[FLINK-21700][security] Add an option to disable credential retrieval on a secure cluster

Author: zuston

docs/layouts/shortcodes/generated/security_auth_kerberos_section.html

@@ -8,6 +8,12 @@
         </tr>
     </thead>
     <tbody>
+        <tr>
+            <td><h5>security.kerberos.fetch.delegation-token</h5></td>
+            <td style="word-wrap: break-word;">true</td>
+            <td>Boolean</td>
+            <td>Indicates whether to fetch the delegation tokens for external services the Flink job needs to contact. Only HDFS and HBase are supported. It is used in Yarn deployments. If true, Flink will fetch HDFS and HBase delegation tokens and inject them into Yarn AM containers.  If false, Flink will assume that the delegation tokens are managed outside of Flink. As a consequence, it will not fetch delegation tokens for HDFS and HBase. You may need to disable this option, if you rely on submission mechanisms, e.g. Apache Oozie, to handle delegation tokens.</td>
+        </tr>
         <tr>
             <td><h5>security.kerberos.login.contexts</h5></td>
             <td style="word-wrap: break-word;">(none)</td>

docs/layouts/shortcodes/generated/security_configuration.html

@@ -14,6 +14,12 @@
             <td>List&lt;String&gt;</td>
             <td>List of factories that should be used to instantiate a security context. If multiple are configured, Flink will use the first compatible factory. You should have a NoOpSecurityContextFactory in this list as a fallback.</td>
         </tr>
+        <tr>
+            <td><h5>security.kerberos.fetch.delegation-token</h5></td>
+            <td style="word-wrap: break-word;">true</td>
+            <td>Boolean</td>
+            <td>Indicates whether to fetch the delegation tokens for external services the Flink job needs to contact. Only HDFS and HBase are supported. It is used in Yarn deployments. If true, Flink will fetch HDFS and HBase delegation tokens and inject them into Yarn AM containers.  If false, Flink will assume that the delegation tokens are managed outside of Flink. As a consequence, it will not fetch delegation tokens for HDFS and HBase. You may need to disable this option, if you rely on submission mechanisms, e.g. Apache Oozie, to handle delegation tokens.</td>
+        </tr>
         <tr>
             <td><h5>security.kerberos.krb5-conf.path</h5></td>
             <td style="word-wrap: break-word;">(none)</td>

flink-core/src/main/java/org/apache/flink/configuration/SecurityOptions.java

@@ -108,6 +108,20 @@ public class SecurityOptions {
                                     + " (for example, `Client,KafkaClient` to use the credentials for ZooKeeper authentication and for"
                                     + " Kafka authentication)");
 
+    @Documentation.Section(Documentation.Sections.SECURITY_AUTH_KERBEROS)
+    public static final ConfigOption<Boolean> KERBEROS_FETCH_DELEGATION_TOKEN =
+            key("security.kerberos.fetch.delegation-token")
+                    .booleanType()
+                    .defaultValue(true)
+                    .withDescription(
+                            "Indicates whether to fetch the delegation tokens for external services the Flink job needs to contact. "
+                                    + "Only HDFS and HBase are supported. It is used in Yarn deployments. "
+                                    + "If true, Flink will fetch HDFS and HBase delegation tokens and inject them into Yarn AM containers.  "
+                                    + "If false, Flink will assume that the delegation tokens are managed outside of Flink. "
+                                    + "As a consequence, it will not fetch delegation tokens for HDFS and HBase. "
+                                    + "You may need to disable this option, if you rely on submission mechanisms, e.g. Apache Oozie, "
+                                    + "to handle delegation tokens.");
+
     // ------------------------------------------------------------------------
     //  ZooKeeper Security Options
     // ------------------------------------------------------------------------

flink-yarn/src/main/java/org/apache/flink/yarn/Utils.java

@@ -197,13 +197,23 @@ private static LocalResource registerLocalResource(
     }
 
     public static void setTokensFor(
-            ContainerLaunchContext amContainer, List<Path> paths, Configuration conf)
+            ContainerLaunchContext amContainer,
+            List<Path> paths,
+            Configuration conf,
+            boolean obtainingDelegationTokens)
             throws IOException {
         Credentials credentials = new Credentials();
-        // for HDFS
-        TokenCache.obtainTokensForNamenodes(credentials, paths.toArray(new Path[0]), conf);
-        // for HBase
-        obtainTokenForHBase(credentials, conf);
+
+        if (obtainingDelegationTokens) {
+            LOG.info("Obtaining delegation tokens for HDFS and HBase.");
+            // for HDFS
+            TokenCache.obtainTokensForNamenodes(credentials, paths.toArray(new Path[0]), conf);
+            // for HBase
+            obtainTokenForHBase(credentials, conf);
+        } else {
+            LOG.info("Delegation token retrieval for HDFS and HBase is disabled.");
+        }
+
         // for user
         UserGroupInformation currUsr = UserGroupInformation.getCurrentUser();
 

flink-yarn/src/main/java/org/apache/flink/yarn/YarnClusterDescriptor.java

@@ -529,6 +529,18 @@ private ClusterClientProvider<ApplicationId> deployInternal(
                         "Hadoop security with Kerberos is enabled but the login user "
                                 + "does not have Kerberos credentials or delegation tokens!");
             }
+
+            boolean fetchToken =
+                    flinkConfiguration.getBoolean(SecurityOptions.KERBEROS_FETCH_DELEGATION_TOKEN);
+            boolean yarnAccessFSEnabled =
+                    flinkConfiguration.get(YarnConfigOptions.YARN_ACCESS) != null;
+            if (!fetchToken && yarnAccessFSEnabled) {
+                throw new IllegalConfigurationException(
+                        String.format(
+                                "When %s is disabled, %s must be disabled as well.",
+                                SecurityOptions.KERBEROS_FETCH_DELEGATION_TOKEN.key(),
+                                YarnConfigOptions.YARN_ACCESS.key()));
+            }
         }
 
         isReadyForDeployment(clusterSpecification);
@@ -1081,13 +1093,16 @@ private ApplicationReport startAppMaster(
         if (UserGroupInformation.isSecurityEnabled()) {
             // set HDFS delegation tokens when security is enabled
             LOG.info("Adding delegation token to the AM container.");
-            List<Path> yarnAccessList =
-                    ConfigUtils.decodeListFromConfig(
-                            configuration, YarnConfigOptions.YARN_ACCESS, Path::new);
-            Utils.setTokensFor(
-                    amContainer,
-                    ListUtils.union(yarnAccessList, fileUploader.getRemotePaths()),
-                    yarnConfiguration);
+            List<Path> pathsToObtainToken = Collections.emptyList();
+            boolean fetchToken =
+                    configuration.getBoolean(SecurityOptions.KERBEROS_FETCH_DELEGATION_TOKEN);
+            if (fetchToken) {
+                List<Path> yarnAccessList =
+                        ConfigUtils.decodeListFromConfig(
+                                configuration, YarnConfigOptions.YARN_ACCESS, Path::new);
+                pathsToObtainToken = ListUtils.union(yarnAccessList, fileUploader.getRemotePaths());
+            }
+            Utils.setTokensFor(amContainer, pathsToObtainToken, yarnConfiguration, fetchToken);
         }
 
         amContainer.setLocalResources(fileUploader.getRegisteredLocalResources());