PG Dump API Reference
The PG Dump API is an admin-only endpoint that can be used to execute
pg_dump on the Postgres instance that Hasura is configured with.
The primary motive of this API is to provide convenience methods to initialise migrations from an existing Hasura instance. But the functionality can be later expanded to do other things such as taking a data dump etc.
All requests are
POST requests to the
POST /v1alpha1/pg_dump HTTP/1.1
"opts": ["-O", "-x", "--schema-only", "--schema", "public"],
opts: Arguments to be passed to the
pg_dumptool. Represented as array of strings. The underlying command that is executed is:
pg_dump $DATABASE_URL $OPTS -f $FILENAME
clean_output: When this optional argument is set to
true, the output SQL from the command is cleaned to remove the following:
- SQL front matter, like SET statements.
CREATE SCHEMA public.
COMMENT ON SCHEMA public is 'standard public schema';
- Comments (
--) and empty newlines.
- Postgres triggers created by Hasura for event triggers.
source: the name of the connected database on which to run
pg_dumpon. If skipped, it is set to
HTTP/1.1 200 OK
SET check_function_bodies = false;
CREATE TABLE public.author (
id integer NOT NULL,
name text NOT NULL
CREATE SEQUENCE public.author_id_seq
START WITH 1
INCREMENT BY 1
ALTER SEQUENCE public.author_id_seq OWNED BY public.author.id;
ALTER TABLE ONLY public.author ALTER COLUMN id SET DEFAULT nextval('public.author_id_seq'::regclass);
Disabling PG Dump API
Since this API can be used to dump all the Postgres data and schema, it can be disabled, especially in production deployments.
enabled-apis flag or the
HASURA_GRAPHQL_ENABLED_APIS env var can
be used to enable/disable this API. By default, The PG DumpAPI is
enabled. To disable it, you need to explicitly state that this API is
not enabled. i.e. remove it from the list of enabled APIs.
# enable only graphql & metadata apis, disable pgdump
See GraphQL Engine server config reference for info on setting the above flag/env var.