Skip to content

Parallel distributed INSERT SELECT into Replicated tables. #47799

@alexey-milovidov

Description

@alexey-milovidov

Describe the situation

If a cluster contains one shard, all of the replicas - allow parallel distributed INSERT SELECT from a replicated table into a replicated table.

How to reproduce

ip-172-31-3-224.eu-central-1.compute.internal :) INSERT INTO github_events_smt SELECT * FROM github_events_web;

INSERT INTO github_events_smt SELECT *
FROM github_events_web

Query id: f08b29a1-5933-4e6e-a098-8508ee8bde5d


0 rows in set. Elapsed: 0.009 sec. 

Received exception from server (version 23.3.1):
Code: 36. DB::Exception: Received from localhost:9000. DB::Exception: Parallel distributed INSERT SELECT is not possible. Reason: distributed reading into Replicated table is supported only from *Cluster table functions, but got MergeTree storage. (BAD_ARGUMENTS)

ip-172-31-3-224.eu-central-1.compute.internal :) INSERT INTO FUNCTION clusterAllReplicas(default, default.github_events_smt) SELECT * FROM github_events_web;

INSERT INTO FUNCTION clusterAllReplicas(default, default.github_events_smt) SELECT *
FROM github_events_web

Query id: 90cdddab-9a5d-4eff-84c5-a2b31c0b75b3


0 rows in set. Elapsed: 0.008 sec. 

Received exception from server (version 23.3.1):
Code: 36. DB::Exception: Received from localhost:9000. DB::Exception: Parallel distributed INSERT SELECT is not possible. Reason: distributed reading is supported only from Distributed engine or *Cluster table functions, but got MergeTree storage. (BAD_ARGUMENTS)

ip-172-31-3-224.eu-central-1.compute.internal :)

Metadata

Metadata

Assignees

Labels

comp-parallel-replicasParallel reading from replicas for query speed (not data replication consistency).performance

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions