Skip to content

Resolve cs_main vs governance vs masternode-sync deadlocks #4293

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
UdjinM6 merged 1 commit into from
Jul 28, 2021
Merged

Resolve cs_main vs governance vs masternode-sync deadlocks #4293

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
84 changes: 50 additions & 34 deletions src/masternode/masternode-sync.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -139,23 +139,22 @@ void CMasternodeSync::ProcessTick(CConnman& connman)
}

nTimeLastProcess = GetTime();
std::vector<CNode*> vNodesCopy = connman.CopyNodeVector(CConnman::FullyConnectedOnly);

// gradually request the rest of the votes after sync finished
if(IsSynced()) {
std::vector<CNode*> vNodesCopy = connman.CopyNodeVector(CConnman::FullyConnectedOnly);
governance.RequestGovernanceObjectVotes(vNodesCopy, connman);
connman.ReleaseNodeVector(vNodesCopy);
return;
}

{
LOCK(cs);
// Calculate "progress" for LOG reporting / GUI notification
double nSyncProgress = double(nTriedPeerCount + (nCurrentAsset - 1) * 8) / (8*4);
LogPrint(BCLog::MNSYNC, "CMasternodeSync::ProcessTick -- nTick %d nCurrentAsset %d nTriedPeerCount %d nSyncProgress %f\n", nTick, nCurrentAsset, nTriedPeerCount, nSyncProgress);
uiInterface.NotifyAdditionalDataSyncProgressChanged(nSyncProgress);

std::vector<CNode*> vNodesCopy = connman.CopyNodeVector(CConnman::FullyConnectedOnly);

for (auto& pnode : vNodesCopy)
{
CNetMsgMaker msgMaker(pnode->GetSendVersion());
Expand Down Expand Up @@ -256,37 +255,10 @@ void CMasternodeSync::ProcessTick(CConnman& connman)
return;
}

// only request obj sync once from each peer, then request votes on per-obj basis
// only request obj sync once from each peer
if(netfulfilledman.HasFulfilledRequest(pnode->addr, "governance-sync")) {
int nObjsLeftToAsk = governance.RequestGovernanceObjectVotes(pnode, connman);
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 received during the last tick.
// We can be pretty sure that we are done syncing.
LogPrintf("CMasternodeSync::ProcessTick -- nTick %d nCurrentAsset %d -- asked for all objects, nothing to do\n", nTick, nCurrentAsset);
// reset nTimeNoObjectsLeft to be able to use the same condition on resync
nTimeNoObjectsLeft = 0;
SwitchToNextAsset(connman);
connman.ReleaseNodeVector(vNodesCopy);
return;
}
nLastTick = nTick;
nLastVotes = governance.GetVoteCount();
}
// will request votes on per-obj basis from each node in a separate loop below
// to avoid deadlocks here
continue;
}
netfulfilledman.AddFulfilledRequest(pnode->addr, "governance-sync");
Expand All @@ -296,11 +268,55 @@ void CMasternodeSync::ProcessTick(CConnman& connman)

SendGovernanceSyncRequest(pnode, connman);

break; //this will cause each peer to get one request each six seconds for the various assets we need
}
}
}
} // cs


if (WITH_LOCK(cs, return nCurrentAsset != MASTERNODE_SYNC_GOVERNANCE)) {
// looped through all nodes and not syncing governance yet/already, release them
connman.ReleaseNodeVector(vNodesCopy);
return;
}

// request votes on per-obj basis from each node
for (auto& pnode : vNodesCopy) {
if(!netfulfilledman.HasFulfilledRequest(pnode->addr, "governance-sync")) {
continue; // to early for this node
}
int nObjsLeftToAsk = governance.RequestGovernanceObjectVotes(pnode, connman);
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 received during the last tick.
// We can be pretty sure that we are done syncing.
LogPrintf("CMasternodeSync::ProcessTick -- nTick %d nCurrentAsset %d -- asked for all objects, nothing to do\n", nTick, MASTERNODE_SYNC_GOVERNANCE);
// reset nTimeNoObjectsLeft to be able to use the same condition on resync
nTimeNoObjectsLeft = 0;
SwitchToNextAsset(connman);
connman.ReleaseNodeVector(vNodesCopy);
return; //this will cause each peer to get one request each six seconds for the various assets we need
return;
}
nLastTick = nTick;
nLastVotes = governance.GetVoteCount();
}
}

// looped through all nodes, release them
connman.ReleaseNodeVector(vNodesCopy);
}
Expand Down