* create a test for message broadcast to peers
* remove dbg comment
* use `prop_assert_eq`
* use a random number of peers with constant version
* fix some docs
* check total peers is equal to active peers
* increase coverage
Co-authored-by: teor <teor@riseup.net>
* Justify that the ErrorSlot Mutex is deadlock-safe
* Document cancellation safety in the async RFC
* Document task starvation in the async RFC
Co-authored-by: Marek <mail@marek.onl>
* Tweak crawler timings so peers are more likely to be available
* Tweak min peer connection interval so we try all peers
* Let other tasks run between fanouts, so we're more likely to choose different peers
* Let other tasks run between retries, so we're more likely to choose different peers
* Let other tasks run after peer crawler DemandDrop
This makes it more likely that peers will become ready.
* Spawn the address book updater on a blocking thread
* Spawn CandidateSet address book operations on blocking threads
* Replace the PeerSet address book with a metrics watch channel
* Fix comment
* Await spawned address book tasks
* Run the address book update tasks concurrently (except for the mutex)
* Explain an internal-only method better
* Fix a typo
Co-authored-by: Alfredo Garcia <oxarbitrage@gmail.com>
Co-authored-by: Alfredo Garcia <oxarbitrage@gmail.com>
* Drop peer services if their cancel handles are dropped
* Exit the client task if the heartbeat task exits
* Allow multiple errors on a connection without panicking
* Explain why we don't need to send an error when the request is cancelled
* Document connection fields
* Make sure connections don't hang due to spurious timer or channel usage
* Actually shut down the client when the heartbeat task exits
* Add tests for unready services
* Close all senders to peer when `Client` is dropped
* Return a Client error if the error slot has an error
* Add tests for peer Client service errors
* Make Client drop and error cleanups consistent
* Use a ClientDropped error when the Client struct is dropped
* Test channel and error state in peer Client tests
* Move all Connection cleanup into a single method
* Add tests for Connection
* fix typo in comment
Co-authored-by: Conrado Gouvea <conrado@zfnd.org>
Co-authored-by: Conrado Gouvea <conrado@zfnd.org>
Co-authored-by: Alfredo Garcia <oxarbitrage@gmail.com>
ZIP-201 describes how a Zcash node should behave when it reaches a
network upgrade activation height. Zebra doesn't implement all the
details specified there, so we need to document what it does implement
and what it doesn't and why.
* Replace usage of `discover::Change` with a tuple
Remove the assumption that a `Remove` variant would never be created
with type changes that allow the compiler to guarantee that assumption.
* Add a `version` field to the `Client` type
Keep track of the peer's reported protocol version.
* Create `LoadTrackedClient` type
A `peer::Client` type wrapper that implements `Load`. This helps with
the creation of a client service that has extra peer information to be
accessed without having to send requests.
* Use `LoadTrackedClient` in `initialize`
Ensure that `PeerSet` receives `LoadTrackedClient`s so that it will be
able to query the peer's protocol version later on.
* Require `LoadTrackedClient` in `PeerSet`
Replace the generic type with a concrete `LoadTrackedClient` so that we
can query its version.
* Create `MinimumPeerVersion` helper type
A type to track the current minimum protocol version for connected
peers based on the current block height.
* Use `MinimumPeerVersion` in handshakes
Keep the code to obtain the current minimum peer protocol version in a
central place.
* Add a `MinimumPeerVersion` instance to `PeerSet`
Prepare it to be able to disconnect from outdated peers based on the
current minimum supported peer protocol version.
* Disconnect from ready services for outdated peers
When the minimum peer protocol version is detected to have changed
(because of a network upgrade), remove all ready services of peers that
became outdated.
* Cancel added unready services of outdated peers
Only add an unready service if it's for a peer that has a supported
protocol version. Otherwise, add it but drop the cancel handle so that
the `UnreadyService` can execute and detect that it was cancelled.
* Avoid adding ready services for outdated peers
If a service becomes ready but it's for a connection to an outdated
peer, drop it.
* Improve comment inside `crawl_and_dial`
Describe an edge case that is also handled but was not explicit.
