As a small developer quality of life improvement, I found many individual unit tests that take longer than around a second to complete, and set them to skip when run under `go test -short`.
On my machine, the wall timings for tests (with `go test -count=1 ./...` and optionally `-short` and `-race`) are roughly:
- Long tests, no race detector: about 1m42s
- Short tests, no race detector: about 17s
- Long tests, race detector enabled: about 2m1s
- Short tests, race detector enabled: about 28s
This PR is split into many commits each touching a single package, with commit messages detailing the approximate timing change per package.
This PR makes vote extensions optional within Tendermint. A new ConsensusParams field, called ABCIParams.VoteExtensionsEnableHeight, has been added to toggle whether or not extensions should be enabled or disabled depending on the current height of the consensus engine. Related to: #8453
* Refactor so building and linting works
This is the first step towards implementing vote extensions: generating
the relevant proto stubs and getting the build and linter to pass.
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Fix typo
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Better describe method given vote extensions
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Fix types tests
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Move CanonicalVoteExtension to canonical types proto defs
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Regenerate protos including latest PBTS synchrony params update
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Inject vote extensions into proposal
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Thread vote extensions through code and fix tests
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Remove extraneous empty value initialization
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Fix lint
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Fix missing VerifyVoteExtension request data
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Explicitly ensure length > 0 to sign vote extension
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Explicitly ensure length > 0 to sign vote extension
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Remove extraneous comment
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Update privval/file.go
Co-authored-by: M. J. Fromberger <fromberger@interchain.io>
* Update types/vote_test.go
Co-authored-by: M. J. Fromberger <fromberger@interchain.io>
* Format
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Fix ABCI proto generation scripts for Linux
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Sync intermediate and goal protos
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Update internal/consensus/common_test.go
Co-authored-by: Sergio Mena <sergio@informal.systems>
* Use dummy value with clearer meaning
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Rewrite loop for clarity
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Panic on ABCI++ method call failure
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Add strong correctness guarantees when constructing extended commit info for ABCI++
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Add strong guarantee in extendedCommitInfo that the number of votes corresponds
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Make extendedCommitInfo function more robust
At first extendedCommitInfo expected votes to be in the same order as
their corresponding validators in the supplied CommitInfo struct, but
this proved to be rather difficult since when a validator set's loaded
from state it's first sorted by voting power and then by address.
Instead of sorting the votes in the same way, this approach simply maps
votes to their corresponding validator's address prior to constructing
the extended commit info. This way it's easy to look up the
corresponding vote and we don't need to care about vote order.
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Remove extraneous validator address assignment
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Sign over canonical vote extension
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Validate vote extension signature against canonical vote extension
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Update privval tests for more meaningful dummy value
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Add vote extension capability to E2E test app
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Disable lint for weak RNG usage for test app
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Use parseVoteExtension instead of custom parsing in PrepareProposal
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Only include extension if we have received txs
It's unclear at this point why this is necessary to ensure that the
application's local app_hash matches that committed in the previous
block.
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Require app_hash from app to match that from last block
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Add contrived (possibly flaky) test to check that vote extensions code works
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Remove workaround for problem now solved by #8229
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* add tests for vote extension cases
* Fix spelling mistake to appease linter
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Collapse redundant if statement
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Formatting
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Always expect an extension signature, regardless of whether an extension is present
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Votes constructed from commits cannot include extensions or signatures
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Pass through vote extension in test helpers
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Temporarily disable vote extension signature requirement
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Expand on vote equality test errors for clarity
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Expand on vote matching error messages in testing
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Allow for selective subscription by vote type
This is an attempt to fix the intermittently failing
`TestPrepareProposalReceivesVoteExtensions` test in the internal
consensus package.
Occasionally we get prevote messages via the subscription channel, and
we're not interested in those. This change allows us to specify what
types of votes we're interested in (i.e. precommits) and discard the
rest.
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Read lock consensus state mutex in test helper to avoid data race
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Revert BlockIDFlag parameter in node test
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Perform additional check in ProcessProposal for special txs generated by vote extensions
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* e2e: check that our added tx does not cause all txs to exceed req.MaxTxBytes
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Only set vote extension signatures when signing is successful
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Remove channel capacity constraint in test helper to avoid missing messages
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Add TODO to always require extension signatures in vote validation
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* e2e: reject vote extensions if the request height does not match what we expect
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* types: remove extraneous call to voteWithoutExtension in test
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Remove unnecessary address parameter from CanonicalVoteExtension
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* privval: change test vote type to precommit since we use an extension
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* privval: update signing logic to cater for vote extensions
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* proto: update field descriptions for vote message
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* proto: update field description for vote extension sig in vote message
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* proto/types: use fixed-length 64-bit integers for rounds in CanonicalVoteExtension
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* consensus: fix flaky TestPrepareProposalReceivesVoteExtensions
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* consensus: remove previously added test helper functionality
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* e2e: add error logs when we get an unexpected height in ExtendVote or VerifyVoteExtension requests
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* node_test: get validator addresses from privvals
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* privval/file_test: optimize filepv creation in tests
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* privval: add test to check that vote extensions are always signed
Signed-off-by: Thane Thomson <connect@thanethomson.com>
* Add a script to check documentation for ToC entries. (#8356)
This script verifies that each document in the docs and architecture directory
has a corresponding table-of-contents entry in its README file. It can be run
manually from the command line.
- Hook up this script to run in CI (optional workflow).
- Update ADR ToC to include missing entries this script found.
* build(deps): Bump async from 2.6.3 to 2.6.4 in /docs (#8357)
Bumps [async](https://github.com/caolan/async) from 2.6.3 to 2.6.4.
