dash/src/masternode-sync.cpp

401 lines
19 KiB
C++
Raw Normal View History

2016-12-20 14:26:45 +01:00
// Copyright (c) 2014-2017 The Dash Core developers
2015-07-15 04:44:58 +02:00
// Distributed under the MIT/X11 software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
2015-07-29 05:26:08 +02:00
#include "activemasternode.h"
#include "checkpoints.h"
2016-04-15 04:54:11 +02:00
#include "governance.h"
#include "validation.h"
2015-07-15 04:44:58 +02:00
#include "masternode.h"
2016-08-28 12:12:14 +02:00
#include "masternode-payments.h"
#include "masternode-sync.h"
#include "masternodeman.h"
#include "netfulfilledman.h"
#include "spork.h"
2015-07-15 04:44:58 +02:00
#include "util.h"
class CMasternodeSync;
CMasternodeSync masternodeSync;
2016-08-28 12:12:14 +02:00
void CMasternodeSync::Fail()
{
nTimeLastFailure = GetTime();
nRequestedMasternodeAssets = MASTERNODE_SYNC_FAILED;
}
2015-08-04 20:21:27 +02:00
void CMasternodeSync::Reset()
Merge #944: V0.12.1.x governance pr - part 1 - base functionality 068c178 Added DBG macro in util.h to facilitate debugging - This macro allows debugging statements (typically printf's or cout's) to be activated or deactivated with a single comment. Uncomment the line: //#define ENABLE_DASH_DEBUG in util.h to enable debugging statements. - When commented any code wrapped with the DBG() macro will simply be removed by the preprocessor. When not commented all such wrapped statements will be present. - For maximum effectiveness it is best that util.h be the first effective include in all source files. It is also possible to enable the macro for a single file by temporarily adding #define ENABLE_DASH_DEBUG to the top of the file. - Code committed to non-development branches should always have the define commented. d125d9b V0.12.1.x -- merging trigger/generic object/superblock changes for testnet phase II - This commit contains the core governance system changes for 0.12.1. Any unrelated changes have either been removed or moved to separate commits. 120724c File mode fixes - Changed mode 0755->0644 on several source files. c7f9e11 Updated todo reminders - Added reminder to revert temporary reduction of number of votes required to trigger superblock to 1 for testing 92adc98 Made CSuperblockManager::IsValidSuperblockHeight an inline function - This is for efficiency since this function is called often and is only 1 line of code. c050ed7 Added comment explaining rationale for no LOCK(cs) in CSuperblock::IsValid dc933fe Removed unused CSuperblockManager::IsBlockValid function decec88 Moved calls to SuperblockManager::IsValidSuperblockHeight into IsSuperblockTriggered. - Since calls to the later function are always protected by the former there's no reason to keep these separate and this simplifies the code in masternode-payments.cpp. 8672885 Reestablished expected value check for non-superblocks in IsBlockValueValid b01cbe0 Changes to IsBlockValueValid to fix rpc test failure a937c76 Changed include order to allow per file activation of the DBG macro d116aa5 Fixed IsValidSuperblockHeight logic - Note this has an effect on testing because we can now only create 1 superblock per day. Devs may need to temporarily change testnet params for easier testing. 2d0c2de Convert superblock payments to CAmount - We assume that payment values in JSON are in units of DASH for consistency with other RPC functions, such as createrawtransaction. 376b833 Revert temporary testing value for nAbsVoteReq - Also ensure that number of votes required is never smaller than 1 8c89f4b Cleaned up CSuperblock error handling - Exceptions are now thrown consistently rather than using a mix of exceptions and return code checking. Exceptions are now caught only in AddNewTrigger when the CSuperblock constructor is called. Unnecessary object status members have been removed. d7c8a6b Removed utilstrencodings header - This appears to help with travis tests, for unknown reasons. c4dfc7a Fixed some minor code review issues 63c3580 Reverted locking change in miner. - This should have been done in the original PR but was overlooked. 4ab72de Fixed variable name to match common practice and bracket formatting 886a678 Improvements to vote conversion code - Replaced redundantly defined function with inclusion of governance-vote.h - Replaced magic numbers with their corresponding constant symbols 0a37966 Reordered governance message handling
