acidjunk / acidjunk/autoleaf-tickets
Cutover from old pricelist env to the new one
- Lenguaje dominante
- Sin datos de lenguaje
- Estrellas
- 0
- Forks
- 0
- Métricas de merge de PR
- Sin PR fusionados en 30 d
Descripción
_Originally: https://github.com/virge-io/pricelist-frontend/issues/365_
After #364 is finished, we need to do a cutover from the old environment to the new one. This will probably be a very hard migration, with some risks attached to it (downtime!), so we need to be prepared.
Will probably need to be done during hours the shops are closed (so at night), to avoid impact on customers.
This cutover plan should first be tested with the staging env, before the production env. (This should go without saying.)
Steps needed:
- Set DNS update times for domain names to 5 mins
- RDS snapshot in old env
- `pg_dump` of current prod db to local dev machine
- `pg_restore` current prod db to RDS in new Virge AWS account
- Update DNS settings to point to new backend and frontend (we may need pricelist-infra changes to support both DNS configs?)
- Check pricelist is working correctly with the new setup
- Set DNS update times to be >5 mins (can be after a few days)
Open questions:
- Can we make use of the DR env to limit impact as well in case something goes wrong?
- I think the domain names are now in Formatics AWS account (under route53).. can we migrate this to the Virge AWS account?
@acidjunk can you add missing steps? I have a less good view of how the DR env and mobile app interact with the pricelist backend/frontend, for example.
Guía de contribución
No hay ninguna guía de contribución indexada para este repositorio
Línea de trabajo
Start by reviewing the completion of #364 and the staging environment, then validate the listed RDS snapshot, pg_dump/pg_restore, DNS, and application checks in order. Resolve the DR, mobile-app, and Route53 ownership questions before proceeding; done means the staging cutover succeeds and the production environment is verified on the new setup without customer impact.
Escrito por el modelo de indexación a partir del texto del issue.
Evaluación
- Stack tecnológico
- aws, postgresql
- Área
- cloud, databases, infrastructure, release
- Tipo de issue
- Nueva funcionalidad
- Dificultad
- 5/5
- Tiempo estimado
- Más de una semana
- Estado de actividad
- Tranquilo
- Claridad
- Bastante claro
- Aptitud para principiantes
- 25/100