* Store masternodes in a map instead of a vector, drop unused functions in CMasternodeMan
* CTxIn vin -> COutPoint outpoint
* do not use CMasternodeMan::Find outside of the class
* update GetMasternodeInfo
* safe version of GetNextMasternodeInQueueForPayment
* fix ProcessMasternodeConnections
* bump CMasternodeMan::SERIALIZATION_VERSION_STRING
* Dont deserialize nVersion into CNode, should fix#9212
* net: deserialize the entire version message locally
This avoids having some vars set if the version negotiation fails.
Also copy it all into CNode at the same site. nVersion and
fSuccessfullyConnected are set last, as they are the gates for the other vars.
Make them atomic for that reason.
* net: don't run callbacks on nodes that haven't completed the version handshake
Since ForEach* are can be used to send messages to all nodes, the caller may
end up sending a message before the version handshake is complete. To limit
this, filter out these nodes. While we're at it, may as well filter out
disconnected nodes as well.
Delete unused methods rather than updating them.
* net: Disallow sending messages until the version handshake is complete
This is a change in behavior, though it's much more sane now than before.
* net: log an error rather than asserting if send version is misused
Also cleaned up the comments and moved from the header to the .cpp so that
logging headers aren't needed from net.h
* Implement conditions for ForEachNode() and ForNode() methods of CConnman.
A change making ForEachNode() and ForNode() methods ignore nodes that
have not completed initial handshake have been backported from Bitcoin.
Unfortunately, some Dash-specific code needs to iterate over all nodes.
This change introduces additional condition argument to these methods.
This argument is a functional object that should return true for nodes
that should be taken into account, not ignored.
Two functional objects are provided in CConnman namespace:
* FullyConnectedOnly returns true for nodes that have handshake completed,
* AllNodes returns true for all nodes.
Overloads for ForEachNode() and ForNode() methods without condition argument
are left for compatibility with non-Dash-specific code.
They use FullyConnectedOnly functional object for condition.
Signed-off-by: Oleg Girko <ol@infoserver.lv>
* Iterate over all nodes in Dash-specific code using AllNodes condition.
Use AllNodes functional object as newly introduced condition argument for
ForEachNode() and ForNode() methods of CConnman to iterate over all nodes
where needed in Dash-specific code.
Signed-off-by: Oleg Girko <ol@infoserver.lv>
* Change sync process:
- IsBlockchainSynced(): drop CheckNodeHeight() and all complicated code, use fInitialDownload in UpdatedBlockTip() to switch initial states
- ProcessTick(): detect sleep mode like it was in IsBlockchainSynced(), not by number of masternodes
* Changes for sync in governance:
- do not keep sync alive on ConfirmInventoryRequest()
- skip some governance actions until we are synced to some level
* do not run CMasternodeMan::UpdateLastPaid() until winners list is synced
* start syncing mn list on the same node right after requesting sporks
* replace nTimeLast<Asset> with the unified nTimeLastBumped, bump on UpdatedBlockTip
* fix comments and LogPrintf-s
* remove excessive MASTERNODE_SYNC_IBD
* a bit more descriptive BumpAssetLastTime in few cases
* serialization: teach serializers variadics
Also add a variadic CDataStream ctor for ease-of-use.
* connman is in charge of pushing messages
The changes here are dense and subtle, but hopefully all is more explicit
than before.
- CConnman is now in charge of sending data rather than the nodes themselves.
This is necessary because many decisions need to be made with all nodes in
mind, and a model that requires the nodes calling up to their manager quickly
turns to spaghetti.
- The per-node-serializer (ssSend) has been replaced with a (quasi-)const
send-version. Since the send version for serialization can only change once
per connection, we now explicitly tag messages with INIT_PROTO_VERSION if
they are sent before the handshake. With this done, there's no need to lock
for access to nSendVersion.
Also, a new stream is used for each message, so there's no need to lock
during the serialization process.
- This takes care of accounting for optimistic sends, so the
nOptimisticBytesWritten hack can be removed.
- -dropmessagestest and -fuzzmessagestest have not been preserved, as I suspect
they haven't been used in years.
* net: switch all callers to connman for pushing messages
Drop all of the old stuff.
* drop the optimistic write counter hack
This is now handled properly in realtime.
