Tombstone and "alive" objects can be both stored in BlobStor. They can
appear during iterating in different order. Metabase returns
`ErrAlreadyRemoved` error if object is inhumed.
Ignore `object.ErrAlreadyRemoved` errors of `metabase.Put`in Shard's
`refillMetabase` operation.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to refill Metabase data with the objects from BlobStor.
Implement `refillMetabase` method which iterates over all objects from
BlobStor and saves them in Metabase.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to be able to process all objects saved in `BlobStor`.
Implement `BlobStor.Iterate` method which iterates over all objects.
Implement `IterateBinaryObjects` and `IterateObjects` helper functions to
simplify the code.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to be able to process all stored objects saved in
`Blobovnicza`.
Implement `Blobovnicza.Iterate` method which iterates over all objects.
Implement `IterateObjects` helper function to simplify the code.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In the previous implementation of the metabase, there was no possibility of
reinitializing the metabase: clearing information about existing objects and
bringing it back to its initial state. This operation can be useful in
cases when the stored metadata about objects has lost (or possibly lost)
relevance, and you need to generate data from scratch. Also at the
initialization stage, static resources of the base were not created -
container-independent buckets.
Make `Metabase.Init` method to allocate graveyard, container-size and
to-move-it buckets in underlying BoltDB instance. Implement `Metabase.Reset`
method: it works like `Init` but clean up all static buckets and removes
other ones. Due to the logical similarity, the methods share a single piece
of code.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to limit disk space used by write-cache. It is almost
impossible to calculate the value exactly. It is proposed to estimate the
size of the cache by the number of objects stored in it.
Track amounts of objects saved in DB and FSTree separately. To do this,
`ObjectCounters` interface is defined. It is generalized to a store of
numbers that can be made persistent (new option `WithObjectCounters`). By
default DB number is calculated as key number in default bucket, and FS
number is set same to DB since it is currently hard to read the actual value
from `FSTree` instance. Each PUT/DELETE operation to DB or FS
increases/decreases corresponding counter. Before each PUT op an overflow
check is performed with the following formula for evaluating the occupied
space: `NumDB * MaxDBSize + NumFS * MaxFSSize`. If next PUT can cause
write-cache overflow, object is written to the main storage.
By default maximum write-cache size is set to 1GB.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to keep track of each local storage change. Log messages are
the most convenient way to do it.
Implement function which writes log message about the completed writing
operation in storage engine.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Do not pass high level `PACK` opcode to
notary parsers. Add opcode amount check.
Delete `PACK` cases in notary parsers.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Make `errIncompletePut` to be a structure which wraps single client error.
Wrap error of the last client into `errIncompletePut` during placement
execution.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation Object service's handler returned const error in
case of failure (full or partial) of PUT operation. This did not even allow
us to roughly guess what the reason is. Not as a complete solution, but to
alleviate some cases where all nodes in a container return the same error,
it is suggested to return the error of the last server that responded.
Return latest server error from placement loop of `iteratePlacement` method
of `distributedTarget` type.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation sticky bit could disrupt access of container
nodes to replication. According to NeoFS specification sticky bit should not
affect the requests sent by nodes from SYSTEM group.
Add role check to `stickyBitCheck`.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Remote RPC node might be in the transition state and produce
events from the past. We should avoid listening such nodes.
To do that subscriber component can await minimal height of
remote node. If minimal height is not reached, then throw
error.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Implement `NotaryContractProcessor` by IR
container processor. Add support for notary
`put` container operation. Do not parse `put`
non-notary notifications in notary enabled
environment.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Add `NotarySignAndInvokeTX` method to morph
client. This function allows invoking notary
request with passed main TX(not creating a
new one). It signs passed main TX with
client's key.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Add `NotaryInvokeNotAlpha` to low-level
client. It creates and sends notary request
that must be signed by Alphabet nodes, but
does not sign it by current node's private
key.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Add preparator for notary requests. Is parses
raw notary requests, checks if it should be
handled by Alphabet node. If handling is required,
returns `NotaryEvent` that contains information
about contract scripthash, method name and
arguments of the call.
