cartography 0.108.0rc1__py3-none-any.whl → 0.109.0rc1__py3-none-any.whl

cartography/intel/aws/elasticache.py

@@ -1,118 +1,132 @@
 import logging
-from typing import Dict
-from typing import List
-from typing import Set
+from typing import Any
 
 import boto3
 import neo4j
 
+from cartography.client.core.tx import load
+from cartography.graph.job import GraphJob
+from cartography.models.aws.elasticache.cluster import ElasticacheClusterSchema
+from cartography.models.aws.elasticache.topic import ElasticacheTopicSchema
 from cartography.stats import get_stats_client
 from cartography.util import aws_handle_regions
 from cartography.util import merge_module_sync_metadata
-from cartography.util import run_cleanup_job
 from cartography.util import timeit
 
 logger = logging.getLogger(__name__)
 stat_handler = get_stats_client(__name__)
 
 
-def _get_topic(cluster: Dict) -> Dict:
-    return cluster["NotificationConfiguration"]
-
-
-def transform_elasticache_topics(cluster_data: List[Dict]) -> List[Dict]:
-    """
-    Collect unique TopicArns from the cluster data
-    """
-    seen: Set[str] = set()
-    topics: List[Dict] = []
-    for cluster in cluster_data:
-        topic = _get_topic(cluster)
-        topic_arn = topic["TopicArn"]
-        if topic_arn not in seen:
-            seen.add(topic_arn)
-            topics.append(topic)
-    return topics
-
-
 @timeit
 @aws_handle_regions
 def get_elasticache_clusters(
     boto3_session: boto3.session.Session,
     region: str,
-) -> List[Dict]:
-    logger.debug(f"Getting ElastiCache Clusters in region '{region}'.")
+) -> list[dict[str, Any]]:
     client = boto3_session.client("elasticache", region_name=region)
     paginator = client.get_paginator("describe_cache_clusters")
-    clusters: List[Dict] = []
+    clusters: list[dict[str, Any]] = []
     for page in paginator.paginate():
-        clusters.extend(page["CacheClusters"])
+        clusters.extend(page.get("CacheClusters", []))
     return clusters
 
