Configure database flags

This page describes how to configure database flags for Cloud SQL, and lists the flags that you can set for your instance. You use database flags for many operations, including adjusting PostgreSQL parameters, adjusting options, and configuring and tuning an instance.

In some cases, setting one flag may require that you set another flag to fully enable the desired functionality.

When you set, remove, or modify a flag for a database instance, the database might be restarted. The flag value is then persisted for the instance until you remove it. If the instance is the source of a replica, and the instance is restarted, the replica is also restarted to align with the current configuration of the instance.

Configure database flags

Set a database flag

Console

  1. In the Google Cloud console, select the project that contains the Cloud SQL instance for which you want to set a database flag.
  2. Open the instance and click Edit.
  3. Scroll down to the Flags section.
  4. To set a flag that has not been set on the instance before, click Add item, choose the flag from the drop-down menu, and set its value.
  5. Click Save to save your changes.
  6. Confirm your changes under Flags on the Overview page.

gcloud

Edit the instance:

gcloud sql instances patch INSTANCE_NAME --database-flags=FLAG1=VALUE1,FLAG2=VALUE2

This command will overwrite all database flags previously set. To keep those and add new ones, include the values for all flags you want set on the instance; any flag not specifically included is set to its default value. For flags that do not take a value, specify the flag name followed by an equals sign ("=").

For example, to set the log_connections and log_min_error_statement flags, you can use the following command:

gcloud sql instances patch INSTANCE_NAME \
  --database-flags=log_connections=on,log_min_error_statement=error

Terraform

To add database flags, use a Terraform resource. 

resource "google_sql_database_instance" "instance" {
  name             = "postgres-instance"
  region           = "us-central1"
  database_version = "POSTGRES_14"
  settings {
    database_flags {
      name  = "log_connections"
      value = "on"
    }
    database_flags {
      name  = "log_min_error_statement"
      value = "error"
    }
    tier = "db-custom-2-7680"
  }
  # set `deletion_protection` to true, will ensure that one cannot accidentally delete this instance by
  # use of Terraform whereas `deletion_protection_enabled` flag protects this instance at the GCP level.
  deletion_protection = false
}

Apply the changes

To apply your Terraform configuration in a Google Cloud project, complete the steps in the following sections.

Prepare Cloud Shell

  1. Launch Cloud Shell.

  2. Set the default Google Cloud project where you want to apply your Terraform configurations.

    You only need to run this command once per project, and you can run it in any directory.

    export GOOGLE_CLOUD_PROJECT=PROJECT_ID

    Environment variables are overridden if you set explicit values in the Terraform configuration file.

Prepare the directory

Each Terraform configuration file must have its own directory (also called a root module).

  1. In Cloud Shell, create a directory and a new file within that directory. The filename must have the .tf extension—for example main.tf. In this tutorial, the file is referred to as main.tf. 
    mkdir DIRECTORY && cd DIRECTORY && touch main.tf

  2. If you are following a tutorial, you can copy the sample code in each section or step.

    Copy the sample code into the newly created main.tf.

    Optionally, copy the code from GitHub. This is recommended when the Terraform snippet is part of an end-to-end solution.

  3. Review and modify the sample parameters to apply to your environment.
  4. Save your changes.
  5. Initialize Terraform. You only need to do this once per directory. 
    terraform init

    Optionally, to use the latest Google provider version, include the -upgrade option: 

    terraform init -upgrade

Apply the changes

  1. Review the configuration and verify that the resources that Terraform is going to create or update match your expectations: 
    terraform plan

    Make corrections to the configuration as necessary.

  2. Apply the Terraform configuration by running the following command and entering yes at the prompt: 
    terraform apply

    Wait until Terraform displays the "Apply complete!" message.

  3. Open your Google Cloud project to view the results. In the Google Cloud console, navigate to your resources in the UI to make sure that Terraform has created or updated them.

Delete the changes

To delete your changes, do the following:

  1. To disable deletion protection, in your Terraform configuration file set the deletion_protection argument to false. 
    deletion_protection =  "false"
  2. Apply the updated Terraform configuration by running the following command and entering yes at the prompt: 
    terraform apply

  1. Remove resources previously applied with your Terraform configuration by running the following command and entering yes at the prompt:

    terraform destroy

REST v1

To set a flag for an existing database:

Before using any of the request data, make the following replacements:

  • project-id: The project ID
  • instance-id: The instance ID

HTTP method and URL: 

PATCH https://sqladmin.googleapis.com/v1/projects/project-id/instances/instance-id

Request JSON body: 

{
  "settings":
  {
    "databaseFlags":
    [
      {
        "name": "flag_name",
        "value": "flag_value"
      }
    ]
  }
}

To send your request, expand one of these options:

You should receive a JSON response similar to the following:

If there are existing flags configured for the database, modify the previous command to include them. The PATCH command overwrites the existing flags with the ones specified in the request.

