Repo with the issue and instructions on how to reproduce https://github.com/poul-kg/hasura-issue-4296

Prerequisites:

• hasura cli 1.1.0

• also tested with v1.2.0-beta.3

• git clone git@github.com:poul-kg/hasura-issue-4296.git

• cd hasura-issue-4296

• docker-compose up -d --build

• hasura migrate apply --admin-secret 12345 --endpoint http://localhost:8080

• hasura console --admin-secret 12345 --endpoint http://localhost:8080

Check event's retry configuration

• open http://localhost:9695/events/manage/triggers/test_event_1/modify , check Retry configuration, it's 30/30/60
• open http://localhost:9695/events/manage/triggers/test_event_2/modify , check Retry configuration, it's 30/30/60

now try to squash
Ctrl+C to stop hasura console

hasura migrate squash --admin-secret 12345 --endpoint http://localhost:8080 --from 1585992945707 --name squashed

select Yes when asked to remove migrations.

run ./remigrate.sh, which will stop docker, remove db volume, start docker, run hasura migrations

then run console

hasura console --admin-secret 12345 --endpoint http://localhost:8080

Check event's retry configuration

• open http://localhost:9695/events/manage/triggers/test_event_1/modify , check Retry configuration, it's 0/10/60
• open http://localhost:9695/events/manage/triggers/test_event_2/modify , check Retry configuration, it's 10/10/60
• open squashed up.yml file:

- args:
    headers: []
    insert:
      columns: '*'
    name: test_event_1
    replace: true
    table:
      name: test_table_1
      schema: public
    webhook: http://localhost:8080/api/test_event_1
  type: create_event_trigger
- args:
    headers: []
    insert:
      columns: '*'
    name: test_event_2
    replace: true
    table:
      name: test_table_2
      schema: public
    webhook: http://localhost/test_event_2
  type: create_event_trigger

there is no information about retry configuration here

Expected behavior

after running squashed migration, I want Retry configuration be preserved

• when I open http://localhost:9695/events/manage/triggers/test_event_1/modify , and check Retry configuration, it should still be 30/30/60
• when I open http://localhost:9695/events/manage/triggers/test_event_2/modify , and check Retry configuration, it should still be 30/30/60

YAML file of squashed migration should be (I'm not sure about format, it's my guess):

I've checked format below, it doesn't work, so probably a migration cli do not support it

- args:
    headers: []
    insert:
      columns: '*'
    name: test_event_1
    retry_conf:
      interval_sec: 30
      num_retries: 30
      timeout_sec: 60
    replace: true
    table:
      name: test_table_1
      schema: public
    webhook: http://localhost:8080/api/test_event_1
  type: create_event_trigger
- args:
    headers: []
    insert:
      columns: '*'
    name: test_event_2
    retry_conf:
      interval_sec: 30
      num_retries: 30
      timeout_sec: 60
    replace: true
    table:
      name: test_table_2
      schema: public
    webhook: http://localhost/test_event_2
  type: create_event_trigger