 
+def transform_elasticache_clusters(
+    clusters: list[dict[str, Any]], region: str
+) -> tuple[list[dict[str, Any]], list[dict[str, Any]]]:
+    cluster_data: list[dict[str, Any]] = []
+    topics: dict[str, dict[str, Any]] = {}
+
+    for cluster in clusters:
+        notification = cluster.get("NotificationConfiguration", {})
+        topic_arn = notification.get("TopicArn")
+        cluster_record = {
+            "ARN": cluster["ARN"],
+            "CacheClusterId": cluster["CacheClusterId"],
+            "CacheNodeType": cluster.get("CacheNodeType"),
+            "Engine": cluster.get("Engine"),
+            "EngineVersion": cluster.get("EngineVersion"),
+            "CacheClusterStatus": cluster.get("CacheClusterStatus"),
+            "NumCacheNodes": cluster.get("NumCacheNodes"),
+            "PreferredAvailabilityZone": cluster.get("PreferredAvailabilityZone"),
+            "PreferredMaintenanceWindow": cluster.get("PreferredMaintenanceWindow"),
+            "CacheClusterCreateTime": cluster.get("CacheClusterCreateTime"),
+            "CacheSubnetGroupName": cluster.get("CacheSubnetGroupName"),
+            "AutoMinorVersionUpgrade": cluster.get("AutoMinorVersionUpgrade"),
+            "ReplicationGroupId": cluster.get("ReplicationGroupId"),
+            "SnapshotRetentionLimit": cluster.get("SnapshotRetentionLimit"),
+            "SnapshotWindow": cluster.get("SnapshotWindow"),
+            "AuthTokenEnabled": cluster.get("AuthTokenEnabled"),
+            "TransitEncryptionEnabled": cluster.get("TransitEncryptionEnabled"),
+            "AtRestEncryptionEnabled": cluster.get("AtRestEncryptionEnabled"),
+            "TopicArn": topic_arn,
+            "Region": region,
+        }
+        cluster_data.append(cluster_record)
+
+        if topic_arn:
+            topics.setdefault(
+                topic_arn,
+                {
+                    "TopicArn": topic_arn,
+                    "TopicStatus": notification.get("TopicStatus"),
+                    "cluster_arns": [],
+                },
+            )["cluster_arns"].append(cluster["ARN"])
+
+    return cluster_data, list(topics.values())
+
+
 @timeit
 def load_elasticache_clusters(
     neo4j_session: neo4j.Session,
-    clusters: List[Dict],
+    clusters: list[dict[str, Any]],
     region: str,
     aws_account_id: str,
     update_tag: int,
 ) -> None:
-    query = """
-    UNWIND $clusters as elasticache_cluster
-        MERGE (cluster:ElasticacheCluster{id:elasticache_cluster.ARN})
-        ON CREATE SET cluster.firstseen = timestamp(),
-            cluster.arn = elasticache_cluster.ARN,
-            cluster.topic_arn = elasticache_cluster.NotificationConfiguration.TopicArn,
-            cluster.id = elasticache_cluster.CacheClusterId,
-            cluster.region = $region
-        SET cluster.lastupdated = $aws_update_tag
-
-        WITH cluster, elasticache_cluster
-        MATCH (owner:AWSAccount{id: $aws_account_id})
-        MERGE (owner)-[r3:RESOURCE]->(cluster)
-        ON CREATE SET r3.firstseen = timestamp()
-        SET r3.lastupdated = $aws_update_tag
-
-        WITH elasticache_cluster, owner
-        WHERE NOT elasticache_cluster.NotificationConfiguration IS NULL
-        MERGE (topic:ElasticacheTopic{id: elasticache_cluster.NotificationConfiguration.TopicArn})
-        ON CREATE SET topic.firstseen = timestamp(),
-            topic.arn = elasticache_cluster.NotificationConfiguration.TopicArn
-        SET topic.lastupdated = $aws_update_tag,
-            topic.status = elasticache_cluster.NotificationConfiguration.Status
-
-        MERGE (topic)-[r:CACHE_CLUSTER]->(cluster)
-        ON CREATE SET r.firstseen = timestamp()
-        SET r.lastupdated = $aws_update_tag
-        WITH cluster, topic
-
-        MERGE (owner)-[r2:RESOURCE]->(topic)
-        ON CREATE SET r2.firstseen = timestamp()
-        SET r2.lastupdated = $aws_update_tag
-    """
     logger.info(
-        f"Loading f{len(clusters)} ElastiCache clusters for region '{region}' into graph.",
+        f"Loading {len(clusters)} ElastiCache clusters for region '{region}' into graph."
     )
-    neo4j_session.run(
-        query,
-        clusters=clusters,
-        region=region,
-        aws_update_tag=update_tag,
-        aws_account_id=aws_account_id,
+    load(
+        neo4j_session,
+        ElasticacheClusterSchema(),
+        clusters,
+        lastupdated=update_tag,
+        Region=region,
+        AWS_ID=aws_account_id,
     )
 
 
 @timeit
-def cleanup(
+def load_elasticache_topics(
     neo4j_session: neo4j.Session,
-    current_aws_account_id: str,
+    topics: list[dict[str, Any]],
+    aws_account_id: str,
     update_tag: int,
 ) -> None:
-    run_cleanup_job(
-        "aws_import_elasticache_cleanup.json",
+    if not topics:
+        return
+    logger.info(f"Loading {len(topics)} ElastiCache topics into graph.")
+    load(
         neo4j_session,
-        {"UPDATE_TAG": update_tag, "AWS_ID": current_aws_account_id},
+        ElasticacheTopicSchema(),
+        topics,
+        lastupdated=update_tag,
+        AWS_ID=aws_account_id,
+    )
+
+
+@timeit
+def cleanup(
+    neo4j_session: neo4j.Session,
+    common_job_parameters: dict[str, Any],
+) -> None:
+    GraphJob.from_node_schema(ElasticacheClusterSchema(), common_job_parameters).run(
+        neo4j_session
+    )
+    GraphJob.from_node_schema(ElasticacheTopicSchema(), common_job_parameters).run(
+        neo4j_session
     )
 