REST v1beta4

To set a flag for an existing database:

Before using any of the request data, make the following replacements:

  • project-id: The project ID
  • instance-id: The instance ID

HTTP method and URL: 

PATCH https://sqladmin.googleapis.com/sql/v1beta4/projects/project-id/instances/instance-id

Request JSON body: 

{
  "settings":
  {
    "databaseFlags":
    [
      {
        "name": "flag_name",
        "value": "flag_value"
      }
    ]
  }
}

To send your request, expand one of these options:

You should receive a JSON response similar to the following:

If there are existing flags configured for the database, modify the previous command to include them. The PATCH command overwrites the existing flags with the ones specified in the request.

Clear all flags to their default values

Console

  1. In the Google Cloud console, select the project that contains the Cloud SQL instance for which you want to clear all flags.
  2. Open the instance and click Edit.
  3. Open the Database flags section.
  4. Click the X next to all of the flags shown.
  5. Click Save to save your changes.

gcloud

Clear all flags to their default values on an instance:

gcloud sql instances patch INSTANCE_NAME \
--clear-database-flags

You are prompted to confirm that the instance will be restarted.

REST v1

To clear all flags for an existing instance:

Before using any of the request data, make the following replacements:

  • project-id: The project ID
  • instance-id: The instance ID

HTTP method and URL: 

PATCH https://sqladmin.googleapis.com/v1/projects/project-id/instances/instance-id

Request JSON body: 

{
  "settings":
  {
    "databaseFlags": []
  }
}

To send your request, expand one of these options:

You should receive a JSON response similar to the following:

REST v1beta4

To clear all flags for an existing instance:

Before using any of the request data, make the following replacements:

  • project-id: The project ID
  • instance-id: The instance ID

HTTP method and URL: 

PATCH https://sqladmin.googleapis.com/sql/v1beta4/projects/project-id/instances/instance-id

Request JSON body: 

{
  "settings":
  {
    "databaseFlags": []
  }
}

To send your request, expand one of these options:

You should receive a JSON response similar to the following:

View current values of database flags

To view all current values of the PostgreSQL settings, log into your instance with the psql client and enter the following statement:

 SELECT name, setting FROM pg_settings;

Note that you can change the value only for supported flags (as listed below).

Determine which database flags have been set for an instance

To see which flags have been set for a Cloud SQL instance:

Console

  1. In the Google Cloud console, select the project that contains the Cloud SQL instance for which you want to see the database flags that have been set.
  2. Select the instance to open its Instance Overview page.

    The database flags that have been set are listed under the Database flags section.

gcloud

Get the instance state:

gcloud sql instances describe INSTANCE_NAME

In the output, database flags are listed under the settings as the collection databaseFlags. For more information about the representation of the flags in the output, see Instances Resource Representation.

REST v1

To list flags configured for an instance:

Before using any of the request data, make the following replacements:

  • project-id: The project ID
  • instance-id: The instance ID

HTTP method and URL: 

GET https://sqladmin.googleapis.com/v1/projects/project-id/instances/instance-id

To send your request, expand one of these options:

You should receive a JSON response similar to the following:

In the output, look for the databaseFlags field.

REST v1beta4

To list flags configured for an instance:

Before using any of the request data, make the following replacements:

  • project-id: The project ID
  • instance-id: The instance ID

HTTP method and URL: 

GET https://sqladmin.googleapis.com/sql/v1beta4/projects/project-id/instances/instance-id

To send your request, expand one of these options:

You should receive a JSON response similar to the following:

In the output, look for the databaseFlags field.

Supported flags

Flags not mentioned below are not supported.

For a given flag, Cloud SQL might support a different value or range from the corresponding PostgreSQL parameter or option.

