fix: move metastore db credentials into a secret (#452)

* fix: move metastore db credentials into a secret * changelog * changed references in documentation/examples * changelog * Update rust/operator-binary/src/controller.rs Co-authored-by: Malte Sander <[email protected]> * Update CHANGELOG.md Co-authored-by: Malte Sander <[email protected]> * consistent ordering --------- Co-authored-by: Malte Sander <[email protected]>
stackabletech · May 16, 2024 · 2eb3f97 · 2eb3f97
1 parent 114bc0c
commit 2eb3f97
Show file tree

Hide file tree

Showing 25 changed files with 227 additions and 66 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -8,6 +8,10 @@ All notable changes to this project will be documented in this file.
 
 - Added documentation/tutorial on using external database drivers ([#449]).
 
+### Fixed
+
+- [BREAKING] Move the metastore `user` and `password` DB credentials out of the CRD into a Secret containing the keys `username` and `password` ([#452]).
+
 ### Changed
 
 - BREAKING: Switch to new image that only contains HMS.
@@ -17,6 +21,7 @@ All notable changes to this project will be documented in this file.
 
 [#447]: https://github.com/stackabletech/hive-operator/pull/447
 [#449]: https://github.com/stackabletech/hive-operator/pull/449
+[#452]: https://github.com/stackabletech/hive-operator/pull/452
 
 ## [24.3.0] - 2024-03-20
 

diff --git a/deploy/helm/hive-operator/crds/crds.yaml b/deploy/helm/hive-operator/crds/crds.yaml
@@ -50,6 +50,9 @@ spec:
                         connString:
                           description: 'A connection string for the database. For example: `jdbc:postgresql://hivehdfs-postgresql:5432/hivehdfs`'
                           type: string
+                        credentialsSecret:
+                          description: A reference to a Secret containing the database credentials. The Secret needs to contain the keys `username` and `password`.
+                          type: string
                         dbType:
                           description: 'The type of database to connect to. Supported are: `postgres`, `mysql`, `oracle`, `mssql` and `derby`. This value is used to configure the jdbc driver class.'
                           enum:
@@ -59,17 +62,10 @@ spec:
                             - oracle
                             - mssql
                           type: string
-                        password:
-                          description: The password for the database user.
-                          type: string
-                        user:
-                          description: The database user.
-                          type: string
                       required:
                         - connString
+                        - credentialsSecret
                         - dbType
-                        - password
-                        - user
                       type: object
                     hdfs:
                       description: HDFS connection specification.

diff --git a/docs/modules/hive/examples/getting_started/hive-postgres-s3.yaml b/docs/modules/hive/examples/getting_started/hive-postgres-s3.yaml
@@ -9,12 +9,20 @@ spec:
   clusterConfig:
     database:
       connString: jdbc:postgresql://postgresql:5432/hive
-      user: hive
-      password: hive
+      credentialsSecret: hive-credentials
       dbType: postgres
     s3:
       reference: minio
   metastore:
     roleGroups:
       default:
         replicas: 1
+---
+apiVersion: v1
+kind: Secret
+metadata:
+  name: hive-credentials
+type: Opaque
+stringData:
+  username: hive
+  password: hive
diff --git a/docs/modules/hive/examples/getting_started/hive-postgres-s3.yaml.j2 b/docs/modules/hive/examples/getting_started/hive-postgres-s3.yaml.j2
@@ -9,12 +9,20 @@ spec:
   clusterConfig:
     database:
       connString: jdbc:postgresql://postgresql:5432/hive
-      user: hive
-      password: hive
+      credentialsSecret: hive-credentials
       dbType: postgres
     s3:
       reference: minio
   metastore:
     roleGroups:
       default:
         replicas: 1
+---
+apiVersion: v1
+kind: Secret
+metadata:
+  name: hive-credentials
+type: Opaque
+stringData:
+  username: hive
+  password: hive
diff --git a/docs/modules/hive/pages/reference/discovery.adoc b/docs/modules/hive/pages/reference/discovery.adoc
@@ -25,13 +25,21 @@ spec:
   clusterConfig:
     database:
       connString: jdbc:postgresql://postgresql:5432/hive
-      user: hive
-      password: hive
+      credentialsSecret: hive-credentials
       dbType: postgres
   metastore:
     roleGroups:
       default: # <3>
         replicas: 2
+---
+apiVersion: v1
+kind: Secret
+metadata:
+  name: hive-credentials
+type: Opaque
+stringData:
+  username: hive
+  password: hive
 ----
 <1> The name of the Hive cluster, which is also the name of the created discovery ConfigMap.
 <2> The namespace of the discovery ConfigMap.