 
@@ -120,24 +134,33 @@ def cleanup(
 def sync(
     neo4j_session: neo4j.Session,
     boto3_session: boto3.session.Session,
-    regions: List[str],
+    regions: list[str],
     current_aws_account_id: str,
     update_tag: int,
-    common_job_parameters: Dict,
+    common_job_parameters: dict[str, Any],
 ) -> None:
     for region in regions:
         logger.info(
-            f"Syncing ElastiCache clusters for region '{region}' in account {current_aws_account_id}",
+            "Syncing ElastiCache clusters for region '%s' in account '%s'.",
+            region,
+            current_aws_account_id,
         )
-        clusters = get_elasticache_clusters(boto3_session, region)
+        raw_clusters = get_elasticache_clusters(boto3_session, region)
+        cluster_data, topic_data = transform_elasticache_clusters(raw_clusters, region)
         load_elasticache_clusters(
             neo4j_session,
-            clusters,
+            cluster_data,
             region,
             current_aws_account_id,
             update_tag,
         )
-    cleanup(neo4j_session, current_aws_account_id, update_tag)
+        load_elasticache_topics(
+            neo4j_session,
+            topic_data,
+            current_aws_account_id,
+            update_tag,
+        )
+    cleanup(neo4j_session, common_job_parameters)
     merge_module_sync_metadata(
         neo4j_session,
         group_type="AWSAccount",

cartography/intel/aws/resourcegroupstaggingapi.py

@@ -1,5 +1,6 @@
 import logging
 from string import Template
+from typing import Any
 from typing import Dict
 from typing import List
 
@@ -56,6 +57,35 @@ def get_short_id_from_lb2_arn(alb_arn: str) -> str:
     return alb_arn.split("/")[-2]
 
 
+def get_resource_type_from_arn(arn: str) -> str:
+    """Return the resource type format expected by the Tagging API.
+
+    The Resource Groups Tagging API requires resource types in the form
+    ``service:resource``. Most ARNs embed the resource type in the fifth segment
+    after the service name. Load balancer ARNs add an extra ``app`` or ``net``
+    component that must be preserved. S3 and SQS ARNs only contain the service
+    name.  This helper extracts the appropriate string so that ARNs can be
+    grouped correctly for API calls.
+    """
+
+    parts = arn.split(":", 5)
+    service = parts[2]
+    if service in {"s3", "sqs"}:
+        return service
+
+    resource = parts[5]
+    if service == "elasticloadbalancing" and resource.startswith("loadbalancer/"):
+        segments = resource.split("/")
+        if len(segments) > 2 and segments[1] in {"app", "net"}:
+            resource_type = f"{segments[0]}/{segments[1]}"
+        else:
+            resource_type = segments[0]
+    else:
+        resource_type = resource.split("/")[0].split(":")[0]
+
+    return f"{service}:{resource_type}" if resource_type else service
+
+
 # We maintain a mapping from AWS resource types to their associated labels and unique identifiers.
 # label: the node label used in cartography for this resource type
 # property: the field of this node that uniquely identified this resource type
@@ -158,27 +188,27 @@ TAG_RESOURCE_TYPE_MAPPINGS: Dict = {
 @aws_handle_regions
 def get_tags(
     boto3_session: boto3.session.Session,
-    resource_type: str,
+    resource_types: list[str],
     region: str,
-) -> List[Dict]:
-    """
-    Create boto3 client and retrieve tag data.
-    """
-    # this is a temporary workaround to populate AWS tags for IAM roles.
-    # resourcegroupstaggingapi does not support IAM roles and no ETA is provided
-    # TODO: when resourcegroupstaggingapi supports iam:role, remove this condition block
-    if resource_type == "iam:role":
-        return get_role_tags(boto3_session)
+) -> list[dict[str, Any]]:
+    """Retrieve tag data for the provided resource types."""
+    resources: list[dict[str, Any]] = []
+
+    if "iam:role" in resource_types:
+        resources.extend(get_role_tags(boto3_session))
+        resource_types = [rt for rt in resource_types if rt != "iam:role"]
+
+    if not resource_types:
+        return resources
 
     client = boto3_session.client("resourcegroupstaggingapi", region_name=region)
     paginator = client.get_paginator("get_resources")
-    resources: List[Dict] = []
-    for page in paginator.paginate(
-        # Only ingest tags for resources that Cartography supports.
-        # This is just a starting list; there may be others supported by this API.
-        ResourceTypeFilters=[resource_type],
-    ):
-        resources.extend(page["ResourceTagMappingList"])
+
+    # Batch resource types into groups of 100
+    # (https://docs.aws.amazon.com/resourcegroupstagging/latest/APIReference/API_GetResources.html)
+    for resource_types_batch in batch(resource_types, size=100):
+        for page in paginator.paginate(ResourceTypeFilters=resource_types_batch):
+            resources.extend(page["ResourceTagMappingList"])
     return resources
 