A | C | D | E | F | G | H | I | J | L | M | O | P | R | S | T | V | W

Cloud SQL Flag Type
Acceptable Values and Notes 		Restart
Required?
anon.algorithm String
The default is sha256.		 No
anon.maskschema String
The default is mask.		 No
anon.restrict_to_trusted_schemas boolean
on | off
The default is off. 		No
anon.salt String
There's no default value.		 No
anon.sourceshema String
The default is public.		 No
auto_explain.log_analyze boolean
on | off
The default is off. 		No
auto_explain.log_buffers boolean
on | off
The default is off. 		No
auto_explain.log_min_duration Integer
-1 ... 2147483647
The default is -1. 		No
auto_explain.log_format String
text|xml|json|yaml
The default is text. 		No
auto_explain.log_level String
debug5|debug4|debug3|debug2|debug1|debug|
info|notice|warning|log
The default is log.
Supported in PostgreSQL 12 and above. 		No
auto_explain.log_nested_statements boolean
on | off
The default is off. 		No
auto_explain.log_settings boolean
on | off
The default is off.
Supported in PostgreSQL 12 and above. 		No
auto_explain.log_timing boolean
on | off
The default is on. 		No
auto_explain.log_triggers boolean
on | off
The default is off. 		No
auto_explain.log_wal boolean
on | off
The default is off.
Supported in PostgreSQL 13 and above. 		No
auto_explain.log_verbose boolean
on | off
The default is off. 		No
auto_explain.sample_rate Float
0 ... 1
The default is 1. 		No
autovacuum boolean
on | off
The default is on.
For usage information that might affect the SLA, see the Operational guidelines for Cloud SQL for PostgreSQL instances. 		No
autovacuum_analyze_scale_factor float
0 ... 100
The default is 0.1. 		No
autovacuum_analyze_threshold integer
0 ... 2147483647
The default is 50.		 No
autovacuum_freeze_max_age integer
100000 ... 2000000000
The default is 200000000.
For usage information that might affect the SLA, see the Operational guidelines for Cloud SQL for PostgreSQL instances. 		Yes
autovacuum_max_workers integer
1 ... varies (see note)
The default is 3. 		Yes
autovacuum_multixact_freeze_max_age integer
10000 ... 2000000000
The default is 400000000.
For usage information that might affect the SLA, see the Operational guidelines for Cloud SQL for PostgreSQL instances. 		Yes
autovacuum_naptime integer
1 ... 2147483s
The default is 60s.		 No
autovacuum_vacuum_cost_delay integer
0 ... 100 ms, or -1 to use the vacuum_cost_delay value
For PostgreSQL 9.6, 10, and 11, the default value is 2 ms.		 No
autovacuum_vacuum_cost_limit integer
0 ... 10000, or -1 to use the vacuum_cost_limit value
The default is -1.		 No
autovacuum_vacuum_insert_scale_factor float
0 ... 100
The default is 0.2.		 No
autovacuum_vacuum_insert_threshold integer
-1 ... 2147483647
The default is 1000.		 No
autovacuum_vacuum_scale_factor float
0 ... 100
The default is 0.2.		 No
autovacuum_vacuum_threshold integer
0 ... 2147483647
The default is 50.		 No
autovacuum_work_mem integer
0 ... 2147483647 KB, or -1 to use the maintenance_work_mem value
The default is -1.		 No
checkpoint_completion_target float
0.0 ... 1.0
The default is 0.9.		 No
checkpoint_timeout integer
30 ... 86,400s
The default is 300s.
For usage information that might affect the SLA, see the Operational guidelines for Cloud SQL for PostgreSQL instances. 		No
checkpoint_warning integer
0 ... 2147483647s
The default is 30s.		 No
cloudsql.allow_passwordless_local_connections boolean
on | off
The default is off. 		No
cloudsql.enable_anon boolean
on | off
The default is off. 		No
cloudsql.enable_auto_explain boolean
on | off
The default is off. 		Yes
cloudsql.enable_maintenance_mode boolean
on | off
The default is off. 		Yes
cloudsql.enable_pgaudit boolean
on | off
The default is off. 		Yes
cloudsql.enable_pg_bigm boolean
on | off
default: off 		Yes
cloudsql.enable_pg_cron boolean
on | off
The default is off.
Supported in PostgreSQL 10 and above. 		Yes
cloudsql.enable_pg_hint_plan boolean
on | off
The default is off. 		Yes
cloudsql.enable_pglogical boolean
on | off
The default is off. 		Yes
cloudsql.enable_pg_squeeze boolean
on | off
The default is off. 		Yes
cloudsql.enable_pg_wait_sampling boolean
on | off 		Yes
cloudsql.iam_authentication boolean
on | off
The default is off. 		No
cloudsql.logical_decoding boolean
on | off
The default is off. 		Yes
cloudsql.max_failed_attempts_user integer
0 ... 10000
The default is 10.
Yes
cloudsql.pg_authid_select_role string
 No
commit_delay integer
0 ... 100000
The default is 0.		 No
commit_siblings integer
0 ... 1000
The default is 5.		 No
constraint_exclusion enumeration
partition | on | off
The default is partition.		 No
cpu_index_tuple_cost float
0.0 ... inf
The default is 0.005.		 No
cpu_operator_cost float
0.0 ... inf
The default is 0.0025.		 No
cpu_tuple_cost float
0.0 ... inf
The default is 0.01.		 No
cron.database_name String.
Supported in PostgreSQL 10 and above. 		Yes
cron.log_statement boolean
on | off
The default is on.
Supported in PostgreSQL 10 and above. 		Yes
