Skip to content

HADOOP-19343: Add hadoop-gcp configuration to core-default.xml and ServiceLoader file. #7916

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
cnauroth wants to merge 1 commit into from
Closed

HADOOP-19343: Add hadoop-gcp configuration to core-default.xml and ServiceLoader file. #7916

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
200 changes: 193 additions & 7 deletions hadoop-common-project/hadoop-common/src/main/resources/core-default.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1285,7 +1285,7 @@

<property>
<name>fs.viewfs.overload.scheme.target.gs.impl</name>
<value>com.google.cloud.hadoop.fs.gcs.GoogleHadoopFS</value>
<value>org.apache.hadoop.fs.gs.GoogleHadoopFileSystem</value>
<description>The GoogleHadoopFS/Google Cloud Storage file system for view
file system overload scheme when child file system and ViewFSOverloadScheme's
schemes are gs.
Expand Down Expand Up @@ -2373,12 +2373,6 @@ The switch to turn S3A auditing on or off.
otherwise fall back to hadoop.tmp.dir </description>
</property>

<property>
<name>fs.AbstractFileSystem.gs.impl</name>
<value>com.google.cloud.hadoop.fs.gcs.GoogleHadoopFS</value>
<description>The AbstractFileSystem for gs: uris.</description>
</property>

<property>
<name>fs.azure.enable.readahead</name>
<value>true</value>
Expand Down Expand Up @@ -4509,4 +4503,196 @@ The switch to turn S3A auditing on or off.
If the value is less than or equal to 0, the cache is disabled entirely.
</description>
</property>

<property>
<name>fs.gs.impl</name>
<value>org.apache.hadoop.fs.gs.GoogleHadoopFileSystem</value>
<description>The FileSystem for gs: uris.</description>
</property>

<property>
<name>fs.AbstractFileSystem.gs.impl</name>
<value>org.apache.hadoop.fs.gs.Gs</value>
<description>The AbstractFileSystem for gs: uris.</description>
</property>

<property>
<name>fs.gs.project.id</name>
<description>
Google Cloud Project ID with access to Google Cloud Storage buckets.
Required only for list buckets and create bucket operations.
</description>
</property>

<property>
<name>fs.gs.working.dir</name>
<value>/</value>
<description>
The directory relative gs: uris resolve in inside the default bucket.
</description>
</property>

<property>
<name>fs.gs.rewrite.max.chunk.size</name>
<value>512m</value>
<description>
Maximum size of object chunk that will be rewritten in a single rewrite
request when fs.gs.copy.with.rewrite.enable is set to true.
</description>
</property>

<property>
<name>fs.gs.bucket.delete.enable</name>
<value>false</value>
<description>
If true, recursive delete on a path that refers to a Cloud Storage bucket
itself or delete on that path when it is empty will result in deletion of
the bucket itself. If false, any operation that normally would have
deleted the bucket will be ignored. Setting to false preserves the typical
behavior of rm -rf / which translates to deleting everything inside of
root, but without clobbering the filesystem authority corresponding to that
root path in the process.
</description>
</property>

<property>
<name>fs.gs.block.size</name>
<value>64m</value>
<description>
The reported block size of the file system. This does not change any
behavior of the connector or the underlying Google Cloud Storage objects.
However, it will affect the number of splits Hadoop MapReduce uses for a
given input.
</description>
</property>

<property>
<name>fs.gs.create.items.conflict.check.enable</name>
<value>true</value>
<description>
Enables a check that ensures that conflicting directories do not exist when
creating files and conflicting files do not exist when creating directories.
</description>
</property>

<property>
<name>fs.gs.marker.file.pattern</name>
<description>
If set, files that match specified pattern are copied last during folder
rename operation.
</description>
</property>

<property>
<name>fs.gs.auth.type</name>
<value>COMPUTE_ENGINE</value>
<description>
What type of authentication mechanism to use for Google Cloud Storage
access. Valid values: APPLICATION_DEFAULT, COMPUTE_ENGINE,
SERVICE_ACCOUNT_JSON_KEYFILE, UNAUTHENTICATED, USER_CREDENTIALS.
</description>
</property>

<property>
<name>fs.gs.auth.service.account.json.keyfile</name>
<description>
The path to the JSON keyfile for the service account when fs.gs.auth.type
property is set to SERVICE_ACCOUNT_JSON_KEYFILE. The file must exist at
the same path on all nodes
</description>
</property>

<property>
<name>fs.gs.auth.client.id</name>
<description>
The OAuth2 client ID.
</description>
</property>

<property>
<name>fs.gs.auth.client.secret</name>
<description>
The OAuth2 client secret.
</description>
</property>

<property>
<name>fs.gs.auth.refresh.token</name>
<description>
The refresh token.
</description>
</property>

<property>
<name>fs.gs.inputstream.support.gzip.encoding.enable</name>
<value>false</value>
<description>
If set to false then reading files with GZIP content encoding (HTTP header
Content-Encoding: gzip) will result in failure (IOException is thrown).

This feature is disabled by default because processing of
GZIP encoded files is inefficient and error-prone in Hadoop and Spark.
</description>
</property>

<property>
<name>fs.gs.outputstream.buffer.size</name>
<value>8m</value>
<description>
Write buffer size used by the file system API to send the data to be
uploaded to Cloud Storage upload thread via pipes. The various pipe types
are documented below.
</description>
</property>

<property>
<name>fs.gs.outputstream.sync.min.interval</name>
<value>0</value>
<description>
Output stream configuration that controls the minimum interval between
consecutive syncs. This allows to avoid getting rate-limited by Google Cloud
Storage. Default is 0 - no wait between syncs. Note that hflush() will
be no-op if called more frequently than minimum sync interval and hsync()
will block until an end of a min sync interval.
</description>
</property>

<property>
<name>fs.gs.inputstream.fadvise</name>
<value>AUTO</value>
<description>
Tunes reading objects behavior to optimize HTTP GET requests for various use
cases. Valid values: SEQUENTIAL, RANDOM, AUTO, AUTO_RANDOM.
</description>
</property>

<property>
<name>fs.gs.fadvise.request.track.count</name>
<value>3</value>
<description>
Self adaptive fadvise mode uses distance between the served requests to
decide the access pattern. This property controls how many such requests
need to be tracked. It is used when AUTO_RANDOM is selected.
</description>
</property>

<property>
<name>fs.gs.inputstream.inplace.seek.limit</name>
<value>8m</value>
<description>
If forward seeks are within this many bytes of the current position, seeks
are performed by reading and discarding bytes in-place rather than opening a
new underlying stream.
</description>
</property>

<property>
<name>fs.gs.inputstream.min.range.request.size</name>
<value>2m</value>
<description>
Minimum size in bytes of the read range for Cloud Storage request when
opening a new stream to read an object.
</description>
</property>

</configuration>
16 changes: 16 additions & 0 deletions hadoop-tools/hadoop-gcp/src/main/resources/META-INF/services/org.apache.hadoop.fs.FileSystem
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements. See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

org.apache.hadoop.fs.gs.GoogleHadoopFileSystem