Improve time to start

[endersonmaia]

TL;DR

With first 2 items, already reduced from ~30s to ~9.5s on my environment.

• Removed prompt service
• Adjusted healthcheck intervals
• Preload database (depends on add initialized postgres database to SDK #178)
• Move messages from removed prompt elsewhere (maybe cartesi status and on another PR?)

---

This pull request includes significant changes to the apps/cli directory, focusing on removing the prompt service from various Docker Compose files and adjusting health check configurations.

Removal of prompt service:

• apps/cli/src/commands/rollups/start.ts: Removed the docker-compose-prompt.yaml reference from the list of Docker Compose files.
• apps/cli/src/compose/rollups/docker-compose-anvil.yaml: Removed the prompt service configuration.
• apps/cli/src/compose/rollups/docker-compose-bundler.yaml: Removed the prompt service configuration.
• apps/cli/src/compose/rollups/docker-compose-database.yaml: Removed the prompt service configuration.
• apps/cli/src/compose/rollups/docker-compose-espresso.yaml: Removed the prompt service configuration.
• apps/cli/src/compose/rollups/docker-compose-explorer.yaml: Removed the prompt service configuration.
• apps/cli/src/compose/rollups/docker-compose-graphql.yaml: Removed the prompt service configuration.
• apps/cli/src/compose/rollups/docker-compose-node.yaml: Removed the prompt service configuration.
• apps/cli/src/compose/rollups/docker-compose-paymaster.yaml: Removed the prompt service configuration.
• apps/cli/src/compose/rollups/docker-compose-prompt.yaml: Removed the entire prompt service file.
• apps/cli/src/compose/rollups/docker-compose-proxy.yaml: Removed the prompt service dependency.

Health check configuration adjustments:

• apps/cli/src/compose/rollups/docker-compose-anvil.yaml: Updated health check intervals, timeouts, and retries.
• apps/cli/src/compose/rollups/docker-compose-database.yaml: Updated health check intervals, timeouts, and retries.
• apps/cli/src/compose/rollups/docker-compose-node.yaml: Updated health check intervals, timeouts, and retries.
• apps/cli/src/compose/rollups/docker-compose-proxy.yaml: Updated health check intervals, timeouts, and retries.

[changeset-bot]

⚠️ No Changeset found

Latest commit: be6865d

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[github-actions]

Coverage Report for ./apps/cli

Status	Category	Percentage	Covered / Total
🔵	Lines	28.14%	293 / 1041
🔵	Statements	28.24%	298 / 1055
🔵	Functions	30.17%	51 / 169
🔵	Branches	26.19%	121 / 462

File Coverage

File	Stmts	Branches	Functions	Lines	Uncovered Lines
Changed Files
apps/cli/src/commands/rollups/start.ts	0%	0%	0%	0%	7-187

Generated in workflow #338 for commit be6865d by the Vitest Coverage Report Action

[tuler]

My idea for the replacement of the prompt was to write a typescript library that monitors the up of a docker compose by listening to events of docker compose events json output. The library user would register containers of "interest" and be notified when they get healthy. Then a client of this library could use some presentation framework like https://listr2.kilic.dev to present prompt messages of each container of interest.
The original output of docker compose up --detached could be suppressed from stdout, as the new output would be more user friendly.

[endersonmaia]

My idea for the replacement of the prompt was to write a typescript library that monitors the up of a docker compose by listening to events of docker compose events json output. The library user would register containers of "interest" and be notified when they get healthy. Then a client of this library could use some presentation framework like https://listr2.kilic.dev to present prompt messages of each container of interest. The original output of docker compose up --detached could be suppressed from stdout, as the new output would be more user friendly.

Please, open an issue to track this.

[tuler]

why db volume in anvil?

[endersonmaia]

because it's the container we have always running that uses the cartesi/sdk image which has the initialized database

ugly, I know

[endersonmaia]

having another container with a sleep infinity looked worse for me

do you have any other idea?

[tuler]

Extend postgres image itself with the db?

[endersonmaia]

you mean, make the cartesi/sdk based on postgres?

or maintain another container image cartesi/postgres ?

[endersonmaia]

Replaced by #179