This feature is in preview. This feature is subject to change. To share feedback and/or issues, contact Support.
The CREATE VIRTUAL CLUSTER
statement creates a new virtual cluster. It is supported only starting a physical cluster replication (PCR) job.
PCR happens between an active primary cluster and a passive standby cluster that accepts updates from the primary cluster. The unit of replication is a virtual cluster, which is part of the underlying infrastructure in the primary and standby clusters. The CockroachDB cluster has:
- The system virtual cluster manages the cluster's control plane and the replication of the cluster's data. Admins connect to the system virtual cluster to configure and manage the underlying CockroachDB cluster, set up PCR, create and manage a virtual cluster, and observe metrics and logs for the CockroachDB cluster and each virtual cluster.
- Each other virtual cluster manages its own data plane. Users connect to a virtual cluster by default, rather than the system virtual cluster. To connect to the system virtual cluster, the connection string must be modified. Virtual clusters contain user data and run application workloads. When PCR is enabled, the non-system virtual cluster on both primary and secondary clusters is named
main
.
For more detail, refer to the Physical Cluster Replication Overview.
Physical cluster replication is only supported in CockroachDB self-hosted clusters.
Required privileges
CREATE VIRTUAL CLUSTER
requires one of the following privileges:
- The
admin
role. - The
MANAGEVIRTUALCLUSTER
system privilege allows the user to run all the relatedVIRTUAL CLUSTER
SQL statements for PCR.
Use the GRANT SYSTEM
statement:
GRANT SYSTEM MANAGEVIRTUALCLUSTER TO user;
Synopsis
Parameters
Parameter | Description |
---|---|
virtual_cluster_name |
The name for the new virtual cluster. |
LIKE virtual_cluster_spec |
Creates a virtual cluster with the same capabilities and settings as another virtual cluster. |
primary_virtual_cluster |
The name of the primary's virtual cluster to replicate. |
primary_connection_string |
The PostgreSQL connection string to the primary cluster. Refer to Connection string for more detail. |
replication_options_list |
Options to modify the PCR streams. Refer to Options. |
Options
Option | Description |
---|---|
New in v24.3: READ VIRTUAL CLUSTER |
(Preview) Create a read-only virtual cluster on the standby cluster, which allows reads of the standby's replicating virtual cluster. For an example, refer to Start a PCR stream with read from standby. |
RETENTION |
Configure a retention window that will control how far in the past you can fail over to. We do not recommend setting RETENTION much higher than the 24-hour default on the standby cluster. Accumulated data from an excessive retention (failover) window could affect queries running on the standby cluster that is active following a failover. |
Connection string
When you initiate a PCR stream from the standby cluster, it is necessary to pass a connection string to the system virtual cluster on the primary cluster:
'postgresql://{replication user}:{password}@{node IP or hostname}:26257?options=-ccluster=system&sslmode=verify-full&sslrootcert=certs/{primary cert}.crt'
To form a connection string similar to the example, include the following values and query parameters. Replace values in {...}
with the appropriate values for your configuration:
Value | Description |
---|---|
{replication user} |
The user on the primary cluster that has the REPLICATION system privilege. Refer to the Create a replication user and password for more detail. |
{password} |
The replication user's password. |
{node ID or hostname} |
The node IP address or hostname of any node from the primary cluster. |
options=ccluster=system |
The parameter to connect to the system virtual cluster on the primary cluster. |
sslmode=verify-full |
The verify-full secure connection type. |
sslrootcert={primary cert} |
The path to the primary cluster's CA certificate on the standby cluster. |
Capabilities
Cockroach Labs does not recommend changing the default capabilities of created virtual clusters.
Capabilities control what a virtual cluster can do. When you start a PCR stream, you can specify a virtual cluster with LIKE
to ensure other virtual clusters on the standby cluster will work in the same way. LIKE
will refer to a virtual cluster on the CockroachDB cluster you're running the statement from.
Examples
Start a PCR stream
To start a PCR stream to the standby of the primary's virtual cluster:
CREATE VIRTUAL CLUSTER main FROM REPLICATION OF main ON 'postgresql://{connection string to primary}';
This will create a main
virtual cluster in the standby cluster. The standby's system virtual cluster will connect to the primary cluster to initiate the PCR job. For details on the PCR stream, refer to the Responses for SHOW VIRTUAL CLUSTER
.
Specify a retention window for a PCR stream
When you initiate a PCR stream, you can specify a retention window to protect data from garbage collection. The retention window controls how far in the past you can fail over to:
CREATE VIRTUAL CLUSTER main FROM REPLICATION OF main ON 'postgresql://{connection string to primary}' WITH RETENTION '36h';
This will initiate a PCR stream from the primary cluster into the standby cluster's new main
virtual cluster. The RETENTION
option allows you to specify a timestamp in the past for failover to the standby cluster. After failover, the standby main
virtual cluster will be transactionally consistent to any timestamp within that retention window.
We do not recommend setting RETENTION
much higher than the 24-hour default on the standby cluster. Accumulated data from an excessive retention (failover) window could affect queries running on the standby cluster that is active following a failover.
Start a PCR stream with read from standby
While physical cluster replication (PCR) is generally available, using the READ VIRTUAL CLUSTER
option to read from a standby cluster is in preview.
New in v24.3:
Use the READ VIRTUAL CLUSTER
option to set up a PCR stream that also creates a read-only virtual cluster on the standby cluster. You can create a PCR job as per the Set Up Physical Cluster Replication guide and then add the option to the CREATE VIRTUAL CLUSTER
statement:
CREATE VIRTUAL CLUSTER main FROM REPLICATION OF main ON 'postgresql://{connection string to primary}' WITH READ VIRTUAL CLUSTER;
View the newly created virtual clusters:
SHOW VIRTUAL CLUSTERS;
You'll find:
- The
main
virtual cluster, which is accepting writes from the primary cluster. - The
main-readonly
virtual cluster, which is a read-only version of themain
virtual cluster.
id | name | data_state | service_mode
-----+---------------+-------------+---------------
1 | system | ready | shared
3 | main | replicating | none
4 | main-readonly | ready | shared
(3 rows)
To read table data from the standby cluster, connect to the readonly
virtual cluster:
cockroach sql --url `"postgresql://root@{node IP or hostname}:{26257}?options=-ccluster=main-readonly&sslmode=verify-full"` --certs-dir=certs
You can only read data on the created readonly
virtual cluster, other operations like SHOW VIRTUAL CLUSTERS
must be run from the system
virtual cluster. To connect to the readonly
virtual cluster, refer to the Connection Reference.
If you started the PCR stream with the READ VIRTUAL CLUSTER
option, failing over with SYSTEM TIME
will destroy the readonly
virtual cluster. If you fail over with LATEST
, the readonly
virtual cluster will remain on the original standby cluster, but will not update with new writes. Use DROP VIRTUAL CLUSTER
to remove the readonly
virtual cluster.
For details on adding a read-only virtual cluster after a failback, refer to the ALTER VIRTUAL CLUSTER
page.