Make Cassandra replication factor configurable (issue #787) (#887)

Add CASSANDRA_REPLICATION_FACTOR environment variable and
--cassandra-replication-factor CLI argument to cassandra_config.py.

Update all four table store constructors (ConfigTableStore,
KnowledgeTableStore, LibraryTableStore, IamTableStore) to accept
an optional replication_factor parameter and use it in keyspace
creation CQL queries.

Thread the replication factor through all service constructors:
Configuration, KnowledgeManager, Librarian, IamService, and
knowledge store Processor.

trustgraph-flow/trustgraph/config/service/config.py

@@ -11,13 +11,14 @@ logger = logging.getLogger(__name__)
 
 class Configuration:
 
-    def __init__(self, push, host, username, password, keyspace):
+    def __init__(self, push, host, username, password, keyspace,
+                 replication_factor=1):
 
         # External function to respond to update
         self.push = push
 
         self.table_store = ConfigTableStore(
-            host, username, password, keyspace
+            host, username, password, keyspace, replication_factor
         )
 
     async def inc_version(self):

trustgraph-flow/trustgraph/cores/knowledge.py

@@ -17,11 +17,12 @@ class KnowledgeManager:
 
     def __init__(
             self, cassandra_host, cassandra_username, cassandra_password,
-            keyspace, flow_config,
+            keyspace, flow_config, replication_factor=1,
     ):
 
         self.table_store = KnowledgeTableStore(
-            cassandra_host, cassandra_username, cassandra_password, keyspace
+            cassandra_host, cassandra_username, cassandra_password, keyspace,
+            replication_factor
         )
 
         self.loader_queue = asyncio.Queue(maxsize=20)

trustgraph-flow/trustgraph/iam/service/iam.py

@@ -245,9 +245,11 @@ def _sign_jwt(kid, private_pem, claims):
 class IamService:
 
     def __init__(self, host, username, password, keyspace,
-                 bootstrap_mode, bootstrap_token=None):
+                 bootstrap_mode, bootstrap_token=None,
+                 replication_factor=1):
         self.table_store = IamTableStore(
             host, username, password, keyspace,
+            replication_factor,
         )
         # bootstrap_mode: "token" or "bootstrap".  In "token" mode the
         # service auto-seeds on first start using the provided

trustgraph-flow/trustgraph/librarian/librarian.py

@@ -28,6 +28,7 @@ class Librarian:
             bucket_name, keyspace, load_document,
             object_store_use_ssl=False, object_store_region=None,
             min_chunk_size=1,  # Default: no minimum (for Garage)
+            replication_factor=1,
     ):
 
         self.blob_store = BlobStore(
@@ -36,7 +37,8 @@ class Librarian:
         )
 
         self.table_store = LibraryTableStore(
-            cassandra_host, cassandra_username, cassandra_password, keyspace
+            cassandra_host, cassandra_username, cassandra_password, keyspace,
+            replication_factor
         )
 
         self.load_document = load_document

trustgraph-flow/trustgraph/storage/knowledge/store.py

@@ -23,7 +23,7 @@ class Processor(FlowProcessor):
         id = params.get("id")
 
         # Use helper to resolve configuration
-        hosts, username, password, keyspace = resolve_cassandra_config(
+        hosts, username, password, keyspace, replication_factor = resolve_cassandra_config(
             host=params.get("cassandra_host"),
             username=params.get("cassandra_username"),
             password=params.get("cassandra_password"),
@@ -59,6 +59,7 @@ class Processor(FlowProcessor):
             cassandra_username = username,
             cassandra_password = password,
             keyspace = keyspace,
+            replication_factor = replication_factor,
         )
 
     async def on_triples(self, msg, consumer, flow):

trustgraph-flow/trustgraph/tables/config.py

@@ -20,9 +20,11 @@ class ConfigTableStore:
     def __init__(
             self,
             cassandra_host, cassandra_username, cassandra_password, keyspace,
+            replication_factor=1,
     ):
 
         self.keyspace = keyspace
+        self.replication_factor = replication_factor
 
         logger.info("Connecting to Cassandra...")
 
@@ -57,12 +59,11 @@ class ConfigTableStore:
 
         logger.debug("Keyspace...")
         
-        # FIXME: Replication factor should be configurable
         self.cassandra.execute(f"""
             create keyspace if not exists {self.keyspace}
                 with replication = {{ 
                    'class' : 'SimpleStrategy', 
-                   'replication_factor' : 1 
+                   'replication_factor' : {self.replication_factor} 
                 }};
         """);
 

trustgraph-flow/trustgraph/tables/iam.py

@@ -28,8 +28,10 @@ class IamTableStore:
             self,
             cassandra_host, cassandra_username, cassandra_password,
             keyspace,
+            replication_factor=1,
     ):
         self.keyspace = keyspace
+        self.replication_factor = replication_factor
 
         logger.info("IAM: connecting to Cassandra...")
 
@@ -57,12 +59,11 @@ class IamTableStore:
         self._prepare_statements()
 
     def _ensure_schema(self):
-        # FIXME: Replication factor should be configurable.
         self.cassandra.execute(f"""
             create keyspace if not exists {self.keyspace}
                 with replication = {{
                     'class' : 'SimpleStrategy',
-                    'replication_factor' : 1
+                    'replication_factor' : {self.replication_factor}
                 }};
         """)
         self.cassandra.set_keyspace(self.keyspace)

trustgraph-flow/trustgraph/tables/knowledge.py

@@ -36,9 +36,11 @@ class KnowledgeTableStore:
     def __init__(
             self,
             cassandra_host, cassandra_username, cassandra_password, keyspace,
+            replication_factor=1,
     ):
 
         self.keyspace = keyspace
+        self.replication_factor = replication_factor
 
         logger.info("Connecting to Cassandra...")
 
@@ -73,12 +75,11 @@ class KnowledgeTableStore:
 
         logger.debug("Keyspace...")
         
-        # FIXME: Replication factor should be configurable
         self.cassandra.execute(f"""
             create keyspace if not exists {self.keyspace}
                 with replication = {{ 
                    'class' : 'SimpleStrategy', 
-                   'replication_factor' : 1 
+                   'replication_factor' : {self.replication_factor} 
                 }};
         """);
 

trustgraph-flow/trustgraph/tables/library.py

@@ -40,9 +40,11 @@ class LibraryTableStore:
     def __init__(
             self,
             cassandra_host, cassandra_username, cassandra_password, keyspace,
+            replication_factor=1,
     ):
 
         self.keyspace = keyspace
+        self.replication_factor = replication_factor
 
         logger.info("Connecting to Cassandra...")
 
@@ -77,12 +79,11 @@ class LibraryTableStore:
 
         logger.debug("Keyspace...")
 
-        # FIXME: Replication factor should be configurable
         self.cassandra.execute(f"""
             create keyspace if not exists {self.keyspace}
                 with replication = {{
                    'class' : 'SimpleStrategy',
-                   'replication_factor' : 1
+                   'replication_factor' : {self.replication_factor}
                 }};
         """);