Make all operations that related to `neofs-api-go` library be placed in `v2`
packages. They parse all v2-versioned structs info `neofs-sdk-go`
abstractions and pass them to the corresponding `acl`/`eacl` packages. `v2`
packages are the only packages that do import `neofs-api-go` library. `eacl`
and `acl` provide public functions that only accepts `sdk` structures.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
- Update `neofs-sdk-go`:
v0.0.0-20211230072947-1fe37df88f80 => v0.0.0-20220113123743-7f3162110659
- Add client interface that duplicates SDK's client behaviour and new
`MultiAddressClient` interface that has method that iterates over wrapped
clients.
- Also start using simple client mode that does not require parsing statuses
outside the SDK library.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
For some data compression makes little sense, as it is already compressed.
This commit allows to leave such data unchanged based on `Content-Type`
attribute. Currently exact, prefix and suffix matching are supported.
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
It is much more convenient to skip source creation.
Also fix some bugs:
1. `cryptoSource.Int63()` now returns number in [0, 1<<63) as required
by `rand.Source` interface.
2. Replace `cryptoSource.Uint63()` with `cryptoSource.Uint64` to allow
generate uint64 numbers directly (see rand.Source64 docs).
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
In previous implementation ACL server was the 1st (except metric server
in some cases) server in pipeline of Object service servers. This led to
the fact that errors of this handler could not be reduced to status
responses.
Nest object ACL server into signature and response servers to support
common response format.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Add `subnet` sub-section to `node` section of storage node config. Add
`entries` value which allows to enumerate subnets for entrance. Add
`exit_zero` value which allows to not enter zero subnet.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Upgrade NeoFS API Go library to version with status returns. Make all API
clients to pull out and return errors from failed statuses. Make signature
service to respond with status if client version supports it.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Add hash of the TX that generated notification
to neofs/netmap event structures. Adapt all
neofs/netmap wrapper calls to new structures.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Rounding is done using the suffix boundary (i.e. floating point value
with `gb` suffix will return size which is an integer number of
megabytes).
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Relay storage node doesn't exec local object operations, so it doesn't need
shard configuration.
Add `required` bool parameter to `engineconfig.IterateShards`. Make it to
panic if it is `true`, and immediately return otherwise. Pass `false` if
node is configured as relay in app (it also prevents panic).
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to block execution of local object storage operations if
node is put into maintenance mode (resume if the node is taken out of
maintenance mode).
Call `BlockExecution` method if `ControlService.SetNetmapStatus` was called
with `MAINTENANCE` status. Call `ResumeExecution` if it was called with
another status.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
All objects in NeoFS must have owner ID. In previous implementation Object
Delete service handler set owner ID from request session token. If removal
was executed w/o a session, object with tombstone was prepared incorrectly.
In order to fix this node should set its own ID and become an owner of the
tombstone object.
Extend `NetworkInfo` interface required by Object.Delete handler with
`LocalNodeID` method which returns `owner.ID` of the local node. Implement
the method on `networkState` component of storage node application which is
updated on each node state change in NeoFS network map. Set owner returned
by `LocalNodeID` call as tombstone object's owner in Delete handler.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
The client needs of the Reputation service are limited and change not often.
Interface changes of the client library should not affect the operation of
various service packages, if they do not change their requirements for
the provided functionality. To localize the use of the base client and
facilitate further support, an auxiliary package is implemented that will
only be used by the Reputation service.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In case we have multiple sections with similar structure (e.g. shards)
having defaults in a single place is easier to work with.
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
There is no point to pass key storage in parameters because
it can be defined on the service level of application.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
```
> source config/example/node.env
config/example/node.env: строка 14: /dns4/s02.neofs.devenv/tcp/8081: Нет такого файла или каталога
config/example/node.env: строка 16: MSK: команда не найдена
config/example/node.env: строка 35: 028f42cfcb74499d7b15b35d9bff260a1c8d27de4f446a627406a382d8961486d6: команда не найдена
config/example/node.env: строка 48: https://rpc2.morph.fs.neo.org:40341: Нет такого файла или каталога
config/example/node.env: строка 49: wss://rpc2.morph.fs.neo.org:40341/ws: Нет такого файла или каталога
config/example/node.env: строка 53: https://rpc2.n3.nspcc.ru:30341: Нет такого файла или каталога
```
New implementation is still basic, but we don't need to parse any env
file, just have a correct example.