cron.log_run boolean
on | off
The default is on.
Supported in PostgreSQL 10 and above. 		Yes
cron.max_running_jobs Integer
0 ... varies
The default is 5.
Supported in PostgreSQL 10 and above. 		Yes
cron.log_min_messages String
debug5|debug4|debug3|debug2|debug1|debug|
info|notice|warning|error|log|fatal|panic
The default is warning.
Supported in PostgreSQL 10 and above. 		No
cursor_tuple_fraction float
0.0 ... 1.0
The default is 0.1.		 No
deadlock_timeout integer
1 ... 2147483647 ms
The default is 1000ms.		 No
default_statistics_target integer
1 ... 10000
The default is 100.		 No
default_tablespace string No
default_transaction_deferrable boolean
on | off
The default is off.		 No
default_transaction_isolation enumeration
serializable | 'repeatable read' | 'read committed' | 'read uncommitted'
The default is 'read committed'.		 No
effective_cache_size integer
The size range is from 10% - 70% of the instance's memory.
Unit is 8 KB.
The default is 40% of VM memory. For example, for a 45GB instance memory, the default value is 18537160 KB.		 No
effective_io_concurrency integer
0 ... 1000
The default is 1.		 No
enable_bitmapscan boolean
on | off
The default is on. 		No
enable_hashagg boolean
on | off
The default is on.		 No
enable_hashjoin boolean
on | off
The default is on.		 No
enable_indexonlyscan boolean
on | off
The default is on.		 No
enable_indexscan boolean
on | off
The default is on.		 No
enable_material boolean
on | off
The default is on		 No
enable_mergejoin boolean
on | off
The default is on.		 No
enable_nestloop boolean
on | off
The default is on.		 No
enable_seqscan boolean
on | off
The default is on.		 No
enable_sort boolean
on | off
The default is on.		 No
enable_tidscan boolean
on | off
The default is on.		 No
force_parallel_mode enumeration
off | on | regress
The default is off.		 No
from_collapse_limit integer
1 ... 2147483647
The default is 8.		 No
geqo boolean
on | off
The default is on.		 No
geqo_effort integer
1 ... 10
The default is 5.		 No
geqo_generations integer
0 ... 2147483647
The default is 0.		 No
geqo_pool_size integer
0 ... 2147483647
The default is 0.		 No
geqo_seed float
0.0 ... 1.0
The default is 0.		 No
geqo_selection_bias float
1.5 ... 2.0
The default is 2.		 No
geqo_threshold integer
2 ... 2147483647
The default is 12.		 No
gin_fuzzy_search_limit integer
0 ... 2147483647
The default is 0.		 No
gin_pending_list_limit integer
64 ... 2147483647KB
The default is 4096KB.		 No
hash_mem_multiplier float
1 ... 1000
The default is 2.		 No
hot_standby_feedback boolean
on | off
The default is off.		 No
huge_pages enumeration
try | off
The default is try.		 Yes
idle_in_transaction_session_timeout integer
0 ... 2147483647 ms
The default is 0.		 No
ivfflat.probes integer
1 ... varies
The default is 1.
Supported in PostgreSQL 11 and above. 		No
join_collapse_limit integer
1 ... 2147483647
The default is 8.		 No
lock_timeout integer
0 ... 2147483647 ms
The default is 0.		 No
log_autovacuum_min_duration integer
0 ... 2147483647 ms, or -1 to disable
The default is 0.		 No
log_checkpoints boolean
on | off
The default is off.		 No
log_connections boolean
on | off
The default is off.		 No
log_disconnections boolean
on | off
The default is off.		 No
log_duration boolean
on | off
The default is off.		 No
log_error_verbosity enumeration
terse | default | verbose
The default is default.		 No
log_executor_stats boolean
on | off
The default is off.		 No
log_hostname boolean
on | off
The default is off.		 No
log_line_prefix String
A printf-style string that's generated at the beginning of each line of a log file.
The default is %m [%p]: [%l-1] db=%d,user=%u, which logs timestamp, process ID, database, and username.
 No
log_lock_waits boolean
on | off
The default is off.		 No
log_min_duration_statement integer
-1 ... 2147483647 ms
The default is -1.		 No
log_min_error_statement enumeration
debug5 | debug4 | debug3 | debug2 | debug1 | info | notice | warning | error | log | fatal | panic
The default is error.		 No
log_min_messages enumeration
debug5 | debug4 | debug3 | debug2 | debug1 | info | notice | warning | error | log | fatal | panic
The default is warning.		 No
log_parser_stats boolean
on | off
The default is off.		 No
log_planner_stats boolean
on | off
The default is off.		 No
log_replication_commands boolean
on | off
The default is off.		 No
log_statement enumeration
none | ddl | mod | all
Set to mod to log all Data definition language (DDL) statements, plus data-modifying statements such as INSERT, UPDATE, DELETE, TRUNCATE
The default is none.		 No
log_statement_stats boolean
on | off
May not be enabled together with log_parser_stats, log_planner_stats, or log_executor_stats.
The default is off.		 No
log_temp_files integer
0 ... 2147483647 KB, or -1 to disable
The default is 0.		 No
log_timezone string
This flag provides Cloud SQL for PostgreSQL users with a way to set the time zone used for timestamps that are written in the server log.

