feat(stackable-operator): Support gitsync via SSH

crates/stackable-operator/CHANGELOG.md

@@ -13,6 +13,8 @@ All notable changes to this project will be documented in this file.
 
 - BREAKING: `ClusterResources` now requires the objects added to implement `DeepMerge`.
   This is very likely a stackable-operator internal change, but technically breaking ([#1118]).
+- BREAKING: Add support for the SSH protocol for pulling git content ([#1121]).
+  This is a user-facing breaking change and should also be highlighted by operators using this functionality.
 - Depend on the patched version of kube-rs available at <https://github.com/stackabletech/kube-rs>,
   ensuring the operators automatically benefit from the fixes ([#1124]).
 
@@ -21,6 +23,7 @@ All notable changes to this project will be documented in this file.
 - BREAKING: `ClusterResources` no longer derives `Eq` ([#1118]).
 
 [#1118]: https://github.com/stackabletech/operator-rs/pull/1118
+[#1121]: https://github.com/stackabletech/operator-rs/pull/1121
 [#1124]: https://github.com/stackabletech/operator-rs/pull/1124
 
 ## [0.100.3] - 2025-10-31

crates/stackable-operator/crds/DummyCluster.yaml

@@ -75,6 +75,83 @@ spec:
               domainName:
                 description: A validated domain name type conforming to RFC 1123, so e.g. not an IP address
                 type: string
+              gitSync:
+                properties:
+                  branch:
+                    default: main
+                    description: |-
+                      The branch to clone; defaults to `main`.
+
+                      Since git-sync v4.x.x this field is mapped to the flag `--ref`.
+                    type: string
+                  credentials:
+                    description: An optional secret used for git access.
+                    nullable: true
+                    oneOf:
+                    - required:
+                      - basicAuthSecretName
+                    - required:
+                      - sshPrivateKeySecretName
+                    properties:
+                      basicAuthSecretName:
+                        description: |-
+                          The name of the Secret used to access the repository via Basic Authentication if it is not public.
+
+                          The referenced Secret must include two fields: `user` and `password`.
+                          The `password` field can either be an actual password (not recommended) or a GitHub token,
+                          as described in the git-sync [documentation].
+
+                          [documentation]: https://github.com/kubernetes/git-sync/tree/v4.2.4?tab=readme-ov-file#manual
+                        type: string
+                      sshPrivateKeySecretName:
+                        description: |-
+                          The name of the Secret used for SSH access to the repository.
+
+                          The referenced Secret must include two fields: `key` and `knownHosts`.
+
+                          [documentation]: https://github.com/kubernetes/git-sync/tree/v4.2.4?tab=readme-ov-file#manual
+                        type: string
+                    type: object
+                  depth:
+                    default: 1
+                    description: The depth of syncing, i.e. the number of commits to clone; defaults to 1.
+                    format: uint32
+                    minimum: 0.0
+                    type: integer
+                  gitFolder:
+                    default: /
+                    description: |-
+                      Location in the Git repository containing the resource; defaults to the root folder.
+
+                      It can optionally start with `/`, however, no trailing slash is recommended.
+                      An empty string (``) or slash (`/`) corresponds to the root folder in Git.
+                    type: string
+                  gitSyncConf:
+                    additionalProperties:
+                      type: string
+                    default: {}
+                    description: |-
+                      A map of optional configuration settings that are listed in the git-sync [documentation].
+
+                      Also read the git-sync [example] in our documentation. These settings are not verified.
+
+                      [documentation]: https://github.com/kubernetes/git-sync/tree/v4.2.4?tab=readme-ov-file#manual
+                      [example]: https://docs.stackable.tech/home/nightly/airflow/usage-guide/mounting-dags#_example
+                    type: object
+                  repo:
+                    description: 'The git repository URL that will be cloned, for example: `https://github.com/stackabletech/airflow-operator` or `ssh://git@github.com:stackable-airflow/dags.git`.'
+                    format: uri
+                    type: string
+                  wait:
+                    default: 20s
+                    description: |-
+                      The synchronization interval, e.g. `20s` or `5m`; defaults to `20s`.
+
+                      Since git-sync v4.x.x this field is mapped to the flag `--period`.
+                    type: string
+                required:
+                - repo
+                type: object
               hostName:
                 type: string
               kerberosRealmName:
@@ -1402,6 +1479,7 @@ spec:
             - clientAuthenticationDetails
             - clusterOperation
             - domainName
+            - gitSync
             - hostName
             - kerberosRealmName
             - opaConfig

crates/stackable-operator/src/crd/git_sync/mod.rs

@@ -7,61 +7,98 @@ use serde::{Deserialize, Serialize};
 use stackable_shared::time::Duration;
 use url::Url;
 
-use crate::versioned::versioned;
+use crate::{crd::git_sync::v1alpha2::Credentials, versioned::versioned};
 
 mod v1alpha1_impl;
+mod v1alpha2_impl;
 
-#[versioned(version(name = "v1alpha1"))]
+#[versioned(version(name = "v1alpha1"), version(name = "v1alpha2"))]
 pub mod versioned {
     pub mod v1alpha1 {
         pub use v1alpha1_impl::{Error, GitSyncResources};
     }
+    pub mod v1alpha2 {
+        pub use v1alpha2_impl::{Error, GitSyncResources};
+    }
 