Co-authored-by: teor <teor@riseup.net>
* Test if calculated minimum peer version is correct
Given an arbitrary best chain tip height, check that the calculated
minimum peer protocol version is the expected value.
* Test if minimum version changes with chain tip
Apply an arbitrary list of chain tip height updates and check that for
each update the minimum peer version is calculated correctly.
* Test minimum peer version changed reports
Simulate a series of best chain tip height updates, and check for
minimum peer version updates at least once between them. Changes should
only be reported once.
* Create a `MockedClientHandle` helper type
Used to create and then track a mock `Client` instance.
* Add `MinimumPeerVersion::with_mock_chain_tip`
An extension method useful for tests, that contains some shared
boilerplate code.
* Bias arbitrary `Version`s to be in valid range
Give a 50% chance for an arbitrary `Version` to be in the range of
previously used values the Zcash network.
* Create a `PeerVersions` helper type
Helps with the creation of mocked client services with arbitrary
protocol versions.
* Create a `PeerSetGuard` helper type
An auxiliary type to a `PeerSet` instance created for testing. It keeps
track of any dummy endpoints of channels created and passed to the
`PeerSet` instance.
* Create a `PeerSetBuilder` helper type
Helps to reduce the code when preparing a `PeerSet` test instance.
* Test if outdated peers are rejected by `PeerSet`
Simulate a set of discovered peers being sent to the `PeerSet`. Ensure
that only up-to-date peers are kept by the `PeerSet` and that outdated
peers are dropped.
* Create `BlockHeightPairAcrossNetworkUpgrades` type
A helper type that allows the creation of arbitrary block height pairs,
where one value is before and the other is at or after the activation
height of an arbitrary network upgrade.
* Test if peers are dropped as they become outdated
Simulate a network upgrade, and check that peers that become outdated
are dropped by the `PeerSet`.
* Remove dbg! macros
Co-authored-by: teor <teor@riseup.net>
* Stop checking the entire AddressBook for each connection attempt
* Stop redundant peer time checks within the address book
* Stop calling `Instant::now` 3 times for each address book update
* Only get the time once each time an address book method is called
* Update outdated comment
* Use an OrderedMap to efficiently store address book peers
* Add address book order tests
* Tweak a log message
* Only retry failed DNS once, then use the other DNS responses
* Limit broadcasts to half the peers
* Use a longer minimum interval for GetAddr requests
* Reduce the syncer and mempool crawler fanouts
* Stop resetting the mempool twice when it starts up
This spawns two crawlers, which send two fanouts,
so it can use up a lot of peers.
Co-authored-by: Conrado Gouvea <conrado@zfnd.org>
Co-authored-by: Alfredo Garcia <oxarbitrage@gmail.com>
* Rewrite PeerSet comments to split long sentences
* Replace peer set integer indexes with address-based indexes
Also improve documentation and logging.
* Security: Stop using peer addresses to choose inventory routing order
* Minor doc and code cleanups
* Stop re-using a drained HashSet
* Replace used `_cancel` with `cancel`
* Reword a comment
* Replace cloned with copied
* Shut down channels and tasks on PeerSet Drop
* Document all the PeerSet fields
* Close the peer set background task handle on shutdown
* Receive background tasks during shutdown
Also, split receiving and polling background tasks into separate methods.
* Limit open inbound connections based on the config
* Log inbound connection errors at debug level
* Test inbound connection limits
* Use clone directly in function call argument lists
* Remove an outdated comment
* Update tests to use an unbounded channel rather than mem::forget
And rename some variables.
* Use a lower limit in a slow test and require that it is exceeded
* Replace some unit tuples with named unit structs
This helps distinguish generic channels and make them type-safe.
Also tidy imports and documentation in `peer_set::set`.