diff --git a/docs/modules/hive/pages/usage-guide/database-driver.adoc b/docs/modules/hive/pages/usage-guide/database-driver.adoc
@@ -145,8 +145,7 @@ spec:
   clusterConfig:
     database:
       connString: jdbc:mysql://mysql:3306/hive  # <1>
-      user: hive  # <2>
-      password: hive
+      credentialsSecret: hive-credentials  # <2>
       dbType: mysql
     s3:
       reference: minio  # <3>
@@ -167,10 +166,19 @@ spec:
                 persistentVolumeClaim:
                   claimName: pvc-hive-drivers
         replicas: 1
+---
+apiVersion: v1
+kind: Secret
+metadata:
+  name: hive-credentials  # <2>
+type: Opaque
+stringData:
+  username: hive
+  password: hive
 ----
 
 <1> The database connection details matching those given when deploying the MySQL Helm chart
-<2> Plain-text Hive credentials will be replaced in an upcoming release!
+<2> Hive credentials are retrieved from a Secret
 <3> A reference to the file store using S3 (this has been omitted from this article for the sake of brevity, but is described in e.g. the xref:getting_started/first_steps.adoc[] guide)
 <4> Use `envOverrides` to set the driver path
 <5> Use `podOverrides` to mount the driver

diff --git a/docs/modules/hive/pages/usage-guide/derby-example.adoc b/docs/modules/hive/pages/usage-guide/derby-example.adoc
@@ -20,13 +20,21 @@ spec:
   clusterConfig:
     database:
       connString: jdbc:derby:;databaseName=/tmp/metastore_db;create=true
-      user: APP
-      password: mine
+      credentialsSecret: hive-credentials
       dbType: derby
   metastore:
     roleGroups:
       default:
         replicas: 1
+---
+apiVersion: v1
+kind: Secret
+metadata:
+  name: hive-credentials
+type: Opaque
+stringData:
+  username: APP
+  password: mine
 ----
 
 WARNING: You should not use the `Derby` database in production. Derby stores data locally which does not work in high availability setups (multiple replicas) and all data is lost after Pod restarts.
@@ -62,8 +70,7 @@ spec:
   clusterConfig:
     database:
       connString: jdbc:derby:;databaseName=/stackable/metastore_db;create=true
-      user: APP
-      password: mine
+      credentialsSecret: hive-credentials
       dbType: derby
     s3:
       inline:
@@ -96,6 +103,15 @@ metadata:
 stringData:
   accessKey: minio-access-key
   secretKey: minio-secret-key
+---
+apiVersion: v1
+kind: Secret
+metadata:
+  name: hive-credentials
+type: Opaque
+stringData:
+  username: APP
+  password: mine
 ----
 
 
@@ -131,11 +147,19 @@ spec:
   clusterConfig:
     database:
       connString: jdbc:postgresql://hive-postgresql.default.svc.cluster.local:5432/hive
-      user: hive
-      password: hive
+      credentialsSecret: hive-credentials
       dbType: postgres
   metastore:
     roleGroups:
       default:
         replicas: 1
+---
+apiVersion: v1
+kind: Secret
+metadata:
+  name: hive-credentials
+type: Opaque
+stringData:
+  username: hive
+  password: hive
 ----
diff --git a/examples/simple-hive-cluster-postgres-s3.yaml b/examples/simple-hive-cluster-postgres-s3.yaml
@@ -22,8 +22,7 @@ spec:
   clusterConfig:
     database:
       connString: jdbc:derby:;databaseName=/tmp/hive;create=true
-      user: APP
-      password: mine
+      credentialsSecret: hive-credentials
       dbType: derby
     s3:
       inline:
@@ -56,3 +55,12 @@ metadata:
 stringData:
   accessKey: minio-access-key
   secretKey: minio-secret-key
+---
+apiVersion: v1
+kind: Secret
+metadata:
+  name: hive-credentials
+type: Opaque
+stringData:
+  username: APP
+  password: mine
diff --git a/examples/simple-hive-cluster.yaml b/examples/simple-hive-cluster.yaml
@@ -10,8 +10,7 @@ spec:
   clusterConfig:
     database:
       connString: jdbc:derby:;databaseName=/tmp/hive;create=true
-      user: APP
-      password: mine
+      credentialsSecret: hive-credentials
       dbType: derby
   metastore:
     roleGroups:
@@ -24,3 +23,12 @@ spec:
               max: "2"
             memory:
               limit: 5Gi
+---
+apiVersion: v1
+kind: Secret
+metadata:
+  name: hive-credentials
+type: Opaque
+stringData:
+  username: APP
+  password: mine
diff --git a/rust/crd/src/affinity.rs b/rust/crd/src/affinity.rs
@@ -49,9 +49,8 @@ mod tests {
           clusterConfig:
             database:
               connString: jdbc:derby:;databaseName=/tmp/hive;create=true
-              user: APP
-              password: mine
               dbType: derby
+              credentialsSecret: mySecret
           metastore:
             roleGroups:
               default:

diff --git a/rust/crd/src/lib.rs b/rust/crd/src/lib.rs
@@ -71,6 +71,12 @@ pub const HADOOP_OPTS: &str = "HADOOP_OPTS";
 pub const HADOOP_HEAPSIZE: &str = "HADOOP_HEAPSIZE";
 pub const JVM_HEAP_FACTOR: f32 = 0.8;
 