* net: remove now-unused ssSend and Fuzz
* net: construct CNodeStates in place
* net: handle version push in InitializeNode
* net: move CBanDB and CAddrDB out of net.h/cpp
This will eventually solve a circular dependency
* net: Create CConnman to encapsulate p2p connections
* net: Move socket binding into CConnman
* net: move OpenNetworkConnection into CConnman
* net: move ban and addrman functions into CConnman
* net: Add oneshot functions to CConnman
* net: move added node functions to CConnman
* net: Add most functions needed for vNodes to CConnman
* net: handle nodesignals in CConnman
* net: Pass CConnection to wallet rather than using the global
* net: Add rpc error for missing/disabled p2p functionality
* net: Pass CConnman around as needed
* gui: add NodeID to the peer table
* net: create generic functor accessors and move vNodes to CConnman
* net: move whitelist functions into CConnman
* net: move nLastNodeId to CConnman
* net: move nLocalHostNonce to CConnman
This behavior seems to have been quite racy and broken.
Move nLocalHostNonce into CNode, and check received nonces against all
non-fully-connected nodes. If there's a match, assume we've connected
to ourself.
* net: move messageHandlerCondition to CConnman
* net: move send/recv statistics to CConnman
* net: move SendBufferSize/ReceiveFloodSize to CConnman
* net: move nLocalServices/nRelevantServices to CConnman
These are in-turn passed to CNode at connection time. This allows us to offer
different services to different peers (or test the effects of doing so).
* net: move semOutbound and semMasternodeOutbound to CConnman
* net: SocketSendData returns written size
* net: move max/max-outbound to CConnman
* net: Pass best block known height into CConnman
CConnman then passes the current best height into CNode at creation time.
This way CConnman/CNode have no dependency on main for height, and the signals
only move in one direction.
This also helps to prevent identity leakage a tiny bit. Before this change, an
attacker could theoretically make 2 connections on different interfaces. They
would connect fully on one, and only establish the initial connection on the
other. Once they receive a new block, they would relay it to your first
connection, and immediately commence the version handshake on the second. Since
the new block height is reflected immediately, they could attempt to learn
whether the two connections were correlated.
This is, of course, incredibly unlikely to work due to the small timings
involved and receipt from other senders. But it doesn't hurt to lock-in
nBestHeight at the time of connection, rather than letting the remote choose
the time.
* net: pass CClientUIInterface into CConnman
* net: Drop StartNode/StopNode and use CConnman directly
* net: Introduce CConnection::Options to avoid passing so many params
* net: add nSendBufferMaxSize/nReceiveFloodSize to CConnection::Options
* net: move vNodesDisconnected into CConnman
* Made the ForEachNode* functions in src/net.cpp more pragmatic and self documenting
* Convert ForEachNode* functions to take a templated function argument rather than a std::function to eliminate std::function overhead
* net: move MAX_FEELER_CONNECTIONS into connman
* Only store and connect to NODE_NETWORK nodes
* Keep addrman's nService bits consistent with outbound observations
* Verify that outbound connections have expected services
* Don't require services in -addnode
* Introduce enum ServiceFlags for service flags
* Introduce REQUIRED_SERVICES constant
* Add hassentinelping to governanceinfo
* sentinelping rpc call
* additional fields in mnp
* sentinel ping implementation
* change sentinel state to byte in mnp
* use adjusted time in sentinel ping
* update nTimeLastWatchdogVote if sentinel ping is actual
* remove unused fields
* bump protocol to 70207
* Fix small issues
- fix the error message text in CActivbeMasternodeUpdateSentinelPing;
- add empty string before public: in CActiveMasternode class declaration;
- rename field sentinelPing in CMasternodePing to sentinelIsActual and change $
- decrease sentinelVersion field size to uint16_t;
* revert proto bump for MIN_... consts
* revert changes in getgovernanceinfo
* Update mn vote time for remote masternodes
- call UpdateWatchdogVoteTime in CMasternodeMan::ProcessMessage
- deserialize masternodeping from the previous version archive without exception
- add ability to set time in UpdateWatchdogVoteTime
- set nTimeLastWatchdogVote to masternode ping sigTime if sentinel is actual
- bump CMasternodeMan::SERIALIZATION_VERSION_STRING
* remove mn state checks and add correct rpc param convertion
* fix var names
* Helper class for version in string and integer form
* String version in sentinel ping
Version format is "x.x.x"
* test for bacward compatibility in serialization
* Change VersionInfo class to convert functions
* fix verification network behaviour:
- use up to 10 connections only
- save state (do not ask same nodes after restart)
- initiate on time based intervals (every 5 minutes), based not on new blocks (should keep connections alive longer)
* revert mWeAskedForVerification serialization
* send verify requests using MAX_POSE_CONNECTIONS as a step
* Flag governance items when MN's are removed
* Remove old broken update mechanism
* Do not flag MN state changes which are irrelevant to validity
* Call AddGovernanceVote
We should remove hash from setAskFor when the message corresponding to previous inv arrives, otherwise it's stays there forever and setAskFor overflows (i.e. AskFor returns immediately without processing).
