<!--
Thanks for sending a pull request!
Here are some tips for you:
1. If this is your first time, please read our contributor guidelines: https://kyuubi.readthedocs.io/en/latest/community/contributions.html
2. If the PR is related to an issue in https://github.com/apache/incubator-kyuubi/issues, add '[KYUUBI #XXXX]' in your PR title, e.g., '[KYUUBI #XXXX] Your PR title ...'.
3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][KYUUBI #XXXX] Your PR title ...'.
-->
### _Why are the changes needed?_
<!--
Please clarify why the changes are needed. For instance,
1. If you add a feature, you can talk about the use case of it.
2. If you fix a bug, you can clarify why it is a bug.
-->
As the Spark driver may launch from another node, when we start the Kyuubi server with an embedded zookeeper and bind the frontend address to `0.0.0.0`, the backend engine can't register correctly.
This pr changes the default behavior of `kyuubi.zookeeper.embedded.client.port.address`, creates the default zookeeper address with local hostname instead of falling back to `kyuubi.frontend.bind.host`.
### _How was this patch tested?_
- [ ] Add some test cases that check the changes thoroughly including negative and positive cases if possible
- [ ] Add screenshots for manual tests if appropriate
- [ ] [Run test](https://kyuubi.readthedocs.io/en/latest/develop_tools/testing.html#running-tests) locally before make a pull request
Closes#1004 from cfmcgrady/fallback-address.
Closes#1004
99f0a143 [Fu Chen] fix test
5e66c5f0 [Fu Chen] create default zk address instead of fall back to frontend bind address
Authored-by: Fu Chen <cfmcgrady@gmail.com>
Signed-off-by: Kent Yao <yao@apache.org>
<!--
Thanks for sending a pull request!
Here are some tips for you:
1. If this is your first time, please read our contributor guidelines: https://kyuubi.readthedocs.io/en/latest/community/contributions.html
2. If the PR is related to an issue in https://github.com/apache/incubator-kyuubi/issues, add '[KYUUBI #XXXX]' in your PR title, e.g., '[KYUUBI #XXXX] Your PR title ...'.
3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][KYUUBI #XXXX] Your PR title ...'.
-->
### _Why are the changes needed?_
<!--
Please clarify why the changes are needed. For instance,
1. If you add a feature, you can talk about the use case of it.
2. If you fix a bug, you can clarify why it is a bug.
-->
### 1. Describe the feature
Kyuubi use `TestingServer` as the EmbeddedZookeeper service in codebase, and require many args to init the service. These `electionPort`, `quorumPort`, `serverId` parameters only work in cluster mode, these ports are not bound to the host.
From hadoop and bookkeeper project, they use `ZooKeeperServer` as a built-in zk service, especially the bookeeper project.
- https://github.com/apache/hadoop/blob/trunk/hadoop-common-project/hadoop-registry/src/main/java/org/apache/hadoop/registry/server/services/MicroZookeeperService.java
- https://github.com/apache/bookkeeper/blob/master/bookkeeper-server/src/main/java/org/apache/bookkeeper/shims/zk/ZooKeeperServerShimImpl.java
### 2. Describe the solution
Improve
- Remove `electionPort`, `quorumPort`, `serverId` which are nerver be used, hadoop and bookkeer do the same
- Replace `TestingServer` with `ZooKeeperServer`
- Remove `curator-test` dependency
### 3. `deleteDataDirectoryOnClose`
The name style for `deleteDataDirectoryOnClose` option, refer to hadoop `core-default.xml`
https://hadoop.apache.org/docs/stable/hadoop-project-dist/hadoop-common/core-default.xml
```
hadoop.security.dns.log-slow-lookups.enabled
hadoop.security.dns.log-slow-lookups.threshold.ms
dfs.client.block.write.replace-datanode-on-failure.policy
```
### _How was this patch tested?_
- [ ] Add some test cases that check the changes thoroughly including negative and positive cases if possible
- [ ] Add screenshots for manual tests if appropriate
- [x] [Run test](https://kyuubi.readthedocs.io/en/latest/tools/testing.html#running-tests) locally before make a pull request
Closes#907 from timothy65535/ky-894.
Closes#894
7d7fde9f [timothy65535] remove useless config
ce06a0fd [timothy65535] limit travis build only master branch
ff825c3f [timothy65535] revert deleteDataDirectoryOnClose config
a336544a [timothy65535] remove javassist
3da0bd55 [timothy65535] remove javassist
0abb44df [timothy65535] [KYUUBI #894] Enhance kyuubi-zookeeper module
Authored-by: timothy65535 <timothy65535@163.com>
Signed-off-by: ulysses-you <ulyssesyou18@gmail.com>
<!--
Thanks for sending a pull request!