2016-08-17 09:08:25 +02:00
{
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAssets = MASTERNODE_SYNC_INITIAL;
nRequestedMasternodeAttempt = 0;
nTimeAssetSyncStarted = GetTime();
nTimeLastBumped = 0;
2016-08-28 12:12:14 +02:00
nTimeLastFailure = 0;
2015-07-15 04:44:58 +02:00
}
void CMasternodeSync::BumpAssetLastTime(std::string strFuncName)
{
if(IsSynced() || IsFailed()) return;
nTimeLastBumped = GetTime();
if(fDebug) LogPrintf("CMasternodeSync::BumpAssetLastTime -- %s\n", strFuncName);
}
std::string CMasternodeSync::GetAssetName()
{
2016-08-28 12:12:14 +02:00
switch(nRequestedMasternodeAssets)
{
2016-08-28 12:12:14 +02:00
case(MASTERNODE_SYNC_INITIAL): return "MASTERNODE_SYNC_INITIAL";
case(MASTERNODE_SYNC_LIST): return "MASTERNODE_SYNC_LIST";
case(MASTERNODE_SYNC_MNW): return "MASTERNODE_SYNC_MNW";
case(MASTERNODE_SYNC_GOVERNANCE): return "MASTERNODE_SYNC_GOVERNANCE";
case(MASTERNODE_SYNC_FAILED): return "MASTERNODE_SYNC_FAILED";
case MASTERNODE_SYNC_FINISHED: return "MASTERNODE_SYNC_FINISHED";
default: return "UNKNOWN";
}
}
2016-08-28 12:12:14 +02:00
void CMasternodeSync::SwitchToNextAsset()
2015-07-15 04:44:58 +02:00
{
2016-08-28 12:12:14 +02:00
switch(nRequestedMasternodeAssets)
2015-07-15 04:44:58 +02:00
{
case(MASTERNODE_SYNC_FAILED):
throw std::runtime_error("Can't switch to next asset from failed, should use Reset() first!");
break;
case(MASTERNODE_SYNC_INITIAL):
ClearFulfilledRequests();
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAssets = MASTERNODE_SYNC_LIST;
LogPrintf("CMasternodeSync::SwitchToNextAsset -- Starting %s\n", GetAssetName());
2015-07-15 04:44:58 +02:00
break;
case(MASTERNODE_SYNC_LIST):
LogPrintf("CMasternodeSync::SwitchToNextAsset -- Completed %s in %llds\n", GetAssetName(), GetTime() - nTimeAssetSyncStarted);
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAssets = MASTERNODE_SYNC_MNW;
LogPrintf("CMasternodeSync::SwitchToNextAsset -- Starting %s\n", GetAssetName());
2015-07-15 04:44:58 +02:00
break;
case(MASTERNODE_SYNC_MNW):
LogPrintf("CMasternodeSync::SwitchToNextAsset -- Completed %s in %llds\n", GetAssetName(), GetTime() - nTimeAssetSyncStarted);
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAssets = MASTERNODE_SYNC_GOVERNANCE;
LogPrintf("CMasternodeSync::SwitchToNextAsset -- Starting %s\n", GetAssetName());
2015-07-15 04:44:58 +02:00
break;
case(MASTERNODE_SYNC_GOVERNANCE):
LogPrintf("CMasternodeSync::SwitchToNextAsset -- Completed %s in %llds\n", GetAssetName(), GetTime() - nTimeAssetSyncStarted);
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAssets = MASTERNODE_SYNC_FINISHED;
uiInterface.NotifyAdditionalDataSyncProgressChanged(1);
//try to activate our masternode if possible
activeMasternode.ManageState();
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
// TODO: Find out whether we can just use LOCK instead of:
// TRY_LOCK(cs_vNodes, lockRecv);
// if(lockRecv) { ... }
Backport Bitcoin PR#9609: net: fix remaining net assertions (#1575) + Dashify * 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>
2017-08-17 20:37:22 +02:00
g_connman->ForEachNode(CConnman::AllNodes, [](CNode* pnode) {
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
netfulfilledman.AddFulfilledRequest(pnode->addr, "full-sync");
});
2017-05-31 05:49:09 +02:00
LogPrintf("CMasternodeSync::SwitchToNextAsset -- Sync has finished\n");
2015-07-15 04:44:58 +02:00
break;
}
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAttempt = 0;
nTimeAssetSyncStarted = GetTime();