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>
`fallbackTime` is delta b/w `ValidUntilBlock` of
the main transaction and block when `fallback`
transaction is sent.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
In recent changes, the locality criterion for a node has been changed to
compare public keys.
Remove no longer used `IsLocalAddress` function and `LocalAddressSource`
interface.
Signed-off-by: Leonard Lyubich <leonard@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>
Announced keys of storage nodes are required for many application components
to function.
Define a unified interface for the utility for working with public keys of
nodes. Add a method to check if the key has been advertised by the local
node in the application context.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is a need to process public keys of the placement result.
Implement `Node.PublicKey` method which returns storage node's key announced
in netmap.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation `placement.Traverser.Next` method returned slice
of `network.AddressGroup` elements. There is a need to process keys of
storage nodes besides network addresses for intra-container communication.
Wrap `network.AddressGroup` in a new type `placement.Node` that summarizes
the storage node information required for communication. Return slice of
`Node` instances from `Traverser.Next` method. Fix compilation breaks in
dependent packages.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
IDs come from NeoFS contract in big endian, but it is customary to write in
the node logs in little endian.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Shard should try to read object headers from write-cache if it is enabled.
Extend `writecache.Cache` interface with `Head` method. Call the method in
`Shard.Head` if `Shard.hasWriteCache` returns true.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Write cache should be able to execute HEAD operations according to spec.
Add simple implementation of `Head` method through the `Get` one. Leave
notes for future optimization.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Caching is performed inside `GetNativeContractHash` method of neo-go client,
so the additional cache level is redundant.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is no need to continue iterating over Neo RPC endpoints in case of
some address-independent error (e.g. NeoFS logic error).
Unwrap and immediately return `neofsError` errors from loop in
`iterateClients`.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Implement `error` interface on new `neofsError` type which is a wrapper over
NeoFS-specific error. Wrap all `Client` errors except neo-go client API ones
into `neofsError`. Wrapped errors are going to be used for multi-endpoint
loop control.
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>
As soon as resulting list and cache operate with the attribute pointer,
we can add attribute structure immediately and set restored key and
value of the attribute later.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Attributes are linked to each other through parents, so they can
be returned in any order. However, it will be better to return
the list in consistent order to reduce entropy.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Parser should reuse existing attributes from the cache to update list
of the parent links in it. Parent links should be unique.
Signed-off-by: Alex Vanin <alexey@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>
In previous implementation of Policer's job queue the same list of objects for
processing was selected at each iteration. This was caused by consistent
return of `engine.List` function.
Use `rand.Shuffle` function to compose pseudo-random list of all objects in
order to approximately evenly distribute objects to work.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
To encode attributes with semicolon or slash, use
backslash as escaped character.
Example:
User-Agent:NeoFS\/0.23
Signed-off-by: Alex Vanin <alexey@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>
Do not read `without_notary` config value from env.
Make morph client constructor return client without
notary support. Enabling notary support should be done
with public `EnableNotarySupport` method separately.
Notary availability is deducted with client. Further,
if notary is presented on chain its support is
enabled at the corresponding client.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Add `MkdirAllX` function that wraps `os.MkdirAll`
and adds `x` permission for user and group.
This is used to make all created dirs openable
for user and group regardless of the read config
permissions.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
When Inner Ring node starts, it should sync nearest epoch tick
event based on the block of the latest epoch. Otherwise epoch
ticking can be stopped, because ballots or notary transactions
are valid for limited period of time.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Get innerring list and committee in innerring
indexer from interfaces, not from structs.
Passing IR fetcher to indexer constructor depends
on having Notary contract in sidechain(use NeoFS
role or netmap contract).
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Add `IrFetcherWithNotary` and `IrFetcherWithoutNotary`
that can fetch IR keys with and without usage of
notary contract.
