Added sync logs for every service separately to provide the ability to
have a custom logger for each service. This commit makes the code follow
the zap usages rules: `Sync calls the underlying Core's Sync method,
flushing any buffered log entries. Applications should take care to
call Sync before exiting.`
Signed-off-by: Ekaterina Pavlova <ekt@morphbits.io>
Start of some services is bound to blockchain subscriptions, and thus,
can't be run before the blockchain notifications dispatcher.
Signed-off-by: Ekaterina Pavlova <ekt@morphbits.io>
Perform initialisation of StateRoot service with designated
StateValidator's node list in the service constructor. There's no need
to wait until the next role update event, and it may lead to inaccuraces
in service work on SIGHUP/server restart.
A part of #3228.
Signed-off-by: Anna Shaleva <shaleva.ann@nspcc.ru>
go.uber.org/atomic deprecated CAS methods in version 1.10 (that introduced
CompareAndSwap), so we need to fix it.
Signed-off-by: Roman Khimov <roman@nspcc.ru>
Everywhere including examples, external interop APIs, bindings generators
code and in other valuable places. A couple of `interface{}` usages are
intentionally left in the CHANGELOG.md, documentation and tests.
And include some node-specific configurations there with backwards
compatibility. Note that in the future we'll remove Ledger's
fields from the ProtocolConfiguration and it'll be possible to access them in
Blockchain directly (not via .Ledger).
The other option tried was using two configuration types separately, but that
incurs more changes to the codebase, single structure that behaves almost like
the old one is better for backwards compatibility.
Fixes#2676.
It's more generic and convenient than MillisecondsPerBlock. This setting is
made in backwards-compatible fashion, but it'll override SecondsPerBlock if
both are used. Configurations are specifically not changed here, it's
important to check compatibility.
Fixes#2675.
Blockchain's notificationDispatcher sends events to channels and these
channels must be read from. Unfortunately, regular service shutdown procedure
does unsubscription first (outside of the read loop) and only then drains the
channel. While it waits for unsubscription request to be accepted
notificationDispatcher can try pushing more data into the same channel which
will lead to a deadlock. Reading in the same method solves this, any number of
events can be pushed until unsub channel accepts the data.
They test both module and service which is a bit wrong, but separating these
tests will lead to some duplication, so it's OK for now to have them in the
higher-order package (service imports module).
Notice that it makes the node accept Extensible payloads with any category
which is the same way C# node works. We're trusting Extensible senders,
improper payloads are harmless until they DoS the network, but we have some
protections against that too (and spamming with proper category doesn't differ
a lot).
These are arbitrary numbers, but that's what C# node uses. maxRetries could be
set lower because we have exponential backoff anyway, but this works too.
And don't add/resend it multiple times.
1. We can be in a setup with one SV only and no AddSignature() called at all.
2. AddSignature() might add M-1 signatures and our signature should be the
last one to complete MPTRoot, but we'll never do that.
It's only used to sign/verify it and is not a part of the structure. It's
still neded in consensus.Payload though because that's the way dbft library
is.