 
@@ -210,6 +240,9 @@ def _load_tags_tx(
             r.firstseen = timestamp()
     """,
     )
+    if not tag_data:
+        return
+
     query = INGEST_TAG_TEMPLATE.safe_substitute(
         resource_label=TAG_RESOURCE_TYPE_MAPPINGS[resource_type]["label"],
         property=TAG_RESOURCE_TYPE_MAPPINGS[resource_type]["property"],
@@ -262,6 +295,26 @@ def compute_resource_id(tag_mapping: Dict, resource_type: str) -> str:
     return resource_id
 
 
+def _group_tag_data_by_resource_type(
+    tag_data: List[Dict],
+    tag_resource_type_mappings: Dict,
+) -> Dict[str, List[Dict]]:
+    """Group raw tag data by the resource types Cartography supports."""
+
+    grouped: Dict[str, List[Dict]] = {rtype: [] for rtype in tag_resource_type_mappings}
+    for mapping in tag_data:
+        rtype = get_resource_type_from_arn(mapping["ResourceARN"])
+        if rtype in grouped:
+            grouped[rtype].append(mapping)
+        else:
+            logger.debug(
+                "Unknown tag resource type %s from ARN %s",
+                rtype,
+                mapping["ResourceARN"],
+            )
+    return grouped
+
+
 @timeit
 def cleanup(neo4j_session: neo4j.Session, common_job_parameters: Dict) -> None:
     run_cleanup_job(
@@ -285,8 +338,14 @@ def sync(
         logger.info(
             f"Syncing AWS tags for account {current_aws_account_id} and region {region}",
         )
+        all_tag_data = get_tags(
+            boto3_session, list(tag_resource_type_mappings.keys()), region
+        )
+        grouped = _group_tag_data_by_resource_type(
+            all_tag_data, tag_resource_type_mappings
+        )
         for resource_type in tag_resource_type_mappings.keys():
-            tag_data = get_tags(boto3_session, resource_type, region)
+            tag_data = grouped.get(resource_type, [])
             transform_tags(tag_data, resource_type)  # type: ignore
             logger.info(
                 f"Loading {len(tag_data)} tags for resource type {resource_type}",

cartography/intel/aws/secretsmanager.py

@@ -7,6 +7,7 @@ import neo4j
 
 from cartography.client.core.tx import load
 from cartography.graph.job import GraphJob
+from cartography.models.aws.secretsmanager.secret import SecretsManagerSecretSchema
 from cartography.models.aws.secretsmanager.secret_version import (
     SecretsManagerSecretVersionSchema,
 )
@@ -14,7 +15,6 @@ from cartography.stats import get_stats_client
 from cartography.util import aws_handle_regions
 from cartography.util import dict_date_to_epoch
 from cartography.util import merge_module_sync_metadata
-from cartography.util import run_cleanup_job
 from cartography.util import timeit
 
 logger = logging.getLogger(__name__)
@@ -32,6 +32,37 @@ def get_secret_list(boto3_session: boto3.session.Session, region: str) -> List[D
     return secrets
 