BumpAssetLastTime("CMasternodeSync::SwitchToNextAsset");
2015-07-15 04:44:58 +02:00
}
std::string CMasternodeSync::GetSyncStatus()
{
2016-08-28 12:12:14 +02:00
switch (masternodeSync.nRequestedMasternodeAssets) {
case MASTERNODE_SYNC_INITIAL: return _("Synchronization pending...");
case MASTERNODE_SYNC_LIST: return _("Synchronizing masternodes...");
case MASTERNODE_SYNC_MNW: return _("Synchronizing masternode payments...");
2016-08-28 12:12:14 +02:00
case MASTERNODE_SYNC_GOVERNANCE: return _("Synchronizing governance objects...");
case MASTERNODE_SYNC_FAILED: return _("Synchronization failed");
case MASTERNODE_SYNC_FINISHED: return _("Synchronization finished");
default: return "";
}
}
2015-07-29 06:16:11 +02:00
void CMasternodeSync::ProcessMessage(CNode* pfrom, std::string& strCommand, CDataStream& vRecv)
{
if (strCommand == NetMsgType::SYNCSTATUSCOUNT) { //Sync status count
2015-07-29 06:16:11 +02:00
//do not care about stats if sync process finished or failed
if(IsSynced() || IsFailed()) return;
2015-07-29 06:16:11 +02:00
int nItemID;
int nCount;
vRecv >> nItemID >> nCount;
LogPrintf("SYNCSTATUSCOUNT -- got inventory count: nItemID=%d nCount=%d peer=%d\n", nItemID, nCount, pfrom->id);
2015-07-30 10:51:48 +02:00
}
}
void CMasternodeSync::ClearFulfilledRequests()
2015-07-30 10:51:48 +02:00
{
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
// TODO: Find out whether we can just use LOCK instead of:
// TRY_LOCK(cs_vNodes, lockRecv);
// if(!lockRecv) return;
2015-07-30 15:44:18 +02:00
Backport Bitcoin PR#9609: net: fix remaining net assertions (#1575) + Dashify * 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>
2017-08-17 20:37:22 +02:00
g_connman->ForEachNode(CConnman::AllNodes, [](CNode* pnode) {
netfulfilledman.RemoveFulfilledRequest(pnode->addr, "spork-sync");
netfulfilledman.RemoveFulfilledRequest(pnode->addr, "masternode-list-sync");
netfulfilledman.RemoveFulfilledRequest(pnode->addr, "masternode-payment-sync");
netfulfilledman.RemoveFulfilledRequest(pnode->addr, "governance-sync");
netfulfilledman.RemoveFulfilledRequest(pnode->addr, "full-sync");
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
});
2015-07-29 06:16:11 +02:00
}
2016-08-28 12:12:14 +02:00
void CMasternodeSync::ProcessTick()
2015-07-15 04:44:58 +02:00
{
2016-08-28 12:12:14 +02:00
static int nTick = 0;
if(nTick++ % MASTERNODE_SYNC_TICK_SECONDS != 0) return;
// reset the sync process if the last call to this function was more than 60 minutes ago (client was in sleep mode)
static int64_t nTimeLastProcess = GetTime();
if(GetTime() - nTimeLastProcess > 60*60) {
LogPrintf("CMasternodeSync::HasSyncFailures -- WARNING: no actions for too long, restarting sync...\n");
Reset();
SwitchToNextAsset();
return;
}
nTimeLastProcess = GetTime();
2015-07-19 10:19:54 +02:00
// reset sync status in case of any other sync failure
if(IsFailed()) {
if(nTimeLastFailure + (1*60) < GetTime()) { // 1 minute cooldown after failed sync
LogPrintf("CMasternodeSync::HasSyncFailures -- WARNING: failed to sync, trying again...\n");
Reset();
SwitchToNextAsset();
}
return;
}
// gradually request the rest of the votes after sync finished
if(IsSynced()) {
std::vector<CNode*> vNodesCopy = g_connman->CopyNodeVector();
governance.RequestGovernanceObjectVotes(vNodesCopy);
g_connman->ReleaseNodeVector(vNodesCopy);
return;
2015-07-18 01:49:41 +02:00
}
2015-07-15 04:44:58 +02:00
// Calculate "progress" for LOG reporting / GUI notification
2016-08-28 12:12:14 +02:00
double nSyncProgress = double(nRequestedMasternodeAttempt + (nRequestedMasternodeAssets - 1) * 8) / (8*4);
LogPrintf("CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d nRequestedMasternodeAttempt %d nSyncProgress %f\n", nTick, nRequestedMasternodeAssets, nRequestedMasternodeAttempt, nSyncProgress);
uiInterface.NotifyAdditionalDataSyncProgressChanged(nSyncProgress);
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
std::vector<CNode*> vNodesCopy = g_connman->CopyNodeVector();
