded from inputs as follows:
1. Decoding an input with the outputs;
2. Compare the decoded input with the original input.
For instance, in RS-6-3, assume that outputs [d1, p1] are decoded from inputs
[d0, d2, d3, d4, d5, p0]. Then the verification is done by decoding d0 from
[d1, d2, d3, d4, d5, p1], and comparing the original and decoded data of d0.
When an EC reconstruction task goes wrong, the comparison will fail with high
probability.
Then the task will also fail and be retried by NameNode.
The next reconstruction will succeed if the condition triggered the failure is
gone.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-dev-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-dev-h...@hadoop.apache.org
