Update database cluster attributes

You can change the following database cluster attributes with the GDC console or the gdcloud CLI:

• Database password for the admin user
• External connections (enabled/disabled)
• Availability level
• Backup enabled and backup retention days
• Database flags
• CPU, memory, or storage allocated to the database cluster

For information on how to modify an attribute, see the workflow corresponding to the attribute type you want to update:

Console

High Availability

Enable or disable same zone high availability. See the Configure high availability section for more information.

Data Protection

For all database cluster types, you can enable or disable Data protection:

1. Go to Data protection and click edit Edit to access the data protection settings form. If automated backups are enabled, you can also configure how long the backups are to be retained.
2. After making changes, click Save to have the updates applied to your database cluster.

Connectivity

For all database cluster types, you can choose whether the database cluster can be accessed only from inside the GDC project, or if it is accessible from outside the project as well:

1. Go to Connectivity and click edit Edit.
2. Update your connectivity configurations.

3. Click Save to apply your changes.

   Important: Updating your connectivity configurations will also update the connectivity endpoint and connection string fields. Make sure to use the latest endpoint and connection string.

Instances

For all database cluster types, you can edit the primary instance properties:

1. Go to Primary Instance and click edit Edit.

2. Choose between High Performance, Standard, Minimum, or Custom configurations. The Custom option lets you choose the number of CPUs, memory, and storage for your primary instance.

   Note: You can increase or decrease the amount of CPU and memory, but you won't be able to select storage options smaller than your current storage size.

3. Updating your primary instance's CPU, memory, or storage will require a restart. Select Save and Restart to apply your changes.

For all database cluster types, this form also lets you add, modify, or remove database flags. The set of flags available is predetermined by GDC. See the Configure database flags section for more information.

gdcloud CLI

Use the following command to update a database cluster: sh gdcloud database clusters update CLUSTER_NAME [options]

Replace the following:

• CLUSTER_NAME with the name of the database cluster to update.

For the full list of options, see the command reference or run sh gdcloud database clusters update --help

API

Data Protection

For all database cluster types, you can enable data protection feature by creating a BackupPlan resource for that database cluster.

apiVersion: postgresql.dbadmin.gdc.goog/v1
kind: BackupPlan
metadata:
  name: DBCLUSTER_NAME-backupplan
  namespace: USER_PROJECT
spec:
  dbclusterRef: DBCLUSTER_NAME
  backupRetainDays: RETENTION_DAYS

Replace the following variables:

• DBCLUSTER_NAME: the name of the database cluster to enable data protection for.
• USER_PROJECT: the name of the user project where the database cluster will be created.
• RETENTION_DAYS: the number of days for which the operator retains this backup. It must be an integer between 1 and 90.

You can also configure data protection settings like retention days by editing the corresponding BackupPlan resource.

To turn off data protection for a database cluster, delete the corresponding BackupPlan resource previously created.

Reset Administrator User Password

For all database cluster types, you can reset the database password for the administrator user.

1. Create a password secret.

   apiVersion: v1
   kind: Secret
   metadata:
     name: db-pw-DBCLUSTER_NAME
     namespace: USER_PROJECT
   type: Opaque
   data:
     DBCLUSTER_NAME: "BASE64_PASSWORD"
   

2. Patch the DBCluster.

   kubectl patch dbcluster.postgresql.dbadmin.gdc.goog DBCLUSTER_NAME -p '{"spec":{"primarySpec": {"adminUser": {"passwordRef": {"name": "db-pw-DBCLUSTER_NAME"}}}}}' --type=merge -n USER_PROJECT
   

Replace the following variables:

• DBCLUSTER_NAME: the name of the database cluster.
• USER_PROJECT: the name of the user project where the database cluster will be updated.
• BASE64_PASSWORD: the base64 encoding of the database's administrator password.

It takes about a minute for the new password to take effect.

Configure database flags

Database images that ship with GDC come with default settings. However, you can customize the default database engine to satisfy requirements for your applications. Database clusters provide predefined flags that you can set using the GDC console or gdcloud CLI:

Console

1. In the navigation menu, select Database Service.

2. From the database cluster list, click the database cluster you want to configure database flags for.

3. In the Instances in your database cluster section, click Edit Primary.

   

4. In the Flags section, click Add a Database Flag.

5. Choose your flag and enter the value. If the value you input is not valid, the GDC console gives you instructions on how to satisfy the required range or value type.

6. Click Done.

7. To set the configuration, click Save. For some flags to take effect, you must restart the database cluster. For this case, click Save and Restart.

8. To confirm your new flag is set, return to the Instances in your database cluster section of your database cluster and verify the new flag and value are visible.

To edit your database flags, return to the Flags section and modify the existing flags. Hover over a flag component and click the Delete icon to remove a database flag.

gdcloud

You can configure new database flags for your cluster or reset all existing flags back to their default values:

• To configure a database flag for your database cluster, run:

  gdcloud database clusters update CLUSTER_NAME \
      --database-flags DB_FLAGS
  

  Replace the following:

  • CLUSTER_NAME: the name of the database cluster.
  • DB_FLAGS: the comma-separated list of database flags to set on the database running in the database cluster. Each database flag and value are set as a key-value pair. Flags without a value can be defined without a string following the = character.
  Important: When updating an existing database cluster with new flags, you must include all the previously set flags in addition to the new flags you want to configure. Failure to include your existing flags to the database flag update command sets them back to the default values.

  For example, the following command sets several database flags for the test-db-cluster:

  gdcloud database clusters update test-db-cluster \
      --database-flags max_allowed_packet=55555,skip_grant_tables=,log_output=1
  

  For a list of available database flags, see Available database flags.