Both can be hidden behind
`InnerRingKeys() (keys.PublicKeys, error)` interface.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Add initialization of `updateInnerRing` method name
in config on startup.
Rename var `setInnerRing` => `updateInnerRing`.
Rename method `SetInnerRing` => `UpdateInnerRing`
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
We should go through every key in main chain list to merget lists
as fast as possible. Previously we drop main chain traversing as
soon as we have no more new keys to add. Instead we should try
to go for old keys in the list and add it as more as we can.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Add sorting of passed public keys in `newAlphabetList`
function before Alphabet list merge to make returned
results more expected.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
If `WITHOUT_MAINNET` environmental variable is
`true`:
- Do not read `NeoFS` and `processing`
script-hashes from envs;
- Do not init Governance processor;
- Do not init NeoFS processor.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Storage node should only reflect container Put requests to sidechain w/o any
validation. All verification steps are performed by IR.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Container can have version later than the current lib version, and it should
not be considered as invalid. But it should still have adequate version.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Create `version` package and implement `IsValid` function which checks if
version is not earlier than the genesis one.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Audit client is used to send audit results and they should be
signed by inner ring node itself on order to be saved in
smart contract.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Add `netmapCandidates` method to `netmap` client
wrapper. Method parses node storages candidates
for the next epoch.
Signed-off-by: Pavel Karpy <carpawell@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>
Make Netmap service to return single network address on `LocalNodeInfo` RPC
in order to older clients.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Implement `sort.Interface` interface on `AddressGroup` and perform sorting
in `AddressGroup.FromIterator` method. Addresses with enabled TLS are "less"
in terms of slice position.
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>
In future server info will contain multiple endpoints whose string
representation will be worse suited to the cache key.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In future server info will contain multiple endpoints whose string
representation will be worse suited to the cache key.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make placement `Traverser.Next` method to return ``[]network.AddressGroup`
in order to support multiple addresses of the storeage nodes.
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>
Define `network.AddressGroup` type which represents group of network
addresses. This type is going to be used to support group-address of the
storage node.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Values of object attributes must not be empty according to NeoFS
specification.
Make `FormatValidator.Validate` method to return an error if at least one
attribute has empty value.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Keys of object attributes must be unique according to NeoFS specification.
Make `FormatValidator.Validate` method to return an error if at least one
attribute is duplicated.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Add `CompositeValidator` that wraps
`netmap.NodeValidator`s and implements
`NodeValidator` interface itself.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Validation checks:
1. if address can be parsed by network package;
2. if address contains correct amount of protocols;
3. if address's protocols are in correct order.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
The only thing we need hashes for is to process notifications.
Balance contract if left for now, as it has some initialization.
Signed-off-by: Evgenii Stratonikov <evgeniy@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>
Remove `Encapsulate`, `Decapsulate` and `IPAddrString` methods of `Address`
since they are used in unit tests only.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Return tyype is clear from the method's signature and docs, there is no
point in reflecting it in the name.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Panic if internal `manet.DialArgs` call returns error since this is
unexpected according to `AddressFromString` implementation.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Add sanity checks of GlobalTrust value:
check if "got manager" is real manager for
peer with building managers for peer.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Test invocations are used in `Invoke` method to calculate
consumed gas. We can check return code and return error
if panic happened in contract.
This is already done the same way in `TestInvoke` method.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Serve `ControlService` instance on configured endpoint (do not serve if not
specified). Read allowed keys from config.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Set health status to
* `UNDEFINED` during Server construction;
* `STARTING` on `Server.Start` call;
* `READY` after successful `Server.Start` call;
* `SHUTTING_DOWN` on `Server.Stop` call.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Implement `ControlServiceServer` on `Server` type. The `Server` requires all
requests to be signed with keys from the so-called whitelist. To obtain
health status, it uses the abstraction in the form of `HealthChecker`
interface.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Define `ControlService` for IR similar to the one from storage node. Add
`HealthStatus` RPC which returns health status of the IR application.
