go/oasis-node/cmd/debug/consim: Initial import

[Yawning]

No description provided.

[codecov]

Codecov Report

Merging #2784 into master will increase coverage by 0.20%.
The diff coverage is 33.33%.

@@            Coverage Diff             @@
##           master    #2784      +/-   ##
==========================================
+ Coverage   66.98%   67.19%   +0.20%     
==========================================
  Files         342      343       +1     
  Lines       32968    32987      +19     
==========================================
+ Hits        22084    22165      +81     
+ Misses       8186     8129      -57     
+ Partials     2698     2693       -5     

Impacted Files	Coverage Δ
go/consensus/tendermint/abci/mux.go	72.62% <ø> (ø)
go/consensus/tendermint/abci/mux_mock.go	0.00% <0.00%> (ø)
go/consensus/tendermint/abci/state.go	53.12% <60.00%> (-0.05%)	⬇️
go/storage/mkvs/lookup.go	69.38% <0.00%> (-2.05%)	⬇️
go/worker/storage/committee/checkpointer.go	66.37% <0.00%> (-1.77%)	⬇️
go/worker/compute/merge/committee/node.go	75.87% <0.00%> (-0.25%)	⬇️
go/storage/metrics.go	88.04% <0.00%> (ø)
go/worker/compute/txnscheduler/committee/node.go	66.35% <0.00%> (+0.61%)	⬆️
go/runtime/committee/client.go	77.15% <0.00%> (+1.29%)	⬆️
... and 11 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 537104d...0518095. Read the comment docs.

[pro-wh]

this all looks okay. some overarching comments:

• nothing exercises this in CI, so it'll rot
• this doesn't exercise the full functionality of staking, which looks at commit info stuff. the current simulator leaves those fields blank

[kostko]

Yeah we should run this as part of the daily test pipeline (see .buildkite/longtests.pipeline.yml).

[Yawning]

this doesn't exercise the full functionality of staking, which looks at commit info stuff. the current simulator leaves those fields blank

At some point, this becomes an exercise in re-implementing tendermint.

Yeah we should run this as part of the daily test pipeline (see .buildkite/longtests.pipeline.yml).

Do you honestly thing this will ever do anything more than burn CI resources? I highly doubt that this codebase will catch anything.

[kostko]

I don't see why this would catch anything more than what the existing long-term workload does, it will more probably catch less. But if we have it, we should at least run it somewhere, otherwise why have it at all?

[pro-wh]

this doesn't exercise the full functionality of staking, which looks at commit info stuff. the current simulator leaves those fields blank

At some point, this becomes an exercise in re-implementing tendermint.

I'm talking about sending fake results of a commit, not actually running a decentralized fault tolerance protocol.

Yeah we should run this as part of the daily test pipeline (see .buildkite/longtests.pipeline.yml).

Do you honestly thing this will ever do anything more than burn CI resources? I highly doubt that this codebase will catch anything.

a short smoketest on CI, not a long workload like this is intended to do. it'll catch code changes that break this.

[pro-wh]

I don't see why this would catch anything more than what the existing long-term workload does, it will more probably catch less. But if we have it, we should at least run it somewhere, otherwise why have it at all?

ok this PR is gonna need a description