Verification creates new conections, which we also use to sync but then we disconnect on ProcessMasternodeConnections every minute which at least could lead to excessive node load (for our peers). But could also interfere with sync process itself and we'd better keep them separate.
* Store time we saw mnb last time, bump sync timeout
if we received seen mnb but we are too close to MASTERNODE_NEW_START_REQUIRED_SECONDS
* Reset blockchain sync status if new blocks were accepted during sync
* Add some debug log output
* wait for at least one new block to be accepted
* bump CGovernanceManager-Version
* Few mn list sync changes:
- add more mn states
- only remove spent
- send mn ping in addition to mn announce on sync
- manage mn announces more carefully
- expire mns created from broadcasts with invalid ping
- old fWaitForPing logic for old nodes, should be helpful during migration period
This also adjusts active mn auto-start logic accordingly.
Should also store/check node we asked for mn list entry, not outpoint only. This should help to get mn list in sync when some nodes refuse to answer thus blocking such requests for the same outpoint for the next 3h and increasing mn list inconsistency.
- some were not used, some were included twice, some were in the wrong place, some were missing (but it compiled because some were in the wrong place)
- organized a bit better, grouped dash specific includes in original bitcoin files, should save some time solving conflicts when/if merging patches later
* Revert behaviour introduced in 5e1a6afe7f, make nLastDsq local variable instead of being network-wide one. Should fix mixing for new clients and for those who was offline for too long.
* fix docs
* Since we send all mnb's now regardless of mn state, ping check for sigTime being too old is obsolete (and wrong).
Also removing fRequireEnabled, this logic is deprecated too it seems.
* remove (pre-)enabled check in CMasternodeMan::Add
* locks in PS
* lock in governance
* locks in IS
* lock in ProcessGetData
* locks in CMasternodeSync
* centralize mnodeman.Check call
* locks order in mnpayments
* use current block chainTip when possible (less locks)
* add missing lock in CountInputsWithAmount
* fix deadlock RequestLowDataPaymentBlocks/IsTransactionValid
* LOCK2 in CheckMnbAndUpdateMasternodeList, CheckAndUpdate, SendVerifyRequest
* LOCK(cs) is not needed here
* Decouple governance init actions from serialization
Should fix this:
```
Assertion failed: lock governance.cs not held in governance-classes.cpp:117; locks held:
cs_Shutdown init.cpp:200 (TRY)
cs ./governance.h:195
cs governance.cpp:835
Abort trap: 6
```
* PoSe changes:
- use helpers to alter nPoSeBanScore within predefined range only
- use nPoSeBanHeight instead of timeout of inactivity to ban masternodes till some block in the future (currently should block for the whole payment cycle)
- add log output on pose score increase in CheckSameAddr
Squashed:
* Replaced unsafe mnodeman.Find function with Get in governance-vote.cpp
* Reject unparsable governance objects
* Implemented sentinel watchdog objects (separated out from locking changes)
* Added WATCHDOG support to rpcgovernance.cpp
* Implemented WATCHDOG_EXPIRED state for masternodes
* Added serialization of watchdog timestamps
* Masternode fixes
- Added version check to CMasternodeMan deserialization
- Added several missing locking calls in CMasternodeMan
* Fixed missing member initialization in CMasternode constructor and added more logging
* Added MASTERNODE_WATCHDOG_MAX_SECONDS to governanceinfo
* Added masternodewatchdogmaxseconds info to getgovernanceinfo help
* Make masternodes remain in WATCHDOG_EXPIRED state unless removed or collateral expires
* Allow watchdog object creation by WATCHDOG_EXPIRED MN
* Fixed MN validation logic for governance object creation
* Count total masternodes instead of enabled masternodes in masternode-sync
* Transition out of WATCHDOG_EXPIRED state if the watchdog is inactive
* Fixed IsWatchdogExpired bug
* Fixed rate check for watchdog objects and no longer check MN state when validating governance objects
* Applied PR #1061 patch
* Ported locking changes from other branch
* Require only 1 block between new watchdog objects
* Accept pings for WATCHDOG_EXPIRED masternodes
* Lock CmasternodeMan::cs in CmasternodeMan::ProcessMessage
* Several governance changes
- Fixed uninitialized value in CGovernancePayment class
- Return an error on submission if any superblock payment cannot be parsed
- Added logging more statements
* Explicitly initialize all governance object members
* Fix deadlock
* Fixed non-threadsafe access to masternode in activemasternode.cpp
* Revert added wallet lock
* Changed CActiveMasternode so that watchdog expired nodes can still send pings
* Modified CActiveMasternode to run pinger regardless of state when MN is in list
* Added voter and time information to getvotes command
* Improved CActiveMasternode state management
* Implemented GetInfo functions for more efficient thread-safe access to masternode information
