Currently we will take all leaves and give that to the overseer on
startup, but this is a bad idea when the finality is lagging for
example. There can be many of unfinalized leaves, we don't even need to
look at anymore. To solve this, the pr adds a maximum of 4 leaves we
forward to the overseer and the pr also checks that we only pass uncles
of the best block.
* Lol
* Add the features
* Remove some more runtime logger init calls
* Make companion check work
* Revert "Make companion check work"
This reverts commit a255c798076466c0fa20a4db713fc712772c2b4d.
* Update Substrate
Currently the first last timestamp is `0` and that leads to the first
slot being `0` which is reserved for genesis. There is some debug assert
in BABE that complains about this in Cumulus :D
* Track the collation status in the collator protocol
This pr changes the collator protocol to track the status of a
collation. This is mainly used to log the status of a collation when it
is removed to inform the user if a collation maybe never reached a validator.
* Update node/network/collator-protocol/src/collator_side.rs
Co-authored-by: Robert Habermeier <rphmeier@gmail.com>
* Moare logging
Co-authored-by: Robert Habermeier <rphmeier@gmail.com>
As futures most of the time are catching panics and we don't check this,
it could happen that we have some statement that isn't correct but the
test succeeds successfully.
* Notify collators about seconded collation
This pr adds functionality to inform a collator that its collation was
seconded by a parachain validator. Before this signed statement was only
gossiped over the validation substream. Now, we explicitly send the
seconded statement to the collator after it was validated successfully.
Besides that it changes the `CollatorFn` to return an optional result
sender that is informed when the build collation was seconded by a
parachain validator.
* Add test
* Make sure we only send `Seconded` statements
* Make sure we only receive valid statements
* Review feedback
We only checked all validity votes, ignoring invalid votes. So, the
condidition could not hold. Besides fixing the panic, I removed some old
cruft that isn't required anymore.
* Compress the PoV block before sending it over the network
This pr changes the way we send PoV blocks over the network. We now
compress the PoV block before it is send over the network. This should
reduce the size significant for PoVs which contain the runtime WASM for
example.
* Preallocate 1KB
* Try something..
* Switch to zstd and some renamings
* Make compression/decompression fail in browsers
* Use some sane maximum value
* Update roadmap/implementers-guide/src/types/network.md
Co-authored-by: Andronik Ordian <write@reusable.software>
* Fix and add test
* add
Co-authored-by: Andronik Ordian <write@reusable.software>
Co-authored-by: Robert Habermeier <rphmeier@gmail.com>
* Improve logging to make debugging parachains easier
This pr should make debugging parachains easier, by printing more
information about the validation process.
* 🤦
* moare
* Convert to debug
* Do not send empty view updates to peers
It happened that we send empty view updates to our peers, because we
only updated our finalized block. This could lead to situations where we
overwhelmed sub systems with too many messages. On Rococo this lead to
constant restarts of our nodes, because some node apparently was
finalizing a lot of blocks.
To prevent this, the pr is doing the following:
1. If a peer sends us an empty view, we report this peer and decrease it
reputation.
2. We ensure that we only send a view update when the `heads` changed
and not only the `finalized_number`.
3. We do not send empty `ActiveLeavesUpdates` from the overseer, as this
makes no sense to send these empty updates. If some subsystem is relying
on the finalized block, it needs to listen for the overseer signal.
* Update node/network/bridge/src/lib.rs
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
* Don't work if they're are no added heads
* Fix test
* Ahhh
* More fixes
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
* Add one Jaeger span per relay parent
This adds one Jaeger span per relay parent, instead of always creating
new spans per relay parent. This should improve the UI view, because
subsystems are now grouped below one common span.
* Fix doc tests
* Replace `PerLeaveSpan` to `PerLeafSpan`
* More renaming
* Moare
* Update node/subsystem/src/lib.rs
Co-authored-by: Andronik Ordian <write@reusable.software>
* Skip the spans
* Increase `spec_version`
Co-authored-by: Andronik Ordian <write@reusable.software>
This pr adds support to change the session length of a Rococo chain at
genesis. This is rather useful because Rococo has a session length of
1 hour, while on rococo-local you will now get 1 minute. This improves
the dev experience, because a parachain is only going live at the
start of a new session.
* Fix bug and further optimizations in availability distribution
- There was a bug that resulted in only getting one candidate per block
as the candidates were put into the hashmap with the relay block hash as
key. The solution for this is to use the candidate hash and the relay
block hash as key.