* Link to the tower balance crate from docs
Co-authored-by: Alfredo Garcia <oxarbitrage@gmail.com>
Co-authored-by: Alfredo Garcia <oxarbitrage@gmail.com>
* Rename internal network requests for wide transaction IDs
fastmod TransactionsByHash TransactionsById zebra*
fastmod AdvertiseTransactions AdvertiseTransactionIds zebra*
fastmod MempoolTransactions MempoolTransactionIds zebra*
fastmod TransactionHashes TransactionIds zebra*
* Update network transaction request/response comments
* Rename a transaction hash method for wide transaction IDs
fastmod transaction_hashes transaction_ids zebra-network
* Add UnminedTxId methods and conversions for InventoryHash
* Map WtxIds to unmined transaction network messages
Also, use UnminedTxId and UnminedTx in:
* Zebra's internal request and response format, and
* external Zcash network protocol messages.
* Enable WtxId mempool inventory tracking for peers
* Further clarify transaction IDs
* Use Witnessed rather than Wide for transaction IDs
And rename narrow to legacy when it only applies to v1-v4 transactions.
Otherwise, rename it to mined ID.
* Rename a missed binding
* Remove an incorrectly named binding
Co-authored-by: Janito Vaqueiro Ferreira Filho <janito.vff@gmail.com>
- stop putting inbound addresses in the address book
- drop address book entries that can't be used for outbound connections
- distinguish between temporary inbound and permanent outbound peer
addresses
- also create variants to handle proxy connections
(but don't use them yet)
- avoid tracking connection state for isolated connections
- document security constraints for the address book and peer set
## Motivation
This PR is motivated by the regression identified in https://github.com/ZcashFoundation/zebra/issues/1349. That PR notes that the metrics stopped working for most of the crates other than `zebrad`.
## Solution
This PR resolves the regression by deduplicating the `metrics` crate dependency. During a recent change we upgraded the metrics version in `zebrad` and a couple other of our crates, but we never updated the dependencies in `zebra-state`, `zebra-consensus`, or `zebra-network`. This caused the metrics macros to attempt to retrieve the current metrics exporter through the wrong function. We would install the metrics exporter in `0.13`, but then attempt to look it up through the `0.12` crate, which contains a different instance of the metrics exporter static variable which is unset. Doing this causes the metrics macros to return `None` for the current exporter after which they just silently give up.
## Related Issues
closes https://github.com/ZcashFoundation/zebra/issues/1349
## Follow Up Work
I noticed we have quite a few duplicate dependencies in our tree. We might be able to save some compilation time by auditing those and deduplicating them as much as possible.
- https://github.com/ZcashFoundation/zebra/issues/1582
Co-authored-by: teor <teor@riseup.net>
* warn: if there are no peers at all
* info: if there are no ready peers
* trace: the number of ready and unready peers for every request
Log at most one warn or info log per minute, to avoid flooding the
terminal with log lines. Suppress warn and info logs for the first
minute, while the peer set is starting up.
Closes#1183.
The peer set maintains a preselected ready service that it can use to
perform power-of-two-choices (p2c) routing of requests. Ready services
are stored by key (socket address) in an `IndexMap`, and the preselected
service is represented by an `Option<usize>` indexing that map. This
means that whenever the set of ready services changes (e.g., a service
is removed from the peer set, or a service is taken to be used to
process a request), the preselected index is invalidated. The original
P2C-only implementation maintained this invariant but did not document
it.
The change to inventory-based routing introduced a bug by failing to
maintain this invariant and appropriately invalidate the preselected
index. However, this was only noticeable approximately 1/N of the time
on the next request after an inventory-directed request, so the bug
occurred infrequently. Luckily, the use of `.expect` caused the bug to
be an immediate panic, making it possible to identify by inspecting all
uses of the ready service map.
This addresses at least three pain points:
- we were affected by bugs that were already fixed in git, but not in
the released crate;
- we can use service combinators to transform requests and responses;
- we can use the hedge middleware.
The version in git is still marked as 0.3.1 but these changes will be
part of tower 0.4: https://github.com/tower-rs/tower/issues/431
This commit makes several related changes to the network code:
- adds a `TransactionsByHash(HashSet<transaction::Hash>)` request and
`Transactions(Vec<Arc<Transaction>>)` response pair that allows
fetching transactions from a remote peer;
- adds a `PushTransaction(Arc<Transaction>)` request that pushes an
unsolicited transaction to a remote peer;
- adds an `AdvertiseTransactions(HashSet<transaction::Hash>)` request
that advertises transactions by hash to a remote peer;
- adds an `AdvertiseBlock(block::Hash)` request that advertises a block
by hash to a remote peer;
Then, it modifies the connection state machine so that outbound
requests to remote peers are handled properly:
- `TransactionsByHash` generates a `getdata` message and collects the
results, like the existing `BlocksByHash` request.