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Make the implementation of network info source (Netmap V2 service
dependency) to read MillisecondsPerBlock sidechain parameter and NeoFS
network parameters depending on the client version.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
SN tries to keep 1:3 proportion of GAS and
notary balances respectively. If that proportion
has been messed(means that notary balance is
lower than required) it sends half of its
GAS balance to the notary service.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
After storage engine started to limit number of PUT operations there is no
need to limited worker pool in Object Put service.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Add `shard_pool_size` config to `storage` section. Set app default to 20.
Pass the value to `WithShardPoolSize` option.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Do not log in options constructors. Also failure to
initialize compression module (possibly due to invalid options) is
certainly an error deserving proper treatment.
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
In previous implementation of reputation client constructor compared network
addresses to check the membership of the network map.
Replace network addresses comparison with public key comparison.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to have the ability to expand the data needed for client
construction.
Replace `network.AddressGroup` parameter of client cache interfaces with
`client.NodeInfo`.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation Object PUT used single pool of workers for local
and remote ops, but these ops are heterogeneous.
Use remote/local pool for remote/local operations in PUT service. At first
the pools are configured with the same size.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Implement `RefillMetabase` method to `shardconfig.Config` type which reads
`refill_metabase` config value. Pass the result to `WithRefillMetabase`
option.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Add `size_limit` config value to write-cache section. Add `SizeLimit` method
of `writecache.Config` type. Pass its value to `WithMaxCacheSize` option on
app construction.
Reflect config update in examples. Cover new value in unit test.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Bootstrap process involves `addPeer` contract
call => it is necessary to make deposit before
booting up.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Create non-notary container wrapper and
use it in load announcing since it shouldn't
be invoked as notary request.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Storage Node needs to have notary deposit
for successful notary request sending.
Add notary deposit on startup(and wait for
its acceptance). Add notary deposit timer,
its config in `morph` section and env vars
for its tuning.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Prepare all listening structures for notary events:
rename(add prefix/suffix 'notification') all
notification specific handlers/parsers.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Some software components regulate the way of working with placement arrays
when a local node enters it. In the previous implementation, the locality
criterion was the correspondence between the announced network address
(group) and the address with which the node was configured. However, by
design, network addresses are not unique identifiers of storage nodes in the
system.
Change comparisons by network addresses to comparisons by keys in all
packages with the logic described above. Implement `netmap.AnnouncedKeys`
interface on `cfg` type in the storage node application.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Since morph `Client` works in multi-client mode, there is an error case when
we can not get network magic when all endpoints are unavailable.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to work with a set of Neo RPC nodes in order not to depend
on the failure of some nodes while others are active.
Support "multi-client" mode of morph `Client` entity. If instance is not
"multi-client", it works as before. Constructor `New` creates multi-client,
and each method performs iterating over the fixed set of endpoints until
success. Opened client connections are cached (without eviction for now).
Storage (as earlier) and IR (from now) nodes can be configured with multiple
Neo endpoints. As above, `New` creates multi-client instance, so we don't
need initialization changes on app-side.
`Wait` and `GetDesignateHash` methods of `Client` return an error from now
to detect connection errors. `NotaryEnabled` method is removed as unused.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation Container service handlers didn't cache the
results of `Get` / `GetEACL` / `List` operations. As a consequence of this,
high load on the service caused neo-go client's connection errors. To avoid
this there is a need to use cache. Object service already uses `Get` and
`GetEACL` caches.
Implement cache of `List` results. Share already implemented cache of Object
service with the Container one. Provide new instance of read-only container
storage (defined as an interface)to morph executor's constructor on which
container service is based. Write operations remained unchanged.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Storage node should not try to register itself in network in relay mode.
Implement `needBootstrap` method which checks if node need to bootstrap.
Call `bootstrap` method in `bootstrapNode` function only on true return.
Skip re-bootstrap logic in new epoch event handler on false return.