- We stored received/sent messages with the candidate hash and chunk
index as key. The candidate hash wasn't required in this case, as the
messages are already stored per candidate.
* Update node/core/bitfield-signing/src/lib.rs
Co-authored-by: Robert Habermeier <rphmeier@gmail.com>
* Remove the reverse map
* major refactor of receipts & query_live
* finish refactoring
remove ancestory mapping,
improve relay-parent cleanup & receipts-cache cleanup,
add descriptor to `PerCandidate`
* rename and rewrite query_pending_availability
* add a bunch of consistency tests
* Add some last changes
* xy
* fz
* Make it compile again
* Fix one test
* Fix logging
* Remove some buggy code
* Make tests work again
* Move stuff around
* Remove dbg
* Remove state from test_harness
* More refactor and new test
* New test and fixes
* Move metric
* Remove "duplicated code"
* Fix tests
* New test
* Change break to continue
* Update node/core/av-store/src/lib.rs
* Update node/core/av-store/src/lib.rs
* Update node/core/bitfield-signing/src/lib.rs
Co-authored-by: Fedor Sakharov <fedor.sakharov@gmail.com>
* update guide to match live_candidates changes
* add comment
* fix bitfield signing
Co-authored-by: Robert Habermeier <rphmeier@gmail.com>
Co-authored-by: Bernhard Schuster <bernhard@ahoi.io>
Co-authored-by: Fedor Sakharov <fedor.sakharov@gmail.com>
* Adds consistency checks for the `HostConfiguration`
Besides that it fixes the chain specs to make the consistency checks happy.
* Update runtime/parachains/src/configuration.rs
Co-authored-by: Shawn Tabrizi <shawntabrizi@gmail.com>
* Review feedback and test fixes etc
* Update
* More
* I'm an idiot
* Fix tests...
Co-authored-by: Shawn Tabrizi <shawntabrizi@gmail.com>
* Do not spam when we can not send a message to a job
There are legal reasons why a job ended. If a job failed, the error is
logged. So, we don't need to log an error when we can not send a message
to a job.
* Review feedback
* Rework `ConnectionsRequests`
Instead of implementing the `Stream` trait, this struct now provides a
function `next()`. This enables us to encode into the type system that
it will always return a value or block indefinitely.
* Review feedback
* Add an upper number of maximum parallel runtime api requests
Instead of spawning all runtime api requests in the background and using
all wasm instances. This pr adds a maximum number of parallel requests.
* Update node/core/runtime-api/src/lib.rs
Co-authored-by: Sergei Shulepov <sergei@parity.io>
* Review feedback
* Increase instances
* Add warning
* Update node/core/runtime-api/src/lib.rs
Co-authored-by: Sergei Shulepov <sergei@parity.io>
Co-authored-by: Sergei Shulepov <sergei@parity.io>
* Make it possible for the adder collator to calculate any state
This is very useful for when wanting to have multiple running or when
wanting to restart the collator.
* Update parachain/test-parachains/adder/collator/src/lib.rs
Co-authored-by: Sergei Shulepov <sergei@parity.io>
* Simplify subsystem jobs
This pr simplifies the subsystem jobs interface. Instead of requiring an
extra message that is used to signal that a job should be ended, a job
now ends when the receiver returns `None`. Besides that it changes the
interface to enforce that messages to a job provide a relay parent.
* Drop ToJobTrait
* Remove FromJob
We always convert this message to FromJobCommand anyway.
This pr changes how the runtime api subsystem processes runtime api
requests. Instead of answering all of them in the subsystem task and
thus, making all requests sequential, we now answer them in a background
task. This enables us to serve multiple requests at once.
* Some code cleanup in overseer
- Switches to select! in the overseer run loop to be more fair about
message processing between the different sources.
- Added a check to only send `ActiveLeaves` if the update actually
contains any data.
* Move the check
* Restore old behavior
* Simplify message sending and signal sending to subsystems
* Update node/subsystem/src/lib.rs
* Do not send messages twice in bitfield distribution
This removes a bug which resulted in sending bitfield messages multiple
times by not checking if we already relayed them. Besides that it also
adds an optimization to not relay a message to a peer that send us
this message.
* Review comments
* Break some lines
* Improve collator side of the collator-protocol
This pr improves the collator-protocol implementation of the collator
side. Besides cleaning up code and rewriting it, the following changed:
- Before on `PeerViewChange` we send an advertisment to every peer, now
this only happens for validators.