Implement getters, setters and methods to sign/verify the messages.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make `Validator.VerifyAndUpdate` method to return an error if at least one
of LOCODE-derived attributes is set explicitly. Thus, IR will not confirm
the candidates for the network map who independently set these attributes.
Cover `Validator.VerifyAndUpdate` method with unit tests.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
NeoFS network dictates the limitation of the maximum size of the "finished"
objects. The payload size of an object is recorded in its header. When
writing finished objects, the nodes must check:
* satisfying the constraint;
* matching the value in the header to the number of payload bytes.
Provide value returned by `MaxSizeSource` component to `validatingTarget`.
Check max size value during the stream of the "finished" objects. Check
header value during the streaming and on-close. Check payload size in v2
relay scenario.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Session tokens have limited lifetime in NeoFS. Container processor should
verify lifetime of the incoming tokens.
Define `NetworkState` interface with `Epoch` method to get number of the
current epoch. Use Netmap contract client's wrapper as `NetworkState` of
Container `Processor`. Check values of token lifetime, and deny if:
* NBF value is gt the current epoch;
* IAT is gt the current epoch;
* EXP is le the current epoch.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make IR processor of NeoFS contract to handle `Bind`/`Unbind` notification
events. The processor verifies the format of wallet script hash and public
keys, and call NeoFS ID client wrapper in order to approve adding/removing
keys from NeoFS account.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Pass `TryNotary()` option to constructor of the static client of the NeoFS
ID contract in `NewFromMorph`. This will allow to use client wrapper for key
management in IR application.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Morph event structures defined in `pkg/morph/event` should only carry
notification values without any additional interpretation. All logical work
should be concentrated on app-side. Change `Bind.User` / `Unbind.User` to
return byte slice. Change `Bind.Keys` / `Unbind.Keys` to return `[][]byte`.
`ParseBind` / `ParseUnbind` don't decode data from now.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Implement method `ClientWrapper.ManageKeys` method which provides the
interface to add/remove keys to/from NeoFS account.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Implement `AddKeys` / `RemoveKeys` methods to call `addKey` / `removeKey`
methods of NeoFS ID contract.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Implement wrapper over NeoFS contact's client which allows you to
conveniently interact with the contract. Implement `ManageKeys` method
for binding or unbinding public keys to the NeoFS account.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Implement NeoFS contact's client which is responsible for collecting call
arguments and parsing stack items. Initially key binding and unbinding are
supported.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Update API Go library with introduce `Client.Conn` method. Implement
`ClientCache.CloseAll` method which reads and closes connections of all
cached clients.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In recent change of API Go library `owner.ID.Equal` signature was
implemented.
Replace the comparison of string representations with `Equal` method call
and remove related TODOs.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In recent change of API Go library `Token.Verify` signature was implemented.
Replace previous version-casting approach with new method call in token
signature check stage.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Token of the container session should be written out with container context.
The context should have the verb corresponding to the operation. If an
operation is performed on a fixed container, the session should be
propagated to it or to all user containers
Implement all described checks in validation of `Put` / `Delete` / `SetEACL`
events.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation verification of `SetEACL` events failed on events
without session token. It was caused by redundant tries to verify `nil`
session token.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Session token can be presented `Put`, `Delete` and `SetEACL` notification
events. IR should consider this case as issuing a power of attorney to a
third party. Thus, checking the eligibility for an operation should be
complicated:
- token owner should be the owner of the related container;
- the intent must be signed with a session key;
- the power of attorney must be signed by the owner of the container.
Omitted checks (TBD):
- session token should have container session context;
- the verb of the context should correspond to the operation.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There is no TLS protocol support in
`go-multiaddr` library, but there is
public function that can register any
protocol that can be implemented outside
the library. Also `TLSEnabled` function
for parsing TLS protocol from
`network.Address` was added.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
`bindKey` parameter of `Put` method of `Container` contract was removed in
latest version.