BOOST_FOREACH(CNode* pnode, vNodesCopy)
{
// Don't try to sync any data from outbound "masternode" connections -
// they are temporary and should be considered unreliable for a sync process.
// Inbound connection this early is most likely a "masternode" connection
2017-07-03 15:14:07 +02:00
// initiated from another node, so skip it too.
if(pnode->fMasternode || (fMasterNode && pnode->fInbound)) continue;
// QUICK MODE (REGTEST ONLY!)
if(Params().NetworkIDString() == CBaseChainParams::REGTEST)
{
2016-08-28 12:12:14 +02:00
if(nRequestedMasternodeAttempt <= 2) {
Backport Bitcoin PR#8708: net: have CConnman handle message sending (#1553) * 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
2017-07-27 16:28:05 +02:00
g_connman->PushMessageWithVersion(pnode, INIT_PROTO_VERSION, NetMsgType::GETSPORKS); //get current network sporks
2016-08-28 12:12:14 +02:00
} else if(nRequestedMasternodeAttempt < 4) {
Merge #944: V0.12.1.x governance pr - part 1 - base functionality 068c178 Added DBG macro in util.h to facilitate debugging - This macro allows debugging statements (typically printf's or cout's) to be activated or deactivated with a single comment. Uncomment the line: //#define ENABLE_DASH_DEBUG in util.h to enable debugging statements. - When commented any code wrapped with the DBG() macro will simply be removed by the preprocessor. When not commented all such wrapped statements will be present. - For maximum effectiveness it is best that util.h be the first effective include in all source files. It is also possible to enable the macro for a single file by temporarily adding #define ENABLE_DASH_DEBUG to the top of the file. - Code committed to non-development branches should always have the define commented. d125d9b V0.12.1.x -- merging trigger/generic object/superblock changes for testnet phase II - This commit contains the core governance system changes for 0.12.1. Any unrelated changes have either been removed or moved to separate commits. 120724c File mode fixes - Changed mode 0755->0644 on several source files. c7f9e11 Updated todo reminders - Added reminder to revert temporary reduction of number of votes required to trigger superblock to 1 for testing 92adc98 Made CSuperblockManager::IsValidSuperblockHeight an inline function - This is for efficiency since this function is called often and is only 1 line of code. c050ed7 Added comment explaining rationale for no LOCK(cs) in CSuperblock::IsValid dc933fe Removed unused CSuperblockManager::IsBlockValid function decec88 Moved calls to SuperblockManager::IsValidSuperblockHeight into IsSuperblockTriggered. - Since calls to the later function are always protected by the former there's no reason to keep these separate and this simplifies the code in masternode-payments.cpp. 8672885 Reestablished expected value check for non-superblocks in IsBlockValueValid b01cbe0 Changes to IsBlockValueValid to fix rpc test failure a937c76 Changed include order to allow per file activation of the DBG macro d116aa5 Fixed IsValidSuperblockHeight logic - Note this has an effect on testing because we can now only create 1 superblock per day. Devs may need to temporarily change testnet params for easier testing. 2d0c2de Convert superblock payments to CAmount - We assume that payment values in JSON are in units of DASH for consistency with other RPC functions, such as createrawtransaction. 376b833 Revert temporary testing value for nAbsVoteReq - Also ensure that number of votes required is never smaller than 1 8c89f4b Cleaned up CSuperblock error handling - Exceptions are now thrown consistently rather than using a mix of exceptions and return code checking. Exceptions are now caught only in AddNewTrigger when the CSuperblock constructor is called. Unnecessary object status members have been removed. d7c8a6b Removed utilstrencodings header - This appears to help with travis tests, for unknown reasons. c4dfc7a Fixed some minor code review issues 63c3580 Reverted locking change in miner. - This should have been done in the original PR but was overlooked. 4ab72de Fixed variable name to match common practice and bracket formatting 886a678 Improvements to vote conversion code - Replaced redundantly defined function with inclusion of governance-vote.h - Replaced magic numbers with their corresponding constant symbols 0a37966 Reordered governance message handling