You specify time zones by name. For example, Europe/London is the timezone name for London.

You must update this flag manually, on the primary instance and on all read replicas, to account for it.

Timezone names are case insensitive. You can supply the timezone name in any case.

We support UTC+X as a valid format for this flag, where X is +/-HH:MM.

 Yes
logical_decoding_work_mem integer
64 ... 2147483647
The default is 65536.		 No
maintenance_io_concurrency integer
0 ... 1000
The default is 10.		 No
maintenance_work_mem integer
1024 ... 2147483647 KB
The default is 64 MB.		 No
max_connections integer
14 ... varies (see note)
The default value depends on the amount of memory of the largest instance in the chain of primaries (this instance; and, if it's a read replica, its primary, its primary's primary, etc. all the way up to the root of the replication tree).
Memory (GB) on largest instanceDefault value
tiny (~0.5)25
small (~1.7)50
from 3.75 to < 6100
from 6 to < 7.5200
from 7.5 to < 15400
from 15 to < 30500
from 30 to < 60600
from 60 to < 120800
>=1201,000

The value on replicas must be greater than or equal to the value on the primary. Changes on the primary propagate to replicas that have a value that is lower than the new value on the primary, or that have not been changed from the default value. Such changes on the primary cause the replica to restart.

Yes
max_locks_per_transaction integer
10 ... 2,147,483,647
The default is 64.

The value on replicas must be greater than or equal to the value on the primary. Changes on the primary propagate to replicas that have a value that is lower than the new value on the primary, or that have not been changed from the default value. Such changes on the primary cause the replica to restart.

Yes
max_logical_replication_workers integer
4 ... 8192
The default is 4.
Supported in PostgreSQL 10 and above.		 Yes
max_parallel_maintenance_workers integer
0 ... varies
The default is 2.
Supported in PostgreSQL 11 and above.

The value on replicas must be greater than or equal to the value on the primary. Changes on the primary propagate to replicas that have a value that's lower than the new value on the primary, or that haven't been changed from the default value.

If the value on the primary is default, then the value for the replicas can't be changed. To change the value for the replicas, first, set the value on the primary to an integer.

 No
max_parallel_workers integer
0 ... varies
The default is 8.
Supported in PostgreSQL 10 and above.

The value on replicas must be greater than or equal to the value on the primary. Changes on the primary propagate to replicas that have a value that's lower than the new value on the primary, or that haven't been changed from the default value.

If the value on the primary is default, then the value for the replicas can't be changed. To change the value for the replicas, first, set the value on the primary to an integer.

 No
max_parallel_workers_per_gather integer
0 ... varies
The default is 2.

The value on replicas must be greater than or equal to the value on the primary. Changes on the primary propagate to replicas that have a value that's lower than the new value on the primary, or that haven't been changed from the default value.

If the value on the primary is default, then the value for the replicas can't be changed. To change the value for the replicas, first, set the value on the primary to an integer.