Do not pass bind key of type `[]byte` to `Put` invocation. Remove no longer
needed field from `PutArgs`.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
From `017fb6abed9455c7c99631adcb0bb04d42741f87` used container client is
constructed with enabled notary mode.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
If container is removed via session, then session token should be included
in removal witness.
Write session token from request meta header to `container.RemovalWitness`
structure which is passed to `wrapper.Delete` function.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Approved removed container ID should be stored in sidechain along with
related session token.
Forward session token from `Delete` event to `Wrapper.Delete` method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
`Delete` method of latest `Container` contract accepts binary session token
as an argument.
Provide `DeleteArgs.SetSessionToken` method. Accept session token as a
`[]byte` in `Wrapper.Put` method and attach it to `PutArgs`. Marshal session
token from `RemovalWitness` in `wrapper.Delete` function and pass it to the
method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
The 3rd item of `Delete` container notification event is a byte array of
serialized session token. Parse session token in `ParseDelete` function.
Provide `Delete.SessionToken` method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
NeoFS containers can be removed within a trusted session. There is a need to
take this into account during removal inspection.
Define `SessionToken` / `SetSessionToken` methods on `RemovalWitness` struct
in order to embed `session.Token` to it.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make `wrapper.Delete` function to accept `container.RemovalWitness` struct
instead of its separated elements. `Signature` type is replaced by binary
signature since public key is unused.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Define `RemovalWitness` structure which groups the information required to
prove and verify the removal of a container. This type is going to be used
in container Delete-related methods.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Unmarshal session token from `EACLValues` and write it to resulting
`eacl.Table` structure in `Wrapper.GetEACL` method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
`EACL` method of `Container` contract returns binary session token, key and
signature along with eACL table.
Provide `Signature`, `PublicKey` and `SessionToken` getters from
`EACLValues` structure. Parse and set all values in `Client.EACL` methods.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Approved eACL table should be stored in sidechain along with related session
token.
Forward session token from `SetEACL` event to `Wrapper.PutEACL` method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
The 4th item of `SetEACL` container notification event is a byte array of
serialized session token.
Parse session token in `ParseSetEACL` function. Provide
`SetEACL.SessionToken` method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
If eACL table is set via session, then session token should be written to
it.
Write session token from request meta header to `eacl.Table` structure which
is passed to `wrapper.PutEACL` function.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Rename `PutContext` to `ContextWithToken` and implement its constructor as a
separate function in order to reuse it in other RPCs.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
`SetEACL` method of latest `Container` contract accepts binary session token
as an argument.
Provide `SetEACLArgs.SetSessionToken` method. Accept session token as a
`[]byte` in `Wrapper.PutEACL` method and attach it to `SetEACLArgs`. Marshal
session token from container in `wrapper.PutEACL` function and pass it to
the method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Unmarshal session token from `GetValues` and write it to resulting
`Container` structure in `Wrapper.Get` method. Write key-signature pair from
`GetValues` to resulting `Container` structure in `Wrapper.Get` method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
`Get` method of `Container` contract returns binary session token, key and
signature along with container.
Provide `Signature`, `PublicKey` and `SessionToken` getters from `GetValues`
structure. Parse and set all values in `Client.Get` methods.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Approved container should be stored in sidechain along with related session
token.
Forward session token from `Put` event to `Wrapper.Put` method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
The 4th item of `Put` container notification event is a byte array of
serialized session token.
Parse session token in `ParsePut` function. Provide `Put.SessionToken`
method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
If container is created via session, then session token should be written to
it.
Write session token from request meta header to `Container` structure which
is passed to `wrapper.Put` function.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
`Put` method of latest `Container` contract accepts binary session token as
an argument.
