Validate params in RPC server

[emlautarom1]

Current Behavior

The current RPC server does not validate the parameters on each method, opening up the possibility of errors due to nil dereference panics or unsatisfied invariants.

New Behavior

Each method on the RPC server validates the parameters returning an error in case of failure before doing any actual work.

Breaking Changes

There should not be any breaking changes.

Observations

Currently we are only checking for hidden nil values.

[Hyodar]

LGTM. Can you also add a basic max range check for the checkpoint messages timestamps as well? Just a sanity bound to avoid overloads (e.g. 2h range max between start and end)

[emlautarom1]

@Hyodar added range check to RPC

[Hyodar]

Int test failed with the timestamp range check because the first checkpoint is actually 0 -> current. We could change this without much of an issue so the first checkpoint is toTimestamp - interval -> toTimestamp (feel free to do that), but this reminds me we still then need to chunk GetAggregatedCheckpointMessages calls on the operators if this is ever needed.

[emlautarom1]

We could change this without much of an issue so the first checkpoint is toTimestamp - interval -> toTimestamp

Could you point me out to where this would need to be changed?

[Hyodar]

Could you point me out to where this would need to be changed?

In sendNewCheckpointTask, you can first get the toTimestamp and, if the last timestamp is 0, just do fromTimestamp = toTimestamp - agg.checkpointInterval or such.

[emlautarom1]

There still seems to be an issue here. I've changed the sendNewCheckpointTask interval as follows:

	toTimestamp := block.Time()
	fromTimestamp := lastCheckpointToTimestamp + 1
	if lastCheckpointToTimestamp == 0 {
		fromTimestamp = toTimestamp - uint64(agg.checkpointInterval)
	}

yet the integration test continues to fail.

[Hyodar]

I think you need to get the seconds only, seems to be underflowing. Here's the error:

{"level":"info","ts":1720032516.9505258,"caller":"aggregator/aggregator.go:397","msg":"Aggregator sending new task","fromTimestamp":18446744035429584175,"toTimestamp":1720032559}
{"level":"error","ts":1720032516.9521868,"caller":"chainio/avs_writer.go:88","msg":"Error assembling CreateCheckpointTask tx","stacktrace":"github.com/NethermindEth/near-sffl/core/chainio.(*AvsWriter).SendNewCheckpointTask\n\t/home/runner/work/near-sffl/near-sffl/core/chainio/avs_writer.go:88\ngithub.com/NethermindEth/near-sffl/aggregator.(*Aggregator).sendNewCheckpointTask\n\t/home/runner/work/near-sffl/near-sffl/aggregator/aggregator.go:399"}
{"level":"error","ts":1720032516.952221,"caller":"aggregator/aggregator.go:401","msg":"Aggregator failed to send checkpoint","err":"execution reverted: revert: fromTimestamp greater than toTimestamp","stacktrace":"github.com/NethermindEth/near-sffl/aggregator.(*Aggregator).sendNewCheckpointTask\n\t/home/runner/work/near-sffl/near-sffl/aggregator/aggregator.go:401"}

[emlautarom1]

I think you need to get the seconds only, seems to be underflowing

Good catch, this is not the first time that I've had the same issue.