big-data/hadoop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

HDFS-12927. Update erasure coding doc to address unsupported APIs. (Contributed by Lei (Eddy) Xu)

Browse Source
This commit is contained in:
Lei Xu 2017-12-15 10:20:29 -08:00
parent 1c15b1751c
commit 949be14b08
1 changed files with 5 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
hadoop-hdfs-project/hadoop-hdfs/src/site/markdown/HDFSErasureCoding.md
View File

@ -220,11 +220,14 @@ Below are the details about each command.
Limitations Limitations
----------- -----------
Certain HDFS file write operations, i.e., `hflush`, `hsync` and `append`, Certain HDFS operations, i.e., `hflush`, `hsync`, `concat`, `setReplication`, `truncate` and `append`,
are not supported on erasure coded files due to substantial technical are not supported on erasure coded files due to substantial technical
challenges. challenges.
* `append()` on an erasure coded file will throw `IOException`. * `append()` and `truncate()` on an erasure coded file will throw `IOException`.
* `concat()` will throw `IOException` if files are mixed with different erasure
coding policies or with replicated files.
* `setReplication()` is no-op on erasure coded files.
* `hflush()` and `hsync()` on `DFSStripedOutputStream` are no-op. Thus calling * `hflush()` and `hsync()` on `DFSStripedOutputStream` are no-op. Thus calling
`hflush()` or `hsync()` on an erasure coded file can not guarantee data `hflush()` or `hsync()` on an erasure coded file can not guarantee data
being persistent. being persistent.