- `PushTransaction` generates a `tx` message, and returns `Nil` immediately.
- `AdvertiseTransactions` and `AdvertiseBlock` generate an `inv`
message, and return `Nil` immediately.
Next, it modifies the connection state machine so that messages
from remote peers generate requests to the inbound service:
- `getdata` messages generate `BlocksByHash` or `TransactionsByHash`
requests, depending on the content of the message;
- `tx` messages generate `PushTransaction` requests;
- `inv` messages generate `AdvertiseBlock` or `AdvertiseTransactions`
requests.
Finally, it refactors the request routing logic for the peer set to
handle advertisement messages, providing three routing methods:
- `route_p2c`, which uses p2c as normal (default);
- `route_inv`, which uses the inventory registry and falls back to p2c
(used for `BlocksByHash` or `TransactionsByHash`);
- `route_all`, which broadcasts a request to all ready peers (used for
`AdvertiseBlock` and `AdvertiseTransactions`).
Prior to this change, we required that services that are canceled do not
have a cancel handle in the `cancel_handles` list, based on the
assumption that the handle must have been removed in the process of
canceling this service.
This doesn't holding up though, because it is currently possible for us
to have the same peer connect to us multiple times, the second connect
removes the cancel handle of the original connect and inserts it's own
cancel handle in its place. In this scenario, when the first service is
polled for readiness it will see that it has been canceled and go to
clean itself up, but when it asserts that it doesn't have a cancel
handle it will see the cancel handle of the second connect event, which
uses the same key as the first connect, and fail its debug assertion.
This change removes that debug assert on the assumption that it is okay
for a peer to connect multiple times consecutively, and that the correct
behavior in that case is to just cancel the first connection and
continue as normal.
Prior to this change, the service returned by `zebra_network::init` would spawn background tasks that could silently fail, causing unexpected errors in the zebra_network service.
This change modifies the `PeerSet` that backs `zebra_network::init` to store all of the `JoinHandle`s for each background task it depends on. The `PeerSet` then checks this set of futures to see if any of them have exited with an error or a panic, and if they have it returns the error as part of `poll_ready`.
Co-authored-by: Jane Lusby <jane@zfnd.org>
Prior to this change, the seed subcommand would consistently encounter a panic in one of the background tasks, but would continue running after the panic. This is indicative of two bugs.
First, zebrad was not configured to treat panics as non recoverable and instead defaulted to the tokio defaults, which are to catch panics in tasks and return them via the join handle if available, or to print them if the join handle has been discarded. This is likely a poor fit for zebrad as an application, we do not need to maximize uptime or minimize the extent of an outage should one of our tasks / services start encountering panics. Ignoring a panic increases our risk of observing invalid state, causing all sorts of wild and bad bugs. To deal with this we've switched the default panic behavior from `unwind` to `abort`. This makes panics fail immediately and take down the entire application, regardless of where they occur, which is consistent with our treatment of misbehaving connections.
The second bug is the panic itself. This was triggered by a duplicate entry in the initial_peers set. To fix this we've switched the storage for the peers from a `Vec` to a `HashSet`, which has similar properties but guarantees uniqueness of its keys.
- Add a total peers metric to prevent races between measurements of
ready/unready peers (which can cause the sum to be wrong).
- Add an outbound request counter.
* Don't expose submodules of zebra_network::peer.
* PeerSet, PeerDiscover stubs.
Co-authored-by: Deirdre Connolly <deirdre@zfnd.org>
* Initial work on PeerSet.
This is adapted from the MIT-licensed tower-balance implementation.
* Use PeerSet in the connect stub.
Alfredo Garcia
teor
Janito Vaqueiro Ferreira Filho
Jack Grigg
Jane Lusby
Henry de Valence
Jane Lusby