- It also adds a check that we send an advertisment message only once
for a connected peer.
- If the same validator was part of the current and next group, we
requested to be connected to this validator two times. This is also
fixed now.
- Instead of having only one connection request, we now are being able
to store multiple of them. This is required as we can have multiple
active leafs at any point of time.
* Switch to common `ConnectionRequests`
* Update node/network/collator-protocol/src/collator_side.rs
* Revert "Restrict collating_using_adder_collator test on feature real-overseer (#1947)"
This reverts commit 75da335d66.
* Revert "Revert "Restrict collating_using_adder_collator test on feature real-overseer (#1947)""
This reverts commit 4e92cdfb1046b0e3da7627297050ae71ae3dd6e3.
* Test something
* Adds integration test based on adder collator
This adds an integration test for parachains that uses the adder
collator. The test will start two relay chain nodes and one collator and
waits until 4 blocks are build and enacted by the parachain.
* Make sure the integration test is run in CI
* Fix wasm compilation
* Update parachain/test-parachains/adder/collator/src/lib.rs
Co-authored-by: Sergei Shulepov <sergei@parity.io>
* Update cli/src/command.rs
Co-authored-by: Sergei Shulepov <sergei@parity.io>
We need to distribute the PoV after we have seconded it. Other nodes
that will receive our `Secondded` statement and want to validate the
candidate another time will request this PoV from us.
If a core is freed, we need to make sure it is inserted as free into
availability cores, because we can not be sure that the core is occupied
directly again.
* Make `CandidateHash` a real type
This pr adds a new type `CandidateHash` that is used instead of the
opaque `Hash` type. This helps to ensure on the type system level that
we are passing the correct types.
This pr also fixes wrong usage of `relay_parent` as `candidate_hash`
when communicating with the av storage.
* Update core-primitives/src/lib.rs
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
* Wrap the lines
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
* Do not validate a candidate in candidate selection
The candidate selection subsystem should not validate a candidate, as
this is done by the backing subsystem on a `Second` request. Otherwise
we validate one candidate twice.
* Update candidate-selection.md
* Moare fixes for parachains
- Sending data to a job should always contain a relay parent. Done this
for the provisioner
- Fixed the `select_availability_bitfields` function. It was assuming we
have one core per validator, while we only have one core per parachain.
- Drive by async "rewrite" in proposer
* Make tests compile
* Update primitives/src/v1.rs
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
* Parachain improvements
- Set the parachains configuration in Rococo genesis
- Don't stop the overseer when a subsystem job is stopped
- Several small code changes
* Remove unused functionality
* Return error from the runtime instead of printing it
* Apply suggestions from code review
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
* Update primitives/src/v1.rs
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
* Update primitives/src/v1.rs
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
* Fix test
* Revert "Update primitives/src/v1.rs"
This reverts commit 11fce2785acd1de481ca57815b8e18400f09fd52.
* Revert "Update primitives/src/v1.rs"
This reverts commit d6439fed4f954360c89fb1e12b73954902c76a41.
* Revert "Return error from the runtime instead of printing it"
This reverts commit cb4b5c0830ac516a6d54b2c24197e9354f2b98cb.
* Revert "Fix test"
This reverts commit 0c5fa1b5566d4cd3c55a55d485e707165ce7a59e.
* Update runtime/parachains/src/runtime_api_impl/v1.rs
Co-authored-by: Sergei Shulepov <sergei@parity.io>
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
Co-authored-by: Sergei Shulepov <sergei@parity.io>
* Fixes bug that collator wasn't sending `Declare` message
* Set authority discovery config
* Fixes bug that collator wasn't sending `Declare` message
* Adds real overseer feature and makes the wasm_validation fail with a
proper error
* Adds README
* Remove debug stuff
* Add feature
* Make adder collator use the correct parent when building a new block
* start working on building the real overseer
Unfortunately, this fails to compile right now due to an upstream
failure to compile which is probably brought on by a recent upgrade
to rustc v1.47.
* fill in AllSubsystems internal constructors
* replace fn make_metrics with Metrics::attempt_to_register
* update to account for #1740
* remove Metrics::register, rename Metrics::attempt_to_register
* add 'static bounds to real_overseer type params
* pass authority_discovery and network_service to real_overseer
It's not straightforwardly obvious that this is the best way to handle
the case when there is no authority discovery service, but it seems
to be the best option available at the moment.
