State DB pruning

[rakanalh]

Description

• Add state DB pruning
• Introduce citrea-cli
• Add skeleton for rollback

Linked Issues

• Fixes Pruning: State DB #1193

[jfldde]

Is that supposed to be running against a stopped node ? Wonder what the behaviour would be of rolling back from the cli against a running node. (Pruning feels safe but we might need to be cautious still).
If not, we might want to find a way to prevent running these commands against in-use dbs

[codecov]

Codecov Report

Attention: Patch coverage is 68.19923% with 83 lines in your changes missing coverage. Please review.

Project coverage is 77.4%. Comparing base (36f0e6c) to head (443f9ff).

Files with missing lines	Patch %	Lines
bin/cli/src/main.rs	0.0%	25 Missing ⚠️
bin/cli/src/commands/prune.rs	0.0%	18 Missing ⚠️
...tes/storage-ops/src/pruning/components/state_db.rs	83.1%	14 Missing ⚠️
crates/storage-ops/src/rollback/service.rs	0.0%	13 Missing ⚠️
crates/storage-ops/src/rollback/mod.rs	0.0%	5 Missing ⚠️
bin/cli/src/commands/rollback.rs	0.0%	3 Missing ⚠️
crates/storage-ops/src/pruning/mod.rs	94.6%	3 Missing ⚠️
bin/citrea/src/main.rs	0.0%	1 Missing ⚠️
crates/storage-ops/src/pruning/service.rs	96.2%	1 Missing ⚠️

Additional details and impacted files

Files with missing lines	Coverage Δ
bin/citrea/src/rollup/mod.rs	79.9% <ø> (ø)
crates/common/src/config.rs	94.8% <ø> (ø)
crates/fullnode/src/lib.rs	100.0% <100.0%> (ø)
...reign-sdk/full-node/db/sov-db/src/schema/tables.rs	91.5% <ø> (ø)
.../sovereign-sdk/full-node/db/sov-db/src/state_db.rs	94.1% <100.0%> (+0.3%)	⬆️
...-sdk/module-system/sov-state/src/prover_storage.rs	94.6% <100.0%> (+0.1%)	⬆️
...es/storage-ops/src/pruning/components/ledger_db.rs	100.0% <ø> (ø)
...es/storage-ops/src/pruning/components/native_db.rs	86.3% <100.0%> (ø)
crates/storage-ops/src/pruning/criteria.rs	100.0% <100.0%> (ø)
crates/storage-ops/src/tests.rs	100.0% <100.0%> (ø)
... and 9 more

... and 4 files with indirect coverage changes

[eyusufatik]

Is that supposed to be running against a stopped node ? Wonder what the behaviour would be of rolling back from the cli against a running node. (Pruning feels safe but we might need to be cautious still). If not, we might want to find a way to prevent running these commands against in-use dbs

this is supposed to run against a stopped node.

[eyusufatik]

does this close #1609

[rakanalh]

does this close #1609

No, to prevent the PR from getting larger and harder to review... i am only placing the skeleton for rollback here and will implement in a follow up PR

[rakanalh]

Is that supposed to be running against a stopped node ? Wonder what the behaviour would be of rolling back from the cli against a running node. (Pruning feels safe but we might need to be cautious still). If not, we might want to find a way to prevent running these commands against in-use dbs

The design or Pruner vs PrunerService and Rollback vs RollbackService will enable us to spawn the service in the future when we use rollback to actually perform rolling back on a running node in case of forks. For now, this is not going to be used.

You cannot use the subcommand on an in-use DB because rocksdb locks the directory of any other process is using the database.