 No
max_pred_locks_per_page integer
0 ... 2147483647
The default is 2.
Supported in PostgreSQL 10 and above.		 Yes
max_pred_locks_per_relation integer
-2147483648 ... 2147483647
The default is -2.
Supported in PostgreSQL 10 and above.		 Yes
max_pred_locks_per_transaction integer
64 ... 1048576 		Yes
max_prepared_transactions integer
0 ... varies
Memory size (MB)Max value
0 - 3,84030,000
3,840 - 7,68085,000
7,680 - 15,360200,000
15,360 and above262,000

The value on replicas must be greater than or equal to the value on the primary. Changes on the primary propagate to replicas that have a value that is lower than the new value on the primary, or that have not been changed from the default value. Such changes on the primary cause the replica to restart.

Yes
max_replication_slots integer
10 ... varies
The default is 10. 		Yes
max_standby_archive_delay integer
0 ... 2147483647 ms, or -1 to wait forever 		No
max_standby_streaming_delay integer
0 ... 2147483647 ms, or -1 to wait forever		 No
max_sync_workers_per_subscription integer
2 ... 64
Cannot be greater than max_logical_replication_workers. 		No
max_wal_senders integer
10 ... varies
The default is 10.

The value on replicas must be greater than or equal to the value on the primary. Changes on the primary propagate to replicas that have a value that is lower than the new value on the primary, or that have not been changed from the default value. Such changes on the primary cause the replica to restart.

Yes
max_wal_size integer
2 ... 2147483647

The unit is 16 MB (the WAL file size) for version 9.6 and 1 MB for PostgreSQL 10 and above.

The default value is 1504 MB if instance memory more than or equal to 3.75 GB. The default value is 1 GB if instance memory is less than 3.75 GB.

For usage information that might affect the SLA, see the Operational guidelines for Cloud SQL for PostgreSQL instances. 		No
max_worker_processes integer
8 ... varies

The value on replicas must be greater than or equal to the value on the primary. Changes on the primary propagate to replicas that have a value that is lower than the new value on the primary, or that have not been changed from the default value. Such changes on the primary cause the replica to restart.

Yes
min_parallel_relation_size integer
0 ... 715827882
Unit is 8 KB
Supported only in PostgreSQL 9.6.		 No
min_wal_size integer
32 ... 2147483647

The unit is 16 MB (the WAL file size) for version 9.6 and 1 MB for PostgreSQL 10 and above.

No
old_snapshot_threshold integer
0 ... 86400 min, or -1 to disable
The default is -1.		 Yes
parallel_setup_cost float
0.0 ... inf
The default is 1000.		 No
parallel_tuple_cost float
0.0 ... inf
The default is 0.1.		 No
password_encryption enumeration
md5 | scram-sha-256

The default depends on the PostgreSQL version. For PostgreSQL versions 10 to 13, the default is md5. For PostgreSQL 14 and higher, the default is scram-sha-256.

 No
pg_bigm.enable_recheck boolean
on | off 		No
pg_bigm.gin_key_limit integer
0 ... 2147483647		 No
pg_bigm.similarity_limit float
0.0 ... 1.0		 No
pg_hint_plan.enable_hint boolean
on | off
The default is on. 		No
pg_hint_plan.debug_print String
off|on|detailed|verbose|0|1||2|3|no|yes|false|true
The default is off. 		No
pg_hint_plan.parse_messages String
debug5|debug4|debug3|debug2|debug1|debug|
info|notice|warning|error|log
The default is info. 		No
pg_hint_plan.message_level String
debug5|debug4|debug3|debug2|debug1|debug|
info|notice|warning|error|log
The default is log. 		No
pg_hint_plan.enable_hint_table boolean
on | off
The default is off. 		No
pglogical.batch_inserts boolean
on | off
The default is off. 		Yes
pglogical.conflict_log_level String
The default is LOG.
This flag accepts the same values as log_min_messages. 		No
pglogical.conflict_resolution String
error|apply_remote|keep_local|last_update_wins|first_update_wins The default is apply_remote. 		No
pglogical.extra_connection_options String
Accepts PostgreSQL keyword/value connection strings.
The default is the empty string. 		No
pglogical.synchronous_commit boolean
on | off
The default is on. 		Yes
pglogical.use_spi boolean
on | off
The default is off. 		Yes
pg_stat_statements.max integer
100 ... 2147483647
The default is 5000.		 Yes
pg_stat_statements.save boolean
on | off
The default is on.		 No
pg_stat_statements.track enumeration
none | top | all
The default is top.		 No
pg_stat_statements.track_utility boolean
on | off
The default is on.		 No
pgaudit.log enumeration
read | write | function | role | ddl | misc | misc_set | all|none

You can provide multiple classes using a comma-separated list, and subtract a class by prefacing the class with a - sign. The default is none.