Provide `PutArgs.SetSessionToken` method. Accept session token as a `[]byte`
in `Wrapper.Put` method and attach it to `PutArgs`. Marshal session token
from container in `wrapper.Put` function and pass it to the method.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation wrapper over the Container contract's client
accepted the signature of the eACL table in addition to itself. After recent
changes in API Go lib table carries its signature. Thus, it is redundant
to pass the eACL table signature separately.
Make `wrapper.PutEACL` method to accept `eacl.Table` only.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation wrapper over the Container contract's client
accepted the signature of the container in addition to itself. After recent
changes in API Go lib container carries its signature. Thus, it is redundant
to pass the container signature separately.
Make `wrapper.Put` method to accept `Container` only.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In previous implementation wrapper over the Container contract's client
returned the signature of the eACL table in addition to itself. After recent
changes in API Go lib table carries its signature. Thus, it is redundant to
return the table signature separately.
Make `Wrapper.GetEACL` method to return only `eacl.Table` with error.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In recent API changes `GetExtendedACLResponseBody` carries session token.
In recent API Go lib changes `eacl.Table` structure carries related session
token.
Write session token of eACL table from sidechain to GetExtendedACL response
body in node's `ContainerService` server.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
In recent API changes `GetResponseBody` carries session token and signature.
In recent API Go lib changes `Container` structure carries its session
token and signature.
Write session token and signature of container from sidechain to Get
response body in node's `ContainerService` server.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
As it explained in previous commit, session token also should
be presented in original meta header but can be omitted in higher
layers.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Request meta headers are organized in a layers, where
upper layers re-sign down layers. Bearer token should be
a part of original meta header and it can be omitted in
upper layers. Therefore we need to traverse over linked list
of meta header to the original meta header to get bearer token.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
`eACL` method of `Container` contract returns structure with 4 fields (the
4th is a recently added binary session token).
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Change expected event items to 4 in order to support new binary session
token item. Parsing of the session token will be implemented later.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Construct wrapper over the Container contract client with `TryNotary` option
since it is required to perform invocations of notary contract.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
All client wrappers should use underlying static client with enabled notary
work mode in order to produce invocations of notary contract.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Some of the client wrapper's methods should produce notary contract's
invocations. In previous implementation all wrappers provided separate
methods to do it. Since notary and non-notary invocation scenarios have very
different goals, it makes sense to separate the scenarios of using the
client wrapper at the stage of its creation.
Define `Option` constructor for container client wrapper. Add `TryNotary`
option which enables tries of the notary invocations on underlying static
client. Mark all notary-dedicated methods as deprecated.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
There are two scenarios of invocation of contract methods:
1. do not invoke notary contract;
2. try to invoke notary contract if it is enabled in Client.
Taking this into account, `StaticClient` can work in one of the two described
modes. Based on this, it makes sense at the stage of creating `StaticClient`
to fix the call mode, and the further abstract from it.
Define `StaticClientOption` setters of `StaticClient` optional parameters.
Add `TryNotary` constructor of option which enables notary tries. Call
`NotaryInvoke` on underlying `Client` if the option is provided, otherwise
call `Invoke`. Mark `NotaryInvoke` method of `StaticClient` as deprecated.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Pass session token (byte array) and bind key flag (boolean) arguments to
`Put` method call of `Container` contract.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Change expected event items to 4 in order to support new binary session
token item. Parsing of the session token will be implemented later.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Owner identifier can be calculated from public key. If it matches, no
additional verification of key ownership is required.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Get all owner keys and verify container ID signature until first success. If
none of the keys match, then prohibit deletion. Thus, the delete operation
is only allowed to the owner of the container. With this approach, a
separate check for key ownership is not required.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Make `Get` method of the wrapper over Container contract's client to
accept binary container ID. Create `Get` function similar to the previous
`Get` variation. Use this function in Container service server in the place
where `Get` method was used.
Additionally implement `AsContainerSource` function which allows
to simply compose container Source interface from the wrapper.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Call `Delete` method on the wrapper over the Container contract's client
directly from `Processor.approveDeleteContainer`.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