• To reset all your database flags to their default settings, run:

  gdcloud database clusters update CLUSTER_NAME --clear-database-flags
  

  Replace CLUSTER_NAME with the name of the database cluster.

API

• To configure database flags for your database cluster, run:

    kubectl patch dbcluster.postgresql.dbadmin.gdc.goog DBCLUSTER_NAME -p '{"spec":{"primarySpec": {"parameters": DB_FLAGS}}}' --type=merge -n USER_PROJECT
  

  Replace the following variables:

  • DBCLUSTER_NAME: the name of the database cluster.
  • USER_PROJECT: the name of the user project where the database cluster was created.
  • DB_FLAGS: the database configuration parameters to be added or updated. This should be JSON-formatted object of key-value pairs, where keys are the database parameter names and values are their respective settings. For example, the following sets several database flags:

  "autovacuum": "on", "autovacuum_analyze_threshold": "2147483647", "dbs.enable_anon": "on"
  

  For a list of available database flags, see Available database flags.

• To reset all your database flags to their default settings, run:

  kubectl patch dbcluster.postgresql.dbadmin.gdc.goog DBCLUSTER_NAME --type=json -p='[{"op": "remove", "path": "/spec/primarySpec/parameters"}]' -n USER_PROJECT
  

  Replace the following variables:

  • DBCLUSTER_NAME: the name of the database cluster.
  • USER_PROJECT: the name of the user project where the database cluster was created.

Available database flags

Flag	Range	Type	Need restart
google_columnar_engine.enabled	on, off	Boolean	TRUE
google_columnar_engine.memory_size_in_mb	[128, 2147483647]	Integer	TRUE
anon.algorithm		String	FALSE
anon.maskschema		String	FALSE
anon.restrict_to_trusted_schemas	on, off	Boolean	FALSE
anon.salt		String	FALSE
anon.sourceschema		String	FALSE
dbs.enable_anon	on, off Note:This flag is specific to Database Service.	Boolean	TRUE
dbs.enable_pg_bigm	on, off Note:This flag is specific to Database Service.	Boolean	TRUE
dbs.enable_pg_cron	on, off Note:This flag is specific to Database Service.	Boolean	TRUE
dbs.enable_pg_hint_plan	on, off Note:This flag is specific to Database Service.	Boolean	TRUE
dbs.enable_pg_squeeze	on, off Note:This flag is specific to Database Service.	Boolean	TRUE
dbs.enable_pg_wait_sampling	on, off Note:This flag is specific to Database Service.	Boolean	TRUE
pg_bigm.enable_recheck	on, off	Boolean	FALSE
pg_bigm.gin_key_limit	[0, 2147483647]	Integer	FALSE
pg_bigm.similarity_limit	[0, 1]	Float	FALSE
pg_hint_plan.debug_print	off, on, detailed, verbose, 0, 1, 2, 3, no, yes, false, true	Enum	TRUE
pg_hint_plan.enable_hint	on, off	Boolean	FALSE
pg_hint_plan.enable_hint_table	on, off	Boolean	FALSE
pg_hint_plan.message_level	debug5, debug4, debug3, debug2, debug1, debug, info, notice, warning, error, log	Enum	FALSE
pg_hint_plan.parse_messages	debug5, debug4, debug3, debug2, debug1, debug, info, notice, warning, error, log	Enum	FALSE
pg_stat_statements.max	[100, 2147483647]	Integer	TRUE
pg_stat_statements.save	on, off	Boolean	FALSE
pg_stat_statements.track	none, top, all	Enum	FALSE
pg_stat_statements.track_utility	on, off	Boolean	FALSE
pg_wait_sampling.history_period	[1, 2147483647]	Integer	FALSE
pg_wait_sampling.history_size	[1, 2147483647]	Integer	FALSE
pg_wait_sampling.profile_period	[1, 2147483647]	Integer	FALSE
pg_wait_sampling.profile_pid	on, off	Boolean	FALSE
pg_wait_sampling.profile_queries	on, off	Boolean	FALSE
pgaudit.log	read, write, function, role, ddl, misc, misc_set, all, none , -read, -write, -function, -role, -ddl, -misc, -misc_set, -all, -none	Repeated string	FALSE
pgaudit.log_client	on, off	Boolean	FALSE
pgaudit.log_parameter	on, off	Boolean	FALSE
pgaudit.log_level	debug5, debug4, debug3, debug2, debug1, info, notice, warning, error, log	Enum	FALSE
pgaudit.log_relation	on, off	Boolean	FALSE
pgaudit.log_catalog	on, off	Boolean	FALSE
pgaudit.role		String	FALSE
pgaudit.log_statement_once	on, off	Boolean	FALSE
pglogical.batch_inserts	on, off	Boolean	TRUE
pglogical.conflict_log_level	debug5, debug4, debug3 , debug2, debug1 , info, notice , warning, error , log, fatal , panic	Enum	FALSE
pglogical.conflict_resolution	error, apply_remote, keep_local, last_update_wins, first_update_wins	Enum	FALSE
pglogical.extra_connection_options		String	FALSE
pglogical.synchronous_commit	on, off	Boolean	TRUE
pglogical.use_spi	on, off	Boolean	TRUE
pgtt.enabled	on, off	Boolean	FALSE
plv8.execution_timeout	[1, 65536]	Integer	FALSE
plv8.memory_limit	[256, 3096]	Integer	FALSE
plv8.start_proc		String	FALSE