+// DB credentials
+pub const DB_USERNAME_PLACEHOLDER: &str = "xxx_db_username_xxx";
+pub const DB_PASSWORD_PLACEHOLDER: &str = "xxx_db_password_xxx";
+pub const DB_USERNAME_ENV: &str = "DB_USERNAME_ENV";
+pub const DB_PASSWORD_ENV: &str = "DB_PASSWORD_ENV";
+
 const DEFAULT_METASTORE_GRACEFUL_SHUTDOWN_TIMEOUT: Duration = Duration::from_minutes_unchecked(5);
 
 #[derive(Snafu, Debug)]
@@ -422,16 +428,14 @@ pub struct DatabaseConnectionSpec {
     /// `jdbc:postgresql://hivehdfs-postgresql:5432/hivehdfs`
     pub conn_string: String,
 
-    /// The database user.
-    pub user: String,
-
-    /// The password for the database user.
-    pub password: String,
-
     /// The type of database to connect to. Supported are:
     /// `postgres`, `mysql`, `oracle`, `mssql` and `derby`.
     /// This value is used to configure the jdbc driver class.
     pub db_type: DbType,
+
+    /// A reference to a Secret containing the database credentials.
+    /// The Secret needs to contain the keys `username` and `password`.
+    pub credentials_secret: String,
 }
 
 impl Configuration for MetaStoreConfigFragment {
@@ -492,13 +496,14 @@ impl Configuration for MetaStoreConfigFragment {
                     MetaStoreConfig::CONNECTION_URL.to_string(),
                     Some(hive.spec.cluster_config.database.conn_string.clone()),
                 );
+                // use a placeholder that will be replaced in the start command (also for the password)
                 result.insert(
                     MetaStoreConfig::CONNECTION_USER_NAME.to_string(),
-                    Some(hive.spec.cluster_config.database.user.clone()),
+                    Some(DB_USERNAME_PLACEHOLDER.into()),
                 );
                 result.insert(
                     MetaStoreConfig::CONNECTION_PASSWORD.to_string(),
-                    Some(hive.spec.cluster_config.database.password.clone()),
+                    Some(DB_PASSWORD_PLACEHOLDER.into()),
                 );
                 result.insert(
                     MetaStoreConfig::CONNECTION_DRIVER_NAME.to_string(),

diff --git a/rust/operator-binary/src/command.rs b/rust/operator-binary/src/command.rs
@@ -1,5 +1,6 @@
 use stackable_hive_crd::{
-    HiveCluster, HIVE_METASTORE_LOG4J2_PROPERTIES, HIVE_SITE_XML, STACKABLE_CONFIG_DIR,
+    HiveCluster, DB_PASSWORD_ENV, DB_PASSWORD_PLACEHOLDER, DB_USERNAME_ENV,
+    DB_USERNAME_PLACEHOLDER, HIVE_METASTORE_LOG4J2_PROPERTIES, HIVE_SITE_XML, STACKABLE_CONFIG_DIR,
     STACKABLE_CONFIG_MOUNT_DIR, STACKABLE_LOG_CONFIG_MOUNT_DIR, STACKABLE_TRUST_STORE,
     STACKABLE_TRUST_STORE_PASSWORD, SYSTEM_TRUST_STORE, SYSTEM_TRUST_STORE_PASSWORD,
 };
@@ -59,6 +60,13 @@ pub fn build_container_command_args(
         }
     }
 
+    // db credentials
+    args.extend([
+        format!("echo replacing {DB_USERNAME_PLACEHOLDER} and {DB_PASSWORD_PLACEHOLDER} with secret values."),
+        format!("sed -i \"s|{DB_USERNAME_PLACEHOLDER}|${DB_USERNAME_ENV}|g\" {STACKABLE_CONFIG_DIR}/{HIVE_SITE_XML}"),
+        format!("sed -i \"s|{DB_PASSWORD_PLACEHOLDER}|${DB_PASSWORD_ENV}|g\" {STACKABLE_CONFIG_DIR}/{HIVE_SITE_XML}"),
+    ]);
+
     // metastore start command
     args.push(start_command);