2016-08-17 09:08:25 +02:00
mnodeman.DsegUpdate(pnode);
2016-08-28 12:12:14 +02:00
} else if(nRequestedMasternodeAttempt < 6) {
V0.12.1.x sentinel watchdog pr (#1079) 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
2016-10-17 20:54:28 +02:00
int nMnCount = mnodeman.CountMasternodes();
Backport Bitcoin PR#8708: net: have CConnman handle message sending (#1553) * 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
2017-07-27 16:28:05 +02:00
g_connman->PushMessage(pnode, NetMsgType::MASTERNODEPAYMENTSYNC, nMnCount); //sync payment votes
SendGovernanceSyncRequest(pnode);
} else {
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAssets = MASTERNODE_SYNC_FINISHED;
}
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAttempt++;
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
2015-07-19 01:25:52 +02:00
return;
}
// NORMAL NETWORK MODE - TESTNET/MAINNET
Merge #944: V0.12.1.x governance pr - part 1 - base functionality 068c178 Added DBG macro in util.h to facilitate debugging - This macro allows debugging statements (typically printf's or cout's) to be activated or deactivated with a single comment. Uncomment the line: //#define ENABLE_DASH_DEBUG in util.h to enable debugging statements. - When commented any code wrapped with the DBG() macro will simply be removed by the preprocessor. When not commented all such wrapped statements will be present. - For maximum effectiveness it is best that util.h be the first effective include in all source files. It is also possible to enable the macro for a single file by temporarily adding #define ENABLE_DASH_DEBUG to the top of the file. - Code committed to non-development branches should always have the define commented. d125d9b V0.12.1.x -- merging trigger/generic object/superblock changes for testnet phase II - This commit contains the core governance system changes for 0.12.1. Any unrelated changes have either been removed or moved to separate commits. 120724c File mode fixes - Changed mode 0755->0644 on several source files. c7f9e11 Updated todo reminders - Added reminder to revert temporary reduction of number of votes required to trigger superblock to 1 for testing 92adc98 Made CSuperblockManager::IsValidSuperblockHeight an inline function - This is for efficiency since this function is called often and is only 1 line of code. c050ed7 Added comment explaining rationale for no LOCK(cs) in CSuperblock::IsValid dc933fe Removed unused CSuperblockManager::IsBlockValid function decec88 Moved calls to SuperblockManager::IsValidSuperblockHeight into IsSuperblockTriggered. - Since calls to the later function are always protected by the former there's no reason to keep these separate and this simplifies the code in masternode-payments.cpp. 8672885 Reestablished expected value check for non-superblocks in IsBlockValueValid b01cbe0 Changes to IsBlockValueValid to fix rpc test failure a937c76 Changed include order to allow per file activation of the DBG macro d116aa5 Fixed IsValidSuperblockHeight logic - Note this has an effect on testing because we can now only create 1 superblock per day. Devs may need to temporarily change testnet params for easier testing. 2d0c2de Convert superblock payments to CAmount - We assume that payment values in JSON are in units of DASH for consistency with other RPC functions, such as createrawtransaction. 