Return an error if `ControlService.SetNetmapStatus` is called on relay
node.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
`reBootstrapEnabled` state var is not used. It is going to be used to decide
whether to bootstrap node or not.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Do not init caches for eACL, containers and netmap
if `disable_cache` config options is `true`, use
direct RPC calls instead.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Depending on having notary contract in sidechain get
IR list either from NeoFSAlphabet role either from
netmap contract.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Some users want to specify only wallet section in the SN. It is not
possible if `Key` throws panic on empty value. Instead it should
fallback to wallet section. Panic is suitable if node's key is provided
but invalid.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Changes:
* replace `iotuil` elements with the ones from `os` package;
* replace `os.Filemode` with `fs.FileMode`;
* use `signal.NotifyContext` instead of `NewGracefulContext` (removed).
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to support multiple network addresses of the storage nodes.
Make `BootstrapAddress` to return `network.AddressGroup` (and rename).
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Generalize single gRPC interface of the storage node to a group of
interfaces. Each interface calls the same RPC handler.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation if Control listen endpoint was omitted in config
the gRPC listening endpoint was used instead.
Consider empty address as an option to disable the service.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to support multiple server endpoints for reputation and
container transmission.
Replace `ServerInfo.Address` getter with `ServerInfo.IterateAddresses`
iterator.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make Object Head service to work with `AddressGroup` instead of `Address`
in order to support multiple addresses of the storage node.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make Object Search service to work with `AddressGroup` instead of `Address`
in order to support multiple addresses of the storage node.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make Object Get service to work with `AddressGroup` instead of `Address` in
order to support multiple addresses of the storage node.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make Object Put service to work with `AddressGroup` instead of `Address` in
order to support multiple addresses of the storage node.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make `LocalAddressSource.LocalAddress` method to return `AddressGroup`. Make
`IsLocalAddress` function to accept parameter of type `AddressGroup`. Adopt
the application code with temporary `GroupFromAddress` helper.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Change type of the `ClientCache.Get` method's parameter to `AddressGroup`.
Use `GroupFromAddress` to call the method from the wrappers in order to no
change their interface.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Add group-address `Client` implementation. Return instances of this
implementation from `ClientCache.Get` method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to generalize single-address client to group-address client.
To do this, we can re-implement `Client` interface from NeoFS API Go library
and still use it in the application code. There is a problem with method
`Raw` which must return single-address raw client. So as not to make changes
to API library we need to overload Client interface in order to support
`Raw` method in group-address client implementation.
Define `Client` interface in new `pkg/core/client` package. Completely
inherit API `Client` interface. Add `RawForAddress` method to build raw
client for the single node address. Adopt the application code that used Raw
method to work with new `Client`.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
`network.Address` structure in most cases created once and used read-only.
Replace `AddressFromString` function with `Address.FromString` method with
the same purpose and implementation. Make all libraries to work with value.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Implement `Address.WriteToNodeInfo` method which sets address of `NodeInfo`
structure. Use it in storage node application.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Also use neo-go private key wrapper where possible, as it
already has methods for (un)marshaling.
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Config `NodeInfo` should be used for bootstrap. Separete local node info and
netmap one. Return configured `NodeInfo` if structure from netmap is
missing.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation `Config.Sub` method could lead to the violation
of the internal `path` slice because of `append`. This has been observed on
deeply nested subsections.
Fix `Config.Sub` to copy internal slice in order to prevent violations.
Cover problem case in test config files and unit test.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Dedicated type `LoggerSection` turned out to be redundant since it doesn't
do a hidden logic and just uses `config.Config` API.
Remove `LoggerSection` type and implement `Level` which do the same.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation ENV values were read only if config filepath is
specified. From now ENVs are always read to `Config`.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In some cases viper doesn't interpret `section.value` as a subsection with
`section` name, but value is value still can be accessed through full
pathname.
Fix `Config.Sub` method implementation in order to always interpret
configuration like described above as a subsection. From now method never
returns nil, therefore an additional check has been removed from the `Value`
method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Create `logger` sub-package of `config` package. Implement `LoggerSection`
type of logger sub-section. Add `Level` method to read logger level config
value. Default level is `info`.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Create `config` package nearby storage node application. Implement `Config`
as a wrapper over `viper.Viper` that provides the minimum functionality
required by the application.
The constructor allows you to read the config from the file. Methods are
provided for reading subsections and values from the config tree. Helper
functions are implemented to cast a value to native Go types.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
