- every node should check for "pre-enabled" status of a newly added mn to relay message further (was "enabled" which is wrong - if node is enabled, that's not the first announce, so no need to relay it because most likely we were the one who asked for that mn directly via dseg)
- we should not add ourselves to mn list if announce message has outdated protocol version (report this in log)
- we should remove ourselves from saved list if we start updated mn for the first time and just loaded old list from mncache.dat
- Move heavy `IsVinAssociatedWithPubkey` deep inside of `CheckInputsAndAdd` to be actually executed only once per masternode and only after much simpler checks were executed;
- Verify `GetInputAge` (it uses cache) before trying to enter `AcceptToMemoryPool` (which hash many ifs and checks inside).
This set of rpc commands provide an ability to create masternode broadcast messages offline
and relay them from online node later (messages expire in ~1 hour).
Arguments:
1. command (string or set of strings, required) The command to execute
2. passphrase (string, optional) The wallet passphrase
Available commands:
create-alias - Create single remote masternode broadcast message by assigned alias configured in masternode.conf
create-all - Create remote masternode broadcast messages for all masternodes configured in masternode.conf
decode - Decode masternode broadcast message
relay - Relay masternode broadcast message to the network
Closes#741
- Syncing process is now event based, rather than timeout based. This means the system can tell when it's done with each step and moves on between phases much faster. In initial testing it seems to be about 10-15x faster and has synced everytime successfully.
- Please remove print debugging when the syncing system is proven to be debugged.
- Only send