***** H2.0 Production Release *****

[jbrown-xentity]

User Story

In order to get value from Harvester2.0, data managers want Harvester2.0 released to production.

Acceptance Criteria

• GIVEN Harvester2.0 is built
  AND the admin UI, API, scheduler, public UI are live
  WHEN the DNS cutover happens
  THEN all services are at the correct URL's
  AND the new catalog service is in place.

Background

All the harvester2.0 come to fruition

Security Considerations (required)

None

Sketch

• Should require making all the applications available at the correct URL's (admin UI, h2.0 API, public UI, etc). This may be a combination of cloud.gov route setup and DNS work with AWS.
• Then, will require moving catalog-next as the new catalog service in place of current catalog via cloud.gov route management
• resolve/address all TODO's in all apps

[btylerburton]

Tentative plan is April 9 announcement to Open Data Working Group of two-week window to launch to check production instance of Catalog Next and confirm data integrity.

[rshewitt]

going through @btylerburton decouple flask app pr led me to refreshing my memory on the load manager and how we manage jobs. i don't know if this has been mentioned elsewhere but based on how we determine job slots we would need to seed the job table beforehand. oh never mind, i didn't realize that's handled on harvest source creation