376b833 Revert temporary testing value for nAbsVoteReq - Also ensure that number of votes required is never smaller than 1 8c89f4b Cleaned up CSuperblock error handling - Exceptions are now thrown consistently rather than using a mix of exceptions and return code checking. Exceptions are now caught only in AddNewTrigger when the CSuperblock constructor is called. Unnecessary object status members have been removed. d7c8a6b Removed utilstrencodings header - This appears to help with travis tests, for unknown reasons. c4dfc7a Fixed some minor code review issues 63c3580 Reverted locking change in miner. - This should have been done in the original PR but was overlooked. 4ab72de Fixed variable name to match common practice and bracket formatting 886a678 Improvements to vote conversion code - Replaced redundantly defined function with inclusion of governance-vote.h - Replaced magic numbers with their corresponding constant symbols 0a37966 Reordered governance message handling
2016-08-17 09:08:25 +02:00
{
if(netfulfilledman.HasFulfilledRequest(pnode->addr, "full-sync")) {
// We already fully synced from this node recently,
// disconnect to free this connection slot for another peer.
pnode->fDisconnect = true;
LogPrintf("CMasternodeSync::ProcessTick -- disconnecting from recently synced peer %d\n", pnode->id);
continue;
}
// SPORK : ALWAYS ASK FOR SPORKS AS WE SYNC
if(!netfulfilledman.HasFulfilledRequest(pnode->addr, "spork-sync")) {
// always get sporks first, only request once from each peer
netfulfilledman.AddFulfilledRequest(pnode->addr, "spork-sync");
// get current network sporks
Backport Bitcoin PR#8708: net: have CConnman handle message sending (#1553) * 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
2017-07-27 16:28:05 +02:00
g_connman->PushMessageWithVersion(pnode, INIT_PROTO_VERSION, NetMsgType::GETSPORKS);
LogPrintf("CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d -- requesting sporks from peer %d\n", nTick, nRequestedMasternodeAssets, pnode->id);
}
// MNLIST : SYNC MASTERNODE LIST FROM OTHER CONNECTED CLIENTS
2015-07-15 04:44:58 +02:00
2016-08-28 12:12:14 +02:00
if(nRequestedMasternodeAssets == MASTERNODE_SYNC_LIST) {
LogPrint("masternode", "CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d nTimeLastBumped %lld GetTime() %lld diff %lld\n", nTick, nRequestedMasternodeAssets, nTimeLastBumped, GetTime(), GetTime() - nTimeLastBumped);
// check for timeout first
if(GetTime() - nTimeLastBumped > MASTERNODE_SYNC_TIMEOUT_SECONDS) {
LogPrintf("CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d -- timeout\n", nTick, nRequestedMasternodeAssets);
2016-08-28 12:12:14 +02:00
if (nRequestedMasternodeAttempt == 0) {
LogPrintf("CMasternodeSync::ProcessTick -- ERROR: failed to sync %s\n", GetAssetName());
// there is no way we can continue without masternode list, fail here and try later
Fail();
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
return;
2016-08-28 12:12:14 +02:00
}
SwitchToNextAsset();
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
return;
}
// only request once from each peer
if(netfulfilledman.HasFulfilledRequest(pnode->addr, "masternode-list-sync")) continue;
netfulfilledman.AddFulfilledRequest(pnode->addr, "masternode-list-sync");
if (pnode->nVersion < mnpayments.GetMinMasternodePaymentsProto()) continue;
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAttempt++;
mnodeman.DsegUpdate(pnode);
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
2016-02-04 23:48:23 +01:00
return; //this will cause each peer to get one request each six seconds for the various assets we need
}
2015-07-15 04:44:58 +02:00
// MNW : SYNC MASTERNODE PAYMENT VOTES FROM OTHER CONNECTED CLIENTS
2016-08-28 12:12:14 +02:00
if(nRequestedMasternodeAssets == MASTERNODE_SYNC_MNW) {
LogPrint("mnpayments", "CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d nTimeLastBumped %lld GetTime() %lld diff %lld\n", nTick, nRequestedMasternodeAssets, nTimeLastBumped, GetTime(), GetTime() - nTimeLastBumped);
// check for timeout first
// This might take a lot longer than MASTERNODE_SYNC_TIMEOUT_SECONDS due to new blocks,
// but that should be OK and it should timeout eventually.