 
+def transform_secrets(
+    secrets: List[Dict],
+) -> List[Dict]:
+    """
+    Transform AWS Secrets Manager Secrets to match the data model.
+    """
+    transformed_data = []
+    for secret in secrets:
+        # Start with a copy of the original secret data
+        transformed = dict(secret)
+
+        # Convert date fields to epoch timestamps
+        transformed["CreatedDate"] = dict_date_to_epoch(secret, "CreatedDate")
+        transformed["LastRotatedDate"] = dict_date_to_epoch(secret, "LastRotatedDate")
+        transformed["LastChangedDate"] = dict_date_to_epoch(secret, "LastChangedDate")
+        transformed["LastAccessedDate"] = dict_date_to_epoch(secret, "LastAccessedDate")
+        transformed["DeletedDate"] = dict_date_to_epoch(secret, "DeletedDate")
+
+        # Flatten nested RotationRules.AutomaticallyAfterDays property
+        if "RotationRules" in secret and secret["RotationRules"]:
+            rotation_rules = secret["RotationRules"]
+            if "AutomaticallyAfterDays" in rotation_rules:
+                transformed["RotationRulesAutomaticallyAfterDays"] = rotation_rules[
+                    "AutomaticallyAfterDays"
+                ]
+
+        transformed_data.append(transformed)
+
+    return transformed_data
+
+
 @timeit
 def load_secrets(
     neo4j_session: neo4j.Session,
@@ -40,48 +71,33 @@ def load_secrets(
     current_aws_account_id: str,
     aws_update_tag: int,
 ) -> None:
-    ingest_secrets = """
-    UNWIND $Secrets as secret
-        MERGE (s:SecretsManagerSecret{id: secret.ARN})
-        ON CREATE SET s.firstseen = timestamp()
-        SET s.name = secret.Name, s.arn = secret.ARN, s.description = secret.Description,
-            s.kms_key_id = secret.KmsKeyId, s.rotation_enabled = secret.RotationEnabled,
-            s.rotation_lambda_arn = secret.RotationLambdaARN,
-            s.rotation_rules_automatically_after_days = secret.RotationRules.AutomaticallyAfterDays,
-            s.last_rotated_date = secret.LastRotatedDate, s.last_changed_date = secret.LastChangedDate,
-            s.last_accessed_date = secret.LastAccessedDate, s.deleted_date = secret.DeletedDate,
-            s.owning_service = secret.OwningService, s.created_date = secret.CreatedDate,
-            s.primary_region = secret.PrimaryRegion, s.region = $Region,
-            s.lastupdated = $aws_update_tag
-        WITH s
-        MATCH (owner:AWSAccount{id: $AWS_ACCOUNT_ID})
-        MERGE (owner)-[r:RESOURCE]->(s)
-        ON CREATE SET r.firstseen = timestamp()
-        SET r.lastupdated = $aws_update_tag
-    """
-    for secret in data:
-        secret["LastRotatedDate"] = dict_date_to_epoch(secret, "LastRotatedDate")
-        secret["LastChangedDate"] = dict_date_to_epoch(secret, "LastChangedDate")
-        secret["LastAccessedDate"] = dict_date_to_epoch(secret, "LastAccessedDate")
-        secret["DeletedDate"] = dict_date_to_epoch(secret, "DeletedDate")
-        secret["CreatedDate"] = dict_date_to_epoch(secret, "CreatedDate")
-
-    neo4j_session.run(
-        ingest_secrets,
-        Secrets=data,
+    """
+    Load transformed secrets into Neo4j using the data model.
+    Expects data to already be transformed by transform_secrets().
+    """
+    logger.info(f"Loading {len(data)} Secrets for region {region} into graph.")
+
+    # Load using the schema-based approach
+    load(
+        neo4j_session,
+        SecretsManagerSecretSchema(),
+        data,
+        lastupdated=aws_update_tag,
         Region=region,
-        AWS_ACCOUNT_ID=current_aws_account_id,
-        aws_update_tag=aws_update_tag,
+        AWS_ID=current_aws_account_id,
     )
 
 
 @timeit
 def cleanup_secrets(neo4j_session: neo4j.Session, common_job_parameters: Dict) -> None:
-    run_cleanup_job(
-        "aws_import_secrets_cleanup.json",
-        neo4j_session,
-        common_job_parameters,
+    """
+    Run Secrets cleanup job using the data model.
+    """
+    logger.debug("Running Secrets cleanup job.")
+    cleanup_job = GraphJob.from_node_schema(
+        SecretsManagerSecretSchema(), common_job_parameters
     )
+    cleanup_job.run(neo4j_session)
 
 
 @timeit
@@ -121,8 +137,6 @@ def get_secret_versions(
 
 def transform_secret_versions(
     versions: List[Dict],
-    region: str,
-    aws_account_id: str,
 ) -> List[Dict]:
     """
     Transform AWS Secrets Manager Secret Versions to match the data model.
@@ -203,7 +217,15 @@ def sync(
         )
         secrets = get_secret_list(boto3_session, region)
 
-        load_secrets(neo4j_session, secrets, region, current_aws_account_id, update_tag)
+        transformed_secrets = transform_secrets(secrets)
+
+        load_secrets(
+            neo4j_session,
+            transformed_secrets,
+            region,
+            current_aws_account_id,
+            update_tag,
+        )
 
         all_versions = []
         for secret in secrets:
@@ -216,11 +238,7 @@ def sync(
             )
             all_versions.extend(versions)
 