Here are some tips for you:
1. If this is your first time, please read our contributor guidelines: https://kyuubi.readthedocs.io/en/latest/community/contributions.html
2. If the PR is related to an issue in https://github.com/apache/incubator-kyuubi/issues, add '[KYUUBI #XXXX]' in your PR title, e.g., '[KYUUBI #XXXX] Your PR title ...'.
3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][KYUUBI #XXXX] Your PR title ...'.
-->
### _Why are the changes needed?_
<!--
Please clarify why the changes are needed. For instance,
1. If you add a feature, you can talk about the use case of it.
2. If you fix a bug, you can clarify why it is a bug.
-->
We shall avoid calling cancelJobGroup if the sc is already stopped.
We shall stop the engine to stop serving before we stop the sc during the shutdown hook
### _How was this patch tested?_
- [ ] Add some test cases that check the changes thoroughly including negative and positive cases if possible
- [ ] Add screenshots for manual tests if appropriate
- [ ] [Run test](https://kyuubi.readthedocs.io/en/latest/tools/testing.html#running-tests) locally before make a pull request
Closes#868 from yaooqinn/shutdownhook.
Closes#868
14810875 [Kent Yao] comment
972e2741 [Kent Yao] Avoid calling a stopped SparkContext
Authored-by: Kent Yao <yao@apache.org>
Signed-off-by: Kent Yao <yao@apache.org>
<!--
Thanks for sending a pull request!
Here are some tips for you:
1. If this is your first time, please read our contributor guidelines: https://kyuubi.readthedocs.io/en/latest/community/contributions.html
2. If the PR is related to an issue in https://github.com/NetEase/kyuubi/issues, add '[KYUUBI #XXXX]' in your PR title, e.g., '[KYUUBI #XXXX] Your PR title ...'.
3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][KYUUBI #XXXX] Your PR title ...'.
-->
### _Why are the changes needed?_
<!--
Please clarify why the changes are needed. For instance,
1. If you add a feature, you can talk about the use case of it.
2. If you fix a bug, you can clarify why it is a bug.
-->
### _How was this patch tested?_
- [ ] Add some test cases that check the changes thoroughly including negative and positive cases if possible
- [ ] Add screenshots for manual tests if appropriate
- [ ] [Run test](https://kyuubi.readthedocs.io/en/latest/tools/testing.html#running-tests) locally before make a pull request
Closes#669 from pan3793/cleanup.
Closes#669
2cdc4a3 [Cheng Pan] Cleanup code and fix typos
Authored-by: Cheng Pan <379377944@qq.com>
Signed-off-by: Kent Yao <yao@apache.org>
### _Why are the changes needed?_
When you stop the KyuubiServiceDiscovery, you need to close node in zk and delete localNodePath at first. But when EmbeddedZookeeper is closed before KyuubiServiceDiscovery, the operation that stop kyuubiServer will timeout.
Now use 'org.apache.hadoop.util.ShutdownHookManager' to close each service by different priority.
### _How was this patch tested?_
- [ ] Add some test cases that check the changes thoroughly including negative and positive cases if possible
- [ ] Add screenshots for manual tests if appropriate
- [ ] [Run test](https://kyuubi.readthedocs.io/en/latest/tools/testing.html#running-tests) locally before make a pull request
Closes#632 from zhang1002/dev.
Closes#632
ee21e68 [张宇翔] Change the relevant format
f68fc7a [张宇翔] Merge branch 'master' into dev
128f244 [张宇翔] solve the timeout problem when stop the kyuubiServer
79b3d5a [张宇翔] ## solve the folder naming problem For starting kyuubiServer, the script need to get 'SPARK_HOME' from the file: RELEASE. The rule is: spark-${SPARK_VERSION}-bin-hadoop${HADOOP_VERSION}. But when we package project from source, the script named dist will create the directory about 'SPARK_HOME' by the rule: spark-${SPARK_VERSION}-bin-hadoop${SPARK_HADOOP_VERSION}. So it's the differences between two rules. This change let ${HADOOP_VERSION} as same as ${SPARK_HADOOP_VERSION}.
Lead-authored-by: 张宇翔 <zhang1002@126.com>
Co-authored-by: 张宇翔 <zhangyuxiang03@corp.netease.com>
Signed-off-by: Kent Yao <yao@apache.org>