No
pgaudit.log_catalog boolean
on | off
The default is on. 		No
pgaudit.log_client boolean
on | off
The default is off. 		No
pgaudit.log_level enumeration
debug5 | debug4 | debug3 | debug2 | debug1 | info | notice | warning | error | log
The default is log. Additionally, pgaudit.log_level is enabled only when pgaudit.log_client is on. 		No
pgaudit.log_parameter boolean
on | off
The default is off. 		No
pgaudit.log_relation boolean
on | off
The default is off. 		No
pgaudit.log_statement_once boolean
on | off
The default is off. 		No
pgaudit.role string
There is no default. 		No
pgtt.enabled boolean
on | off
The default is on.		 No
pg_wait_sampling.history_period integer
1 ... 2147483647		 No
pg_wait_sampling.history_size integer
1 ... 2147483647		 No
pg_wait_sampling.profile_period integer
1 ... 2147483647		 No
pg_wait_sampling.profile_pid boolean
on | off		 No
pg_wait_sampling.profile_queries boolean
on | off		 No
random_page_cost float
0.0 ... inf
The default is 4.		 No
rdkit.agent_FP_bit_ratio float
0 ... 3		 No
rdkit.avalon_fp_size integer
64 ... 9192		 No
rdkit.dice_threshold float
0 ... 1		 No
rdkit.difference_FP_weight_agents integer
-10 ... 10		 No
rdkit.difference_FP_weight_nonagents integer
1 ... 20		 No
rdkit.do_chiral_sss boolean
on | off		 No
rdkit.do_enhanced_stereo_sss boolean
on | off		 No
rdkit.featmorgan_fp_size integer
64 ... 9192		 No
rdkit.hashed_atompair_fp_size integer
64 ... 9192		 No
rdkit.hashed_torsion_fp_size integer
64 ... 9192		 No
rdkit.ignore_reaction_agents boolean
on | off		 No
rdkit.init_reaction boolean
on | off		 No
rdkit.layered_fp_size integer
64 ... 9192		 No
rdkit.morgan_fp_size integer
64 ... 9192		 No
rdkit.move_unmmapped_reactants_to_agents boolean
on | off		 No
rdkit.rdkit_fp_size integer
64 ... 9192		 No
rdkit.reaction_difference_fp_size integer
64 ... 9192		 No
rdkit.reaction_difference_fp_type integer
1 ... 3		 No
rdkit.reaction_sss_fp_size integer
64 ... 9192		 No
rdkit.reaction_sss_fp_type integer
1 ... 5		 No
rdkit.sss_fp_size integer
64 ... 4096		 No
rdkit.tanimoto_threshold float
0 ... 1		 No
rdkit.threshold_unmapped_reactant_atoms float
0 ... 1		 No
replacement_sort_tuples integer
0 ... 2147483647 		No
session_replication_role enumeration
origin | replica | local
Can be set only in current session		 -
shared_buffers integer
The size range is from 10% - 60% of the instance's memory.
Unit is 8 KB.
The default is one-third of the instance's total VM memory (in MB). For example, for 45 GB of instance memory, the default value is 15085MB. 		Yes
squeeze.max_xlock_time integer
1 ... 2147483647		 No
squeeze.worker_autostart string
 No
squeeze.worker_role string
 No
ssl_max_protocol_version enumeration
Postgres 14: Sets the maximum SSL/TLS protocol version to use. Valid values are as for ssl_min_protocol_version, with the addition of an empty string, which allows any protocol version to be specified.
Supported in PostgreSQL 12 and above.		 No
ssl_min_protocol_version enumeration
Postgres 14: Sets the minimum SSL/TLS protocol version to use. Valid values are currently: TLSv1, TLSv1.1, TLSv1.2, TLSv1.3.
The default is TLSv1.
Supported in PostgreSQL 12 and above.		 No
standard_conforming_strings boolean
on | off
The default is on.		 No
synchronize_seqscans boolean
on | off
The default is on.		 No
tcp_keepalives_count integer
0 ... 2147483647
The default is 5.		 No
tcp_keepalives_idle integer
0 ... 2147483647
The default is 60.		 No
tcp_keepalives_interval integer
0 ... 2147483647
The default is 60.		 No
temp_buffers integer
100 ... 1,073,741,823
Unit is 8 KB		 No
temp_file_limit integer
1048576 ... 2147483647 KB
The default is 10% of the initial disk size. For example, for a 100-GB disk, the default value is 10262623KB.		 No
TimeZone string
This flag provides Cloud SQL for PostgreSQL users with a way to set the time zone for displaying and analyzing timestamps.

