Flink on yarn 加载失败plugins失效问题解决

flink版本:1.13.6

1. 问题

flink 任务运行在yarn集群,plugins加载失效,导致通过扩展资源获取任务参数失效

2. 问题定位

  1. yarn容器的jar包及插件信息,jar包是正常上传

  2. 源码定位

    加载plugins入口,TaskManagerRunner.class

    PluginUtils.createPluginManagerFromRootFolder

    源码加载扩展资源参数入口TaskManagerRunner.class

    ExternalResourceUtils.createStaticExternalResourceInfoProviderFromConfig

    日志信息

    定位PluginConfig源码

3. 方案

重写覆盖集群的PluginConfig.java,优先从configurtaion获取

bash 复制代码
/*
 * 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.
 */

package org.apache.flink.core.plugin;

import com.alibaba.fastjson.JSON;
import org.apache.commons.lang3.StringUtils;
import org.apache.flink.configuration.*;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import java.io.File;
import java.nio.file.Path;
import java.util.Arrays;
import java.util.Objects;
import java.util.Optional;

/** Stores the configuration for plugins mechanism. */
@SuppressWarnings("OptionalUsedAsFieldOrParameterType")
public class PluginConfig {
    private static final Logger LOG = LoggerFactory.getLogger(PluginConfig.class);

    private static final String PUPU_FLINK_PLUGINS_DIR = "flink_plugins_dir";

    private final Optional<Path> pluginsPath;

    private final String[] alwaysParentFirstPatterns;

    private PluginConfig(Optional<Path> pluginsPath, String[] alwaysParentFirstPatterns) {
        pluginsPath.ifPresent(path -> LOG.info("pluginsPath: {}", path));
        LOG.info("alwaysParentFirstPatterns: {}", Arrays.stream(alwaysParentFirstPatterns).toArray());
        this.pluginsPath = pluginsPath;
        this.alwaysParentFirstPatterns = alwaysParentFirstPatterns;
    }

    public Optional<Path> getPluginsPath() {
        return pluginsPath;
    }

    public String[] getAlwaysParentFirstPatterns() {
        return alwaysParentFirstPatterns;
    }

    public static PluginConfig fromConfiguration(Configuration configuration) {
        return new PluginConfig(
                getPluginsDir(configuration).map(File::toPath),
                CoreOptions.getPluginParentFirstLoaderPatterns(configuration));
    }

    public static Optional<File> getPluginsDir(Configuration configuration) {
        String pluginsDir = configuration.get(ConfigOptions.key(PUPU_FLINK_PLUGINS_DIR).stringType().defaultValue(null));
        if (StringUtils.isBlank(pluginsDir)) {
            pluginsDir =
                    System.getenv()
                            .getOrDefault(
                                    ConfigConstants.ENV_FLINK_PLUGINS_DIR,
                                    ConfigConstants.DEFAULT_FLINK_PLUGINS_DIRS);
        }
        File pluginsDirFile = new File(pluginsDir);
        if (!pluginsDirFile.isDirectory()) {
            LOG.warn("The plugins directory [{}] does not exist.", pluginsDirFile);
            return Optional.empty();
        }
        return Optional.of(pluginsDirFile);
    }

}
相关推荐
zgl_2005377911 小时前
ZGLanguage 解析SQL数据血缘 之 标识提取SQL语句中的目标表
java·大数据·数据库·数据仓库·hadoop·sql·源代码管理
尋有緣11 小时前
力扣1355-活动参与者
大数据·数据库·leetcode·oracle·数据库开发
roman_日积跬步-终至千里13 小时前
【大数据架构-数据中台(1)】解码数据中台:从概念到认知
大数据·架构·dubbo
追风的木木13 小时前
TDengine在NetCore中数据查询的使用
大数据·tdengine
数据猿13 小时前
【金猿CIO展】上海纽约大学信息技术部高级主任常潘:大数据铸基,AI赋能,从数字化校园向智慧有机体的十年跃迁
大数据·人工智能
李@十一₂⁰13 小时前
git多分支管理
大数据·git·elasticsearch
CappuccinoRose14 小时前
流计算概述
python·flink·流计算·数据流·pyflink
yumgpkpm14 小时前
AI评判:信创替代对Cloudera CDH CDP Hadoop大数据平台有何影响?
大数据·hive·oracle·flink·kafka·hbase·cloudera
小四的快乐生活14 小时前
大数据SQL诊断(采集、分析、优化方案)
大数据·数据库·sql
DeepFlow 零侵扰全栈可观测14 小时前
3分钟定位OA系统GC瓶颈:DeepFlow全栈可观测平台实战解析
大数据·运维·人工智能·云原生·性能优化