* select a proper database configuration for the availability store db
* use subdirectory for av-store database path
* apply Basti's patch which avoids needing to parameterize everything on Block
* simplify path extraction
* get all tests to compile
* Fix Prometheus double-registry error
for debugging purposes, added this to node/subsystem-util/src/lib.rs:472-476:
```rust
Some(registry) => Self::try_register(registry).map_err(|err| {
eprintln!("PrometheusError calling {}::register: {:?}", std::any::type_name::<Self>(), err);
err
}),
```
That pointed out where the registration was failing, which led to
this fix. The test still doesn't pass, but it now fails in a new
and different way!
* authorities must have authority discovery, but not necessarily overseer handlers
* fix broken SpawnedSubsystem impls
detailed logging determined that using the `Box::new` style of
future generation, the `self.run` method was never being called,
leading to dropped receivers / closed senders for those subsystems,
causing the overseer to shut down immediately.
This is not the final fix needed to get things working properly,
but it's a good start.
* use prometheus properly
Prometheus lets us register simple counters, which aren't very
interesting. It also allows us to register CounterVecs, which are.
With a CounterVec, you can provide a set of labels, which can
later be used to filter the counts.
We were using them wrong, though. This pattern was repeated in a
variety of places in the code:
```rust
// panics with an cardinality mismatch
let my_counter = register(CounterVec::new(opts, &["succeeded", "failed"])?, registry)?;
my_counter.with_label_values(&["succeeded"]).inc()
```
The problem is that the labels provided in the constructor are not
the set of legal values which can be annotated, but a set of individual
label names which can have individual, arbitrary values.
This commit fixes that.
* get av-store subsystem to actually run properly and not die on first signal
* typo fix: incomming -> incoming
* don't disable authority discovery in test nodes
* Fix rococo-v1 missing session keys
* Update node/core/av-store/Cargo.toml
* try dummying out av-store on non-full-nodes
* overseer and subsystems are required only for full nodes
* Reduce the amount of warnings on browser target
* Fix two more warnings
* InclusionInherent should actually have an Inherent module on rococo
* Ancestry: don't return genesis' parent hash
* Update Cargo.lock
* fix broken test
* update test script: specify chainspec as script argument
* Apply suggestions from code review
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* Update node/service/src/lib.rs
Co-authored-by: Bastian Köcher <bkchr@users.noreply.github.com>
* node/service/src/lib: Return error via ? operator
* post-merge blues
* add is_collator flag
* prevent occasional av-store test panic
* simplify fix; expand application
* run authority_discovery in Role::Discover when collating
* distinguish between proposer closed channel errors
* add IsCollator enum, remove is_collator CLI flag
* improve formatting
* remove nop loop
* Fix some stuff
* Adds test parachain adder collator
* Add sudo to Rococo, change session length to 30 seconds and some renaming
* Update to the latest changes on master
* Some fixes
* Fix compilation
* Update parachain/test-parachains/adder/collator/src/lib.rs
Co-authored-by: Sergei Shulepov <sergei@parity.io>
* Review comments
* Downgrade transaction version
* Fixes
* MOARE
* Register notification protocols
* utils: remove unused error
* av-store: more resilient to some errors
* address review nits
* address more review nits
Co-authored-by: Peter Goodspeed-Niklaus <peter.r.goodspeedniklaus@gmail.com>
Co-authored-by: Andronik Ordian <write@reusable.software>
Co-authored-by: Fedor Sakharov <fedor.sakharov@gmail.com>
Co-authored-by: Robert Habermeier <robert@Roberts-MBP.lan1>
Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com>
Co-authored-by: Max Inden <mail@max-inden.de>
Co-authored-by: Sergey Shulepov <s.pepyakin@gmail.com>
Co-authored-by: Sergei Shulepov <sergei@parity.io>
* Make sure validator discovery works with a delayed peer to validator mapping
Currently the implementation checks on connect of a peer if this peer is
a validator by asking the authority discovery. It can now happen that
the authority discovery is not yet aware that a given peer is an
authority. This can for example happen on start up of the node.
This pr changes the behavior, to make it possible to later associate a
peer to a validator id. Instead of just storing the connected
validators, we now store all connected peers with a vector of associated
validator ids. When we get a request to connect to a given given set of
validators, we start by checking the connected peers. If we didn't find
a validator id in the connected peers, we ask the authority discovery
for the peerid of a given authority id. When the returned peerid is part
of our connected peers set, we cache and return the authority id.