You specify time zones by name. For example, Europe/London is the timezone name for London.

You must update this flag manually, on the primary instance and on all read replicas, to account for it.

Timezone names are case insensitive. You can supply the timezone name in any case.

We support UTC+X as a valid format for this flag, where X is +/-HH.

 Yes
trace_notify boolean
on | off
The default is off.		 No
trace_recovery_messages enumeration
debug5 | debug4 | debug3 | debug2 | debug1 | log | notice | warning | error
The default is log.		 No
trace_sort boolean
on | off
The default is off.		 No
track_activities boolean
on | off
The default is on.		 No
track_activity_query_size integer
100 ... 102400
The default is 1 KB.		 Yes
track_commit_timestamp boolean
on | off
The default is off.		 Yes
track_counts boolean
on | off
The default is on.		 No
track_functions enumeration
none | pl | all
The default is none.		 No
track_io_timing boolean
on | off
The default is off.		 No
vacuum_cost_delay integer
0 ... 100 ms
The default is 0.		 No
vacuum_cost_limit integer
1 ... 10000
The default is 200.		 No
vacuum_failsafe_age integer
0 ... 2100000000
The default is 1600000000.		 No
vacuum_freeze_min_age integer
0 ... 1000000000
The default is 50000000.		 No
vacuum_freeze_table_age integer
0 ... 2000000000
The default is 150000000. 		No
vacuum_multixact_failsafe_age integer
0 ... 2100000000
The default is 1600000000.		 No
vacuum_multixact_freeze_min_age integer
0 ... 1000000000
The default is 5000000. 		No
vacuum_multixact_freeze_table_age integer
0 ... 2000000000
The default is 150000000.		 No
wal_buffers integer
The size range is from -1 - 5% of the instance's memory.
Unit is 8 KB.
The default is 16 MB. 		Yes
wal_compression enumeration
off | on | pglz | lz4 | zstd
The default is off.
No
wal_receiver_timeout integer
0 ... 2147483647
The default is 60 seconds.

This flag affects the WAL sender and receiver. If not appropriately set, this flag affects logical and physical replication. This flag also affects replication performance and latency. A value of zero disables the timeout mechanism. The unit is milliseconds.

No
wal_sender_timeout integer
0 ... 2147483647
The default is 60 seconds.

This flag affects the WAL sender and receiver. If not appropriately set, this flag affects logical and physical replication. This flag also affects replication performance and latency. A value of zero disables the timeout mechanism. The unit is milliseconds.

No
work_mem integer
64 ... 2147483647 KB
The default is 4 MB.		 No

Special flags

bgwriter

PostgreSQL has a background writer (bgwriter) flag. This flag issues writes of new or modified shared buffers. These shared buffers are known as dirty buffers. When the number of clean shared buffers is insufficient, the background writer writes dirty buffers to the file system and marks them as clean.

Two flags associated with the bgwriter flag are bgwriter_delay and bgwriter_lru_maxpages. bgwriter_delay specifies the delay between activity rounds for the background writer in milliseconds (ms), and bgwriter_lru_maxpages specifies how many buffers will be written by the background writer.

The default value for the bgwriter flag is 200 ms. However, if you select a solid state drive (SSD) that's greater than 500 GB, then the value of the bgwriter_delay flag is set to 50, and the value of the bgwriter_lru_maxpages flag is set to 200.

For more information about the background writer, see the PostgreSQL documentation.

session_replication_role

PostgreSQL has the session_replication_role flag, which is designed to be used for logical replication and lets you disable constraint triggers in individual sessions.

Sometimes this flag can also be used for some maintenance operations to circumvent constraint (most often Foreign Key) checks.

This flag can be set in a session by any user which has the REPLICATION property set. The REPLICATION property for any user can be set by cloudsqlsuperuser when one of flags cloudsql.enable_pglogical or cloudsql.logical_decoding is set for the instance.

This flag can not be set for the whole instance.

Troubleshooting

Issue Troubleshooting
You set the time zone for a session, but it expires when you log off.

Connect to the database and set the database time zone to the one you want, either per user or per database.

In Cloud SQL for PostgreSQL, you can specify the following. These settings remain after a session is closed, mimicking a .conf configuration:

ALTER DATABASE dbname SET TIMEZONE TO 'timezone';
ALTER USER username SET TIMEZONE TO 'timezone';

These settings apply only to new connections to the database. To see the change to the time zone, disconnect from the instance and then reconnect to it.

What's next