if(GetTime() - nTimeLastBumped > MASTERNODE_SYNC_TIMEOUT_SECONDS) {
LogPrintf("CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d -- timeout\n", nTick, nRequestedMasternodeAssets);
2016-08-28 12:12:14 +02:00
if (nRequestedMasternodeAttempt == 0) {
LogPrintf("CMasternodeSync::ProcessTick -- ERROR: failed to sync %s\n", GetAssetName());
// probably not a good idea to proceed without winner list
Fail();
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
return;
2016-08-28 12:12:14 +02:00
}
SwitchToNextAsset();
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
2015-07-15 04:44:58 +02:00
return;
}
// check for data
2016-08-28 12:12:14 +02:00
// if mnpayments already has enough blocks and votes, switch to the next asset
// try to fetch data from at least two peers though
if(nRequestedMasternodeAttempt > 1 && mnpayments.IsEnoughData()) {
LogPrintf("CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d -- found enough data\n", nTick, nRequestedMasternodeAssets);
2016-08-28 12:12:14 +02:00
SwitchToNextAsset();
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
2016-05-29 20:35:09 +02:00
return;
2015-08-14 05:56:58 +02:00
}
// only request once from each peer
if(netfulfilledman.HasFulfilledRequest(pnode->addr, "masternode-payment-sync")) continue;
netfulfilledman.AddFulfilledRequest(pnode->addr, "masternode-payment-sync");
2016-08-28 12:12:14 +02:00
if(pnode->nVersion < mnpayments.GetMinMasternodePaymentsProto()) continue;
nRequestedMasternodeAttempt++;
// ask node for all payment votes it has (new nodes will only return votes for future payments)
Backport Bitcoin PR#8708: net: have CConnman handle message sending (#1553) * 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
2017-07-27 16:28:05 +02:00
g_connman->PushMessage(pnode, NetMsgType::MASTERNODEPAYMENTSYNC, mnpayments.GetStorageLimit());
// ask node for missing pieces only (old nodes will not be asked)
mnpayments.RequestLowDataPaymentBlocks(pnode);
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
2016-02-04 23:48:23 +01:00
return; //this will cause each peer to get one request each six seconds for the various assets we need
}
// GOVOBJ : SYNC GOVERNANCE ITEMS FROM OUR PEERS
2016-08-28 12:12:14 +02:00
if(nRequestedMasternodeAssets == MASTERNODE_SYNC_GOVERNANCE) {
LogPrint("gobject", "CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d nTimeLastBumped %lld GetTime() %lld diff %lld\n", nTick, nRequestedMasternodeAssets, nTimeLastBumped, GetTime(), GetTime() - nTimeLastBumped);
// check for timeout first
if(GetTime() - nTimeLastBumped > MASTERNODE_SYNC_TIMEOUT_SECONDS) {
LogPrintf("CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d -- timeout\n", nTick, nRequestedMasternodeAssets);
2016-08-28 12:12:14 +02:00
if(nRequestedMasternodeAttempt == 0) {
LogPrintf("CMasternodeSync::ProcessTick -- WARNING: failed to sync %s\n", GetAssetName());
2016-08-28 12:12:14 +02:00
// it's kind of ok to skip this for now, hopefully we'll catch up later?