     #[derive(Clone, Debug, Deserialize, JsonSchema, PartialEq, Eq, Serialize)]
     #[serde(rename_all = "camelCase")]
     pub struct GitSync {
-        /// The git repository URL that will be cloned, for example: `https://github.com/stackabletech/airflow-operator`.
+        /// The git repository URL that will be cloned, for example: `https://github.com/stackabletech/airflow-operator` or `ssh://git@github.com:stackable-airflow/dags.git`.
         pub repo: Url,
 
         /// The branch to clone; defaults to `main`.
         ///
         /// Since git-sync v4.x.x this field is mapped to the flag `--ref`.
-        #[serde(default = "GitSync::default_branch")]
+        #[serde(default = "v1alpha2::GitSync::default_branch")]
         pub branch: String,
 
         /// Location in the Git repository containing the resource; defaults to the root folder.
         ///
         /// It can optionally start with `/`, however, no trailing slash is recommended.
         /// An empty string (``) or slash (`/`) corresponds to the root folder in Git.
-        #[serde(default = "GitSync::default_git_folder")]
+        #[serde(default = "v1alpha2::GitSync::default_git_folder")]
         pub git_folder: PathBuf,
 
         /// The depth of syncing, i.e. the number of commits to clone; defaults to 1.
-        #[serde(default = "GitSync::default_depth")]
+        #[serde(default = "v1alpha2::GitSync::default_depth")]
         pub depth: u32,
 
         /// The synchronization interval, e.g. `20s` or `5m`; defaults to `20s`.
         ///
         /// Since git-sync v4.x.x this field is mapped to the flag `--period`.
-        #[serde(default = "GitSync::default_wait")]
+        #[serde(default = "v1alpha2::GitSync::default_wait")]
         pub wait: Duration,
 
-        /// The name of the Secret used to access the repository if it is not public.
+        /// A map of optional configuration settings that are listed in the git-sync [documentation].
+        ///
+        /// Also read the git-sync [example] in our documentation. These settings are not verified.
+        ///
+        /// [documentation]: https://github.com/kubernetes/git-sync/tree/v4.2.4?tab=readme-ov-file#manual
+        /// [example]: DOCS_BASE_URL_PLACEHOLDER/airflow/usage-guide/mounting-dags#_example
+        #[serde(default)]
+        pub git_sync_conf: BTreeMap<String, String>,
+
+        /// An optional secret used for git access.
+        #[versioned(changed(
+            since = "v1alpha2",
+            from_name = "credentials_secret",
+            from_type = "Option<String>",
+            upgrade_with = credentials_secret_to_basic_auth,
+            downgrade_with = credentials_to_secret
+        ))]
+        pub credentials: Option<Credentials>,
+    }
+
+    #[derive(strum::Display, Clone, Debug, Deserialize, Eq, JsonSchema, PartialEq, Serialize)]
+    #[serde(rename_all = "camelCase")]
+    #[schemars(rename_all = "camelCase")]
+    pub enum Credentials {
+        /// The name of the Secret used to access the repository via Basic Authentication if it is not public.
         ///
         /// The referenced Secret must include two fields: `user` and `password`.
         /// The `password` field can either be an actual password (not recommended) or a GitHub token,
         /// as described in the git-sync [documentation].
         ///
         /// [documentation]: https://github.com/kubernetes/git-sync/tree/v4.2.4?tab=readme-ov-file#manual
-        pub credentials_secret: Option<String>,
-
-        /// A map of optional configuration settings that are listed in the git-sync [documentation].
+        BasicAuthSecretName(String),
+        /// The name of the Secret used for SSH access to the repository.
         ///
-        /// Also read the git-sync [example] in our documentation. These settings are not verified.
+        /// The referenced Secret must include two fields: `key` and `knownHosts`.
         ///
         /// [documentation]: https://github.com/kubernetes/git-sync/tree/v4.2.4?tab=readme-ov-file#manual
-        /// [example]: DOCS_BASE_URL_PLACEHOLDER/airflow/usage-guide/mounting-dags#_example
-        #[serde(default)]
-        pub git_sync_conf: BTreeMap<String, String>,
+        SshPrivateKeySecretName(String),
+    }
+}
+
+pub fn credentials_to_secret(input: Option<Credentials>) -> Option<String> {
+    if let Some(Credentials::BasicAuthSecretName(credentials_secret)) = input {
+        Some(credentials_secret)
+    } else {
+        None
     }
 }
+
+pub fn credentials_secret_to_basic_auth(input: Option<String>) -> Option<Credentials> {
+    input.map(Credentials::BasicAuthSecretName)
+}

crates/stackable-operator/src/crd/git_sync/v1alpha1_impl.rs

@@ -4,7 +4,6 @@ use k8s_openapi::api::core::v1::{
     Container, EmptyDirVolumeSource, EnvVar, EnvVarSource, SecretKeySelector, Volume, VolumeMount,
 };
 use snafu::{ResultExt, Snafu};
-use stackable_shared::time::Duration;
 use strum::{EnumDiscriminants, IntoStaticStr};
 
 use crate::{
@@ -42,24 +41,6 @@ pub enum Error {
     },
 }
 
-impl GitSync {
-    pub(crate) fn default_branch() -> String {
-        "main".to_string()
-    }
-
-    pub(crate) fn default_git_folder() -> PathBuf {
-        PathBuf::from("/")
-    }
-
-    pub(crate) fn default_depth() -> u32 {
-        1
-    }
-
-    pub(crate) fn default_wait() -> Duration {
-        Duration::from_secs(20)
-    }
-}
-
 /// Kubernetes resources generated from `GitSync` specifications which should be added to the Pod.
 #[derive(Default)]
 pub struct GitSyncResources {