This pull request fixes a panic that exists in both mempools. The panic occurs when the ABCI client misses a response from the ABCI application. This happen when the ABCI client drops the request as a result of a full client queue. The fix here was to loop through the ordered list of recheck-tx in the callback until one matches the currently observed recheck request.
This is, perhaps, the trival final piece of #7075 that I've been
working on.
There's more work to be done:
- push more of the setup into the pacakges themselves
- move channel-based sending/filtering out of the
- simplify the buffering throuhgout the p2p stack.
This patch was needed to pass the buf breakage check for the proto file removed
in #7121, but now that master contains the change we no longer need the patch.
This change removes the partial gRPC interface to the RPC service, which was
deprecated in resolution of #6718.
Details:
- rpc: Remove the client and server interfaces and proto definitions.
- Remove the gRPC settings from the config library.
- Remove gRPC setup for the RPC service in the node startup.
- Fix various test helpers to remove gRPC bits.
- Remove the --rpc.grpc-laddr flag from the CLI.
Note that to satisfy the protobuf interface check, this change also includes a
temporary edit to buf.yaml, that I will revert after this is merged.
This metric describes itself as 'pending' but never actual decrements when the messages are removed from the queue.
This change fixes that by decrementing the metric when the data is removed from the queue.
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.
Fixes#7068. The build-docker rule relies on being able to run make
build-linux, but did not pull the Makefile into the build context.
There are various ways to fix this, but this was probably the smallest.
* ci: use run-multiple.sh for e2e pr tests
* fix labeling
* Update .github/workflows/e2e-nightly-35x.yml
Co-authored-by: M. J. Fromberger <michael.j.fromberger@gmail.com>
Co-authored-by: M. J. Fromberger <michael.j.fromberger@gmail.com>
Fixes#7098. The light client documentation moved to the spec repository.
I was not able to figure out what happened to light-client-protocol.md, it was removed in #5252 but no corresponding file exists in the spec repository. Since the spec also discusses the protocol, this change simply links to the spec and removes the non-functional reference.
Alternatively we could link to the top-level [light client doc](https://docs.tendermint.com/master/tendermint-core/light-client.html) if you think that's better.
This tweaks the connectivity of test configurations, in hopes that more will be viable.
Additionally reduces the prevalence of testing the legacy mempool.
My earlier p2p cleanup code removed support for the p2p tests from the
e2e generator and runner, but missed removing the CI
configuration. This patch remedies that.
While discussing a question about the indexing interface (#7044), we found some
confusion about the intent of the design decisions in ADR 065.
Based on discussion with the original authors of the ADR, this commit adds some
language to the Decisions section to spell out the intentions more clearly, and
to call out future work that this ADR did not explicitly decide about.
Addresses one of the concerns with #7041.
Provides a mechanism (via the RPC interface) to delete a single transaction, described by its hash, from the mempool. The method returns an error if the transaction cannot be found. Once the transaction is removed it remains in the cache and cannot be resubmitted until the cache is cleared or it expires from the cache.
This code hasn't been battle tested, and seems to have grown
increasingly flaky int tests. Given our general direction of reducing
queue complexity over the next couple of releases I think it makes
sense to remove it.
This PR tackles the case of using the e2e application in a long lived testnet. The application continually saves snapshots (usually every 100 blocks) which after a while bloats the size of the application. This PR prunes older snapshots so that only the most recent 10 snapshots remain.
A few notes:
- this is not all the deletion that we can do, but this is the most
"simple" case: it leaves in shims, and there's some trivial
additional cleanup to the transport that can happen but that
requires writing more code, and I wanted this to be easy to review
above all else.
- This should land *after* we cut the branch for 0.35, but I'm
anticipating that to happen soon, and I wanted to run this through
CI.
The race occurred as a result of a goroutine launched by `processPeerUpdate` racing with the `OnStop` method. The `processPeerUpdates` goroutine deletes from the map as `OnStop` is reading from it. This change updates the `OnStop` method to wait for the peer updates channel to be done before closing the peers. It also copies the map contents to a new map so that it will not conflict with the view of the map that the goroutine created in `processPeerUpdate` sees.
This commit should be one of the first to land as part of the v0.36
cycle *after* cutting the 0.35 branch.
The blocksync/v2 reactor was originally implemented as an experiement
to produce an implementation of the blockstack protocol that would be
easier to test and validate, but it was never appropriately
operationalized and this implementation was never fully debugged. When
the p2p layer was refactored as part of the 0.35 cycle, the v2
implementation was not refactored and it was left in the codebase but
not removed. This commit just removes all references to it.
This script is referenced from the release documentation, we should make sure it's functional. This is helpful in generating the "Special Thanks" section of the changelog.
This is intended to fix a test failure that occurs in the p2p state provider. The issue presents as the state provider timing out waiting for the consensus params response.
The reason that this can occur is because the statesync reactor has the possibility of attempting to respond to the params request before the state provider is ready to read it. This results in the reactor hitting the `default` case seen here and then never sending on the channel. The stateprovider will then block waiting for a response and never receive one because the reactor opted not to send it.
William Banfield
dependabot[bot]
Marko
Sam Kleinman
M. J. Fromberger
Callum Waters
Jared Zhou
lklimek