Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat: Migrate deployments to PostgreSQL #1948

Closed
chair28980 opened this issue Aug 24, 2023 · 8 comments
Closed

feat: Migrate deployments to PostgreSQL #1948

chair28980 opened this issue Aug 24, 2023 · 8 comments
Assignees
Labels
E:2.1: Production testing of existing protocols See https://github.com/waku-org/pm/issues/49 for details

Comments

@chair28980
Copy link
Contributor

Ensure that new and existing Waku fleets use PostgreSQL by default.

Priority: Critical for launch

@Ivansete-status
Copy link
Collaborator

Hi @chair28980 !
May kindly elaborate a bit more on this point? I can't quite remember why we need that and I was assuming that the PostgreSQL was specifically interesting from the Status point of view.
( cc @jm-clius to shed some light on this .)

@jm-clius
Copy link
Contributor

jm-clius commented Nov 7, 2023

was specifically interesting from the Status point of view

This is true, so as long as the Status fleets are working as expected with PostgreSQL, migrating other deployments is not critical. I would say though that the new fleets (for the Waku Network, for example) should use postgresql to simplify what we maintain and support, but I think this is already the case.

@fryorcraken
Copy link
Collaborator

fryorcraken commented Nov 24, 2023

migrating other deployments is not critical.

I believe that infra team has been complaining that SQLite is not as easy to manage and hence would prefer a migration to PostgreSQL.
I understand their grievance is mainly around the fact that SQLite needs double the space to reduce the watermarks/refragment/vacuum.

Is this statement still valid? I think infra may consider migrating other deployment as critical.

edit: clarified my comment, original comment below


Is this comment still valid considering the complete from infra regarding SQLite database sizes? @jm-clius @Ivansete-status

@Ivansete-status
Copy link
Collaborator

Weekly Update

  • achieved: Preparing environment to stress one single database with multiple Postgres clients writing and reading simultaneously.
  • next: Extend Postgres benchmarking report from previous results and start analyzing the performance of status.test fleet where three nodes will use one single database.

@Ivansete-status
Copy link
Collaborator

Is this comment still valid considering the complete from infra regarding SQLite database sizes? @jm-clius @Ivansete-status

I think it is still valid. We have the "size" retention policy for SQLite database which can be helpful to simplify the maintenance on fleets != status.*

@fryorcraken
Copy link
Collaborator

I think it is still valid. We have the "size" retention policy for SQLite database which can be helpful to simplify the maintenance on fleets != status.*

As long as infra is happy.

@Ivansete-status
Copy link
Collaborator

I will close this because I don't see a clear need to change the current wakuv2.* archive system, for instance.
On the other hand, Jakub doesn't consider this migration (SQLite -> Postgres) very interesting. Instead, he considers we need to reduce the number of fleets ( @jakubgs - correct me if I said something inaccurate .)

( cc @chair28980 )

@chair28980
Copy link
Contributor Author

Closing in favor of work covered in 2024 Milestone scope.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
E:2.1: Production testing of existing protocols See https://github.com/waku-org/pm/issues/49 for details
Projects
Archived in project
Development

No branches or pull requests

5 participants