-        transformed_data = transform_secret_versions(
-            all_versions,
-            region,
-            current_aws_account_id,
-        )
+        transformed_data = transform_secret_versions(all_versions)
 
         load_secret_versions(
             neo4j_session,

cartography/intel/entra/groups.py

@@ -59,10 +59,29 @@ async def get_group_members(
     return user_ids, group_ids
 
 
+@timeit
+async def get_group_owners(client: GraphServiceClient, group_id: str) -> list[str]:
+    """Get owner user IDs for a given group."""
+    owner_ids: list[str] = []
+    request_builder = client.groups.by_group_id(group_id).owners
+    page = await request_builder.get()
+    while page:
+        if page.value:
+            for obj in page.value:
+                odata_type = getattr(obj, "odata_type", "")
+                if odata_type == "#microsoft.graph.user":
+                    owner_ids.append(obj.id)
+        if not page.odata_next_link:
+            break
+        page = await request_builder.with_url(page.odata_next_link).get()
+    return owner_ids
+
+
 def transform_groups(
     groups: list[Group],
     user_member_map: dict[str, list[str]],
     group_member_map: dict[str, list[str]],
+    group_owner_map: dict[str, list[str]],
 ) -> list[dict[str, Any]]:
     """Transform API responses into dictionaries for ingestion."""
     result: list[dict[str, Any]] = []
@@ -82,6 +101,7 @@ def transform_groups(
             "deleted_date_time": g.deleted_date_time,
             "member_ids": user_member_map.get(g.id, []),
             "member_group_ids": group_member_map.get(g.id, []),
+            "owner_ids": group_owner_map.get(g.id, []),
         }
         result.append(transformed)
     return result
@@ -134,6 +154,12 @@ async def sync_entra_groups(
 
     user_member_map: dict[str, list[str]] = {}
     group_member_map: dict[str, list[str]] = {}
+    group_owner_map: dict[str, list[str]] = {}
+
+    for group in groups:
+        owners = await get_group_owners(client, group.id)
+        group_owner_map[group.id] = owners
+
     for group in groups:
         try:
             users, subgroups = await get_group_members(client, group.id)
@@ -144,7 +170,9 @@ async def sync_entra_groups(
             user_member_map[group.id] = []
             group_member_map[group.id] = []
 
-    transformed_groups = transform_groups(groups, user_member_map, group_member_map)
+    transformed_groups = transform_groups(
+        groups, user_member_map, group_member_map, group_owner_map
+    )
 
     load_tenant(neo4j_session, {"id": tenant_id}, update_tag)
     load_groups(neo4j_session, transformed_groups, update_tag, tenant_id)

cartography/intel/gcp/__init__.py

@@ -391,6 +391,7 @@ def _sync_multiple_projects(
     # Compute data sync
     for project in projects:
         project_id = project["projectId"]
+        common_job_parameters["PROJECT_ID"] = project_id
         logger.info("Syncing GCP project %s for Compute.", project_id)
         _sync_single_project_compute(
             neo4j_session,
@@ -399,10 +400,12 @@ def _sync_multiple_projects(
             gcp_update_tag,
             common_job_parameters,
         )
+        del common_job_parameters["PROJECT_ID"]
 
     # Storage data sync
     for project in projects:
         project_id = project["projectId"]
+        common_job_parameters["PROJECT_ID"] = project_id
         logger.info("Syncing GCP project %s for Storage", project_id)
         _sync_single_project_storage(
             neo4j_session,
@@ -411,10 +414,12 @@ def _sync_multiple_projects(
             gcp_update_tag,
             common_job_parameters,
         )
+        del common_job_parameters["PROJECT_ID"]
 
     # GKE data sync
     for project in projects:
         project_id = project["projectId"]
+        common_job_parameters["PROJECT_ID"] = project_id
         logger.info("Syncing GCP project %s for GKE", project_id)
         _sync_single_project_gke(
             neo4j_session,
@@ -423,10 +428,12 @@ def _sync_multiple_projects(
             gcp_update_tag,
             common_job_parameters,
         )
+        del common_job_parameters["PROJECT_ID"]
 
     # DNS data sync
     for project in projects:
         project_id = project["projectId"]
+        common_job_parameters["PROJECT_ID"] = project_id
         logger.info("Syncing GCP project %s for DNS", project_id)
         _sync_single_project_dns(
             neo4j_session,
@@ -435,14 +442,17 @@ def _sync_multiple_projects(
             gcp_update_tag,
             common_job_parameters,
         )
+        del common_job_parameters["PROJECT_ID"]
 
     # IAM data sync
     for project in projects:
         project_id = project["projectId"]
+        common_job_parameters["PROJECT_ID"] = project_id
         logger.info("Syncing GCP project %s for IAM", project_id)
         _sync_single_project_iam(
             neo4j_session, resources, project_id, gcp_update_tag, common_job_parameters
         )
+        del common_job_parameters["PROJECT_ID"]
 
 
 @timeit