}
SwitchToNextAsset();
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
return;
}
// only request obj sync once from each peer, then request votes on per-obj basis
if(netfulfilledman.HasFulfilledRequest(pnode->addr, "governance-sync")) {
int nObjsLeftToAsk = governance.RequestGovernanceObjectVotes(pnode);
static int64_t nTimeNoObjectsLeft = 0;
// check for data
if(nObjsLeftToAsk == 0) {
static int nLastTick = 0;
static int nLastVotes = 0;
if(nTimeNoObjectsLeft == 0) {
// asked all objects for votes for the first time
nTimeNoObjectsLeft = GetTime();
}
// make sure the condition below is checked only once per tick
if(nLastTick == nTick) continue;
if(GetTime() - nTimeNoObjectsLeft > MASTERNODE_SYNC_TIMEOUT_SECONDS &&
governance.GetVoteCount() - nLastVotes < std::max(int(0.0001 * nLastVotes), MASTERNODE_SYNC_TICK_SECONDS)
) {
// We already asked for all objects, waited for MASTERNODE_SYNC_TIMEOUT_SECONDS
// after that and less then 0.01% or MASTERNODE_SYNC_TICK_SECONDS
// (i.e. 1 per second) votes were recieved during the last tick.
// We can be pretty sure that we are done syncing.
LogPrintf("CMasternodeSync::ProcessTick -- nTick %d nRequestedMasternodeAssets %d -- asked for all objects, nothing to do\n", nTick, nRequestedMasternodeAssets);
// reset nTimeNoObjectsLeft to be able to use the same condition on resync
nTimeNoObjectsLeft = 0;
SwitchToNextAsset();
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
return;
}
nLastTick = nTick;
nLastVotes = governance.GetVoteCount();
}
continue;
}
netfulfilledman.AddFulfilledRequest(pnode->addr, "governance-sync");
2016-05-24 21:25:33 +02:00
if (pnode->nVersion < MIN_GOVERNANCE_PEER_PROTO_VERSION) continue;
2016-08-28 12:12:14 +02:00
nRequestedMasternodeAttempt++;
2016-05-24 21:25:33 +02:00
SendGovernanceSyncRequest(pnode);
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
2016-05-24 21:25:33 +02:00
return; //this will cause each peer to get one request each six seconds for the various assets we need
}
}
2015-07-15 04:44:58 +02:00
}
// looped through all nodes, release them
Backport Bitcoin PR#8085: p2p: Begin encapsulation (#1537) * 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
2017-07-21 11:35:19 +02:00
g_connman->ReleaseNodeVector(vNodesCopy);
}
void CMasternodeSync::SendGovernanceSyncRequest(CNode* pnode)
{
if(pnode->nVersion >= GOVERNANCE_FILTER_PROTO_VERSION) {
CBloomFilter filter;
filter.clear();
Backport Bitcoin PR#8708: net: have CConnman handle message sending (#1553) * 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
2017-07-27 16:28:05 +02:00
g_connman->PushMessage(pnode, NetMsgType::MNGOVERNANCESYNC, uint256(), filter);
}
else {
Backport Bitcoin PR#8708: net: have CConnman handle message sending (#1553) * 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
2017-07-27 16:28:05 +02:00
g_connman->PushMessage(pnode, NetMsgType::MNGOVERNANCESYNC, uint256());
}
}
void CMasternodeSync::UpdatedBlockTip(const CBlockIndex *pindexNew, bool fInitialDownload)
{
if(fDebug) LogPrintf("CMasternodeSync::UpdatedBlockTip -- pindexNew->nHeight: %d fInitialDownload=%d\n", pindexNew->nHeight, fInitialDownload);
// nothing to do here if we failed to sync previousely,
// just wait till status reset after a cooldown (see ProcessTick)
if(IsFailed()) return;
// switch from MASTERNODE_SYNC_INITIAL to the next "asset"
// the first time we are out of IBD mode (and only the first time)
if(!fInitialDownload && !IsBlockchainSynced()) SwitchToNextAsset();
// postpone timeout each time new block arrives while we are syncing
if(!IsSynced()) BumpAssetLastTime("CMasternodeSync::UpdatedBlockTip");
}