chore: randomize the selection of nodes to balance the load

Author: fuziontech

housewatch/clickhouse/backups.py

@@ -30,8 +30,8 @@ def execute_backup(
 ):
     """
     This function will execute a backup on each shard in a cluster
-    This is very similar to run_query_on_shards but it has very specific things for backups
-    specifically around base_backup settings
+    This is very similar to run_query_on_shards but it has very specific params
+    for backups - specifically around base_backup settings
     """
     nodes = get_node_per_shard(cluster)
     responses = []

housewatch/clickhouse/clusters.py

@@ -1,3 +1,4 @@
+import random
 from collections import defaultdict
 from housewatch.clickhouse.client import run_query
 
@@ -38,11 +39,13 @@ def get_node_per_shard(cluster):
     preferred = PreferredReplica.objects.filter(cluster=cluster).values_list("replica", flat=True)
     for shard, n in shards.items():
         preferred_replica_found = False
+        # shuffle the nodes so we don't always pick the first preferred one
+        random.shuffle(n)
         for node in n:
             if node["host_name"] in preferred:
                 nodes.append((shard, node))
                 preferred_replica_found = True
                 break
         if not preferred_replica_found:
-            nodes.append((shard, n[0]))
+            nodes.append((shard, random.choice(n)))
     return nodes

housewatch/settings/__init__.py

@@ -26,7 +26,7 @@
 
 
 sentry_sdk.init(
-    dsn="https://6a05afd8bf4e2d54c81833ca1ff98cca@o607503.ingest.sentry.io/4505874503237633",
+    dsn="https://8874d21e05d62df688505df70c9f053d@o1015702.ingest.us.sentry.io/4507393944846336",
     integrations=[DjangoIntegration()],
     # If you wish to associate users to errors (assuming you are using
     # django.contrib.auth) you may enable sending PII data.