* Added CActiveMasternode debug logging messages
* Fixed initial type setting and error message for incorrect protocol version
* Changes based on code review comments
* Set active state for local mode
* Remove unnecessary call to IsCollateralValid and hence allow superblocks
to propagate
* Added CMasternodeMan::Get overload for masternode vin
* Use vin as masternode identifier instead of public key
* Fixed missing member in copy constructor, improved logging
* Added logging for MasternodeRateCheck failures
* Removed pubkeyMasternode field from CGovernanceObject
* Impose MN trigger creation rate limit only when MN info is synced
* Improve rpcgovernance error handling
- Prevent attempts to prepare trigger objects (and waste the collateral)
- Improve clarity of gobject submit error messages
* Implemented different fees for different types of governance objects
* Added fee amounts to object returned by getgovernanceinfo
* Implement new requireents for Superblock creation
- Superblocks creation requires a valid masternode signature
- Superblock creation no longer requires a collateral fee
- Superblock creation rate is limited to roughly 1 per masternode per cycle
* Fixed getgovernanceinfo help message
* Removed old governance fee constant
* Fixed bug in IsSignatureValid and added debugging code
* Fixed parent hash variable index and added debugging code
* Modified GetBudgetSystemCollateralTX to take fee amount parameter
* Changes due to code review comments
- Naming changes
- Removed confusing comment
da406e3 Masternode sync improvements
- add simple helpers for few more sync states (use them where appropriate instead of old code + rpc output)
- use new helpers to avoid meaningless message processing
- actually fail if sync shouldn't continue due to lack of info, make sure Reset is used to quit failed state
- mn hash compatibility with 70103
- ignore some requests while syncing
- fix locking/initializing in sync
- do not ban for old mnw
- split budget/governance messages/invs
513506f Fixing AddRef() usage
Using AddRef() in ConnectNode() for existing connections doesn't feel right considering how refs are released in ThreadSocketHandler(). I guess this could be the reason that sometimes refs stay >0 no matter what and nodes stuck in vNodesDisconnected forever which means that node never get deleted and FinalizeNode signal is never fired which in its turn means that for example mapBlocksInFlight can't be cleaned properly and then blocks stuck.
This commit should solve the issue by:
- removing AddRef() for existing connections
- adding AddRef() in CNode's constructor using the same conditions as in ThreadSocketHandler()
- addding AddRef() in ConnectNode() and Release() in ThreadSocketHandler() for mixing nodes
- removing explicit calls to Release() (back to `pnode->fDisconnect = true` in `CMasternodeMan::ProcessMasternodeConnections`)
9da4a83 fix names/comments
- use "" account address instead
- remove `collateralPubKey` and `SetCollateralAddress()` from darksend and create `dummyScriptPubkey` and `InitDummyScriptPubkey()` in masternodeman instead
- initialize `dummyScriptPubkey` before loading cache otherwise MNs could be removed because of the invalid tx
- every node should check for "pre-enabled" status of a newly added mn to relay message further (was "enabled" which is wrong - if node is enabled, that's not the first announce, so no need to relay it because most likely we were the one who asked for that mn directly via dseg)
- we should not add ourselves to mn list if announce message has outdated protocol version (report this in log)
- we should remove ourselves from saved list if we start updated mn for the first time and just loaded old list from mncache.dat
- Move heavy `IsVinAssociatedWithPubkey` deep inside of `CheckInputsAndAdd` to be actually executed only once per masternode and only after much simpler checks were executed;
- Verify `GetInputAge` (it uses cache) before trying to enter `AcceptToMemoryPool` (which hash many ifs and checks inside).
- make logthreadnames work correctly without logtimestamps (and with emty strings again)
- fix crash in debug (it3)
- log debug info about locks in one line
- show actual DS rounds instead of n/a in coin control while in debug mode
- fix initial string in debug.log: "Bitcoin version" -> "Dash version"
This set of rpc commands provide an ability to create masternode broadcast messages offline
and relay them from online node later (messages expire in ~1 hour).
Arguments:
1. command (string or set of strings, required) The command to execute
2. passphrase (string, optional) The wallet passphrase
Available commands:
create-alias - Create single remote masternode broadcast message by assigned alias configured in masternode.conf
create-all - Create remote masternode broadcast messages for all masternodes configured in masternode.conf
decode - Decode masternode broadcast message
relay - Relay masternode broadcast message to the network
Closes#741
