feat: deprecate stuck statue and auto re-onboard stuck repos [CM-1098] (#4034)

Signed-off-by: Mouad BANI <mouad-mb@outlook.com>

Author: mbani01

backend/src/database/migrations/U1776428661__dropStuckFlagFromRepositoryProcessing.sql

@@ -0,0 +1 @@
+ALTER TABLE git."repositoryProcessing" DROP COLUMN IF EXISTS "stuckRequiresReOnboard";

backend/src/database/migrations/V1776428661__dropStuckFlagFromRepositoryProcessing.sql

@@ -31,7 +31,6 @@
     rp.branch,
     rp."maintainerFile",
     rp."lastMaintainerRunAt",
-    rp."stuckRequiresReOnboard",
     rp."reOnboardingCount"
 """
 
@@ -166,7 +165,6 @@ async def acquire_recurrent_repo() -> Repository | None:
     states_to_exclude = (
         RepositoryState.PENDING,
         RepositoryState.PROCESSING,
-        RepositoryState.STUCK,
         RepositoryState.PENDING_REONBOARD,
         RepositoryState.AUTH_REQUIRED,
     )

services/apps/git_integration/src/crowdgit/database/crud.py

@@ -21,7 +21,6 @@ class ErrorCode(str, Enum):
     CLEANUP_FAILED = "cleanup-failed"
     PARENT_REPO_INVALID = "parent-repo-invalid"
     REONBOARDING_REQUIRED = "reonboarding-required"
-    STUCK_REPO = "stuck-repo"
     REPO_AUTH_REQUIRED = "repo-auth-required"
     RATE_LIMITED = "rate-limited"
     ACCESS_FORBIDDEN = "access-forbidden"
@@ -37,7 +36,6 @@ class RepositoryState(str, Enum):
     COMPLETED = "completed"
     FAILED = "failed"
     REQUIRES_PARENT = "requires_parent"  # fork repo without valid parent repo in out system
-    STUCK = "stuck"  # requires manual resolution
     PENDING_REONBOARD = "pending_reonboard"  # re-onboarding deferred until weekend
     AUTH_REQUIRED = "auth_required"  # private repo or repo requiring authentication
 

services/apps/git_integration/src/crowdgit/enums.py

@@ -116,12 +116,6 @@ class ReOnboardingRequiredError(CrowdGitError):
     error_code: ErrorCode = ErrorCode.REONBOARDING_REQUIRED
 
 
-@dataclass
-class StuckRepoError(CrowdGitError):
-    error_message = "Repos stuck in processing state for a long time"
-    error_code: ErrorCode = ErrorCode.STUCK_REPO
-
-
 @dataclass
 class RepoAuthRequiredError(CrowdGitError):
     error_message: str = "Repository requires authentication (likely private or deleted)"

services/apps/git_integration/src/crowdgit/errors.py

@@ -41,10 +41,6 @@ class Repository(BaseModel):
     parent_repo: Repository | None = Field(
         None, description="The parent repository (in case of fork) object from our database"
     )
-    stuck_requires_re_onboard: bool = Field(
-        default=False,
-        description="Indicates if the stuck repository is resolved by a re-onboarding",
-    )
     re_onboarding_count: int = Field(
         default=0,
         description="Tracks the number of times this repository has been re-onboarded. Used to identify unreachable commits via activity.attributes.cycle matching pattern onboarding-{reOnboardingCount}",
@@ -71,7 +67,6 @@ def from_db(cls, db_data: dict[str, Any]) -> Repository:
             "maintainerFile": "maintainer_file",
             "lastMaintainerRunAt": "last_maintainer_run_at",
             "forkedFrom": "forked_from",
-            "stuckRequiresReOnboard": "stuck_requires_re_onboard",
             "reOnboardingCount": "re_onboarding_count",
         }
         for db_field, model_field in field_mapping.items():

services/apps/git_integration/src/crowdgit/models/repository.py

@@ -14,7 +14,6 @@
     ParentRepoInvalidError,
     ReOnboardingRequiredError,
     RepoAuthRequiredError,
-    StuckRepoError,
 )
 
 # Import configured loguru logger from crowdgit.logger
@@ -117,14 +116,9 @@ async def _ensure_repo_not_stuck(self, repository: Repository):
         # handling
         if repo_stuck:
             logger.warning(
-                f"Repo {repository.url} is stuck for {processing_duration_hours} hours!"
+                f"Repo {repository.url} is stuck for {processing_duration_hours} hours — queuing for re-onboarding"
             )
-            if repository.stuck_requires_re_onboard:
-                logger.warning(
-                    f"Repo {repository.url} is stuck due to force-push or dangling commit. Will be re-onboarded"
-                )
-                raise ReOnboardingRequiredError()
-            raise StuckRepoError()
+            raise ReOnboardingRequiredError()
 
     async def _process_repositories(self):
         """
@@ -258,11 +252,6 @@ async def _process_single_repository(self, repository: Repository):
 
             logger.info("Incremental processing completed successfully")
             processing_state = RepositoryState.COMPLETED
-        except StuckRepoError:
-            logger.error(
-                f"Repo {repository.url} is stuck for unkown reason, marking it as stuck until manually resolved!"
-            )
-            processing_state = RepositoryState.STUCK
         except ReOnboardingRequiredError:
             logger.info(f"Repo {repository.url} needs re-onboarding, deferring until weekend")
             processing_state = RepositoryState.PENDING_REONBOARD