* Update node/network/bridge/Cargo.toml
Co-authored-by: Pierre Krieger <pierre.krieger1708@gmail.com>
* Update node/network/bridge/src/validator_discovery.rs
Co-authored-by: Pierre Krieger <pierre.krieger1708@gmail.com>
* Update `Cargo.lock`
Co-authored-by: Pierre Krieger <pierre.krieger1708@gmail.com>
* Improve and unify testing facilities
This improves the testing facilities by making the test client easier to
use. It also removes code that is not required for the test client.
Besides that it also moves the test service and test client under
`node/test`.
* Update Cargo.lock
* Update node/test/client/src/block_builder.rs
Co-authored-by: Andronik Ordian <write@reusable.software>
* Remove explicit lifetime annotation
* Fix warnings and add extra `BlockBuilderExt`
Co-authored-by: Andronik Ordian <write@reusable.software>
* Make register/deregister parachain dispatchables
This makes `register_parachain` and `deregister_parachain` of
`Registrar` dispatchables.
Besides that it brings back the functionality of the test node to
register a parachain.
* Fix tests
* PR review comments
* Make `AllSubsystems` usage easier in tests
This makes the usage of `AllSubsystems` easier in tests by introducing
new methods.
- `dummy` initializes `AllSubsystems` with all systems set to dummy
- `replace_*` to replace any subsystem
Besides that this pr adds a `ForwardSubsystem` that is also useful for
tests. This subsystem will forward all incoming messages to the given channel.
* Update node/overseer/src/lib.rs
Co-authored-by: Andronik Ordian <write@reusable.software>
* Update node/subsystem/src/lib.rs
Co-authored-by: Andronik Ordian <write@reusable.software>
* Update node/subsystem/src/lib.rs
Co-authored-by: Andronik Ordian <write@reusable.software>
* Move ForwardSubsystem and add a test
* Break some lines
Co-authored-by: Andronik Ordian <write@reusable.software>
This pr changes the collator interface function to return an optional
collation instead of a collation. This is required as the parachain
itself can fail to generate a valid collation for various reason. Now if
the collation fails it will return `None`.
Besides that the pr adds some `RuntimeDebug` derive for `ValidationData`
and removes some whitespaces.
* Prepare Polkadot to be used by Cumulus
This begins to make Polkadot usable from Cumulus.
* Remove old test
* Reset some changes
* More reverts
* MOARE
* Don't use `unimplemented!`
* Make parachain validation wasm executor functional
- Increase the size of the validation result in the shared memory. The
validation result holds the new runtime when a runtime upgrade is
scheduled. So, we need to give it enough memory to send the data between
the validator and the wasm execution host.
- Add the `CallInWasmExt`. This is required when doing a runtime upgrade
to check that we upgrade to something meaningful.
* Update parachain/src/wasm_executor/mod.rs
* Update parachain/src/wasm_executor/mod.rs
Co-authored-by: Nikolay Volf <nikvolf@gmail.com>
Co-authored-by: Nikolay Volf <nikvolf@gmail.com>
* Support `build-spec` for other chains than Polkadot
The problem when building a chain specification is that you require the
native runtime to parse the json file (assuming the chain spec is not
raw yet). Before this pr we could only overwrite the native runtime when
running the node using `force_*`. This pr now adds support to load the
native runtime when the filename starts with the name of the chain. So,
when usng `build-spec --chain rococo-something-else.jon` it will use the
rococo native runtime to load the chain spec.
* Apply suggestions from code review
Co-authored-by: Andronik Ordian <write@reusable.software>
Co-authored-by: Andronik Ordian <write@reusable.software>
* Rewrite client handling
We are supporting muliple polkadot-like chains and all have different
client types. This pr reworks the client handling by having all of them
in one enum combined. Besides that, there is added a special trait
`ExecuteWithClient` to use the internal client.
* Apply suggestions from code review
Co-authored-by: Robert Habermeier <rphmeier@gmail.com>
* Up the versions
* Fix Cargo.lock
* Fix merge conflict
* ......................
* ....v2
* yep
* I'm dumb...
* Browser lol
Co-authored-by: Robert Habermeier <rphmeier@gmail.com>
Bastian Köcher