- [Release notes](https://github.com/caolan/async/releases)
- [Changelog](https://github.com/caolan/async/blob/v2.6.4/CHANGELOG.md)
- [Commits](https://github.com/caolan/async/compare/v2.6.3...v2.6.4)
---
updated-dependencies:
- dependency-name: async
dependency-type: indirect
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* privval/file_test: reset vote ext sig before signing
Signed-off-by: Thane Thomson <connect@thanethomson.com>
Co-authored-by: M. J. Fromberger <fromberger@interchain.io>
Co-authored-by: Sergio Mena <sergio@informal.systems>
Co-authored-by: William Banfield <wbanfield@gmail.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
This change implements the logic for the PrepareProposal ABCI++ method call. The main logic for creating and issuing the PrepareProposal request lives in execution.go and is tested in a set of new tests in execution_test.go. This change also updates the mempool mock to use a mockery generated version and removes much of the plumbing for the no longer used ABCIResponses.
* event: Added Events after evidence validation; evidence: refactored AddEvidence
Added context and Metrics as parameter for the pool constructor
* evidence: pushed event firing into evidence pool and added metrics to represent the size of the evpool
* state: fixed parameters of evpool mock functions
* evidence: added test to confirm events are generated
* Removed obsolete EvidenceEventPublisher interface
* evidence: pool removed error on missing eventbus
Our test cases spew a lot of files and directories around $TMPDIR. Make more
thorough use of the testing package's TempDir methods to ensure these are
cleaned up.
In a few cases, this required plumbing test contexts through existing helper
code. In a couple places an explicit path was required, to work around cases
where we do global setup during a TestMain function. Those cases probably
deserve more thorough cleansing (preferably with fire), but for now I have just
worked around it to keep focused on the cleanup.
This pull request merges in the changes for implementing Proposer-based timestamps into `master`. The power was primarily being done in the `wb/proposer-based-timestamps` branch, with changes being merged into that branch during development. This pull request represents an amalgamation of the changes made into that development branch. All of the changes that were placed into that branch have been cleanly rebased on top of the latest `master`. The changes compile and the tests pass insofar as our tests in general pass.
### Note To Reviewers
These changes have been extensively reviewed during development. There is not much new here. In the interest of making effective use of time, I would recommend against trying to perform a complete audit of the changes presented and instead examine for mistakes that may have occurred during the process of rebasing the changes. I gave the complete change set a first pass for any issues, but additional eyes would be very appreciated.
In sum, this change set does the following:
closes#6942
merges in #6849
Defines a different concrete type that satisfies the service interface for a seed node.
update the seed node unit test to assert the new type.
Fixes#6775
This continues the push of plumbing contexts through tendermint. I
attempted to find all goroutines in the production code (non-test) and
made sure that these threads would exit when their contexts were
canceled, and I believe this PR does that.
Some of our tests were creating a psql event sink and expecting
it to report (or not report) certain kinds of errors. These tests
were ill-founded in a couple of ways:
1. Tests that required the Postgres driver were not loading it.
This led to spurious successes on tests that wanted "some error"
from the sink constructor, but didn't exercise the right path.
2. Tests that wanted a Postgres sink to succeed without a database.
These tests "passed" because they weren't actually establishing a
connection to the database, but if they had would have failed for
the lack of one.
To fix this:
- Load the postgres driver in tests that need it.
- Verify connectivity before reporting successful creation of a PSQL event sink.
- Remove tests that wanted a psql sink without a database, since that case
is already tested elsewhere.
This is a very small change, but removes a method from the
`service.Service` interface (a win!) and forces callers to explicitly
pass loggers in to objects during construction rather than (later)
injecting them. There's not a real need for this kind of lazy
construction of loggers, and I think a decent potential for confusion
for mutable loggers.
The main concern I have is that this changes the constructor API for
ABCI clients. I think this is fine, and I suspect that as we plumb
contexts through, and make changes to the RPC services there'll be a
number of similar sorts of changes to various (quasi) public
interfaces, which I think we should welcome.
This follows the same model as we did in the p2p package.
Rework the indexer service constructor to take a struct of arguments,
that makes it easier to construct the optional settings.
Deprecate but do not remove the existing constructor.
Clean up node initialization a little bit.
This is part of the work described by #7156.
Remove "unbuffered subscriptions" from the pubsub service.
Replace them with a dedicated blocking "observer" mechanism.
Use the observer mechanism for indexing.
Add a SubscribeWithArgs method and deprecate the old Subscribe
method. Remove SubscribeUnbuffered entirely (breaking).
Rework the Subscription interface to eliminate exposed channels.
Subscriptions now use a context to manage lifecycle notifications.
Internalize the eventbus package.
This PR adds an initial set of metrics for use ABCI. The initial metrics enable the calculation of timing histograms and call counts for each of the ABCI methods. The metrics are also labeled as either 'sync' or 'async' to determine if the method call was performed using ABCI's `*Async` methods.
An example of these metrics is included here for reference:
```
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="0.0001"} 0
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="0.0004"} 5
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="0.002"} 12
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="0.009"} 13
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="0.02"} 13
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="0.1"} 13
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="0.65"} 13
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="2"} 13
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="6"} 13
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="25"} 13
tendermint_abci_connection_method_timing_bucket{chain_id="ci",method="commit",type="sync",le="+Inf"} 13
tendermint_abci_connection_method_timing_sum{chain_id="ci",method="commit",type="sync"} 0.007802058000000001
tendermint_abci_connection_method_timing_count{chain_id="ci",method="commit",type="sync"} 13
```
These metrics can easily be graphed using prometheus's `histogram_quantile(...)` method to pick out a particular quantile to graph or examine. I chose buckets that were somewhat of an estimate of expected range of times for ABCI operations. They start at .0001 seconds and range to 25 seconds. The hope is that this range captures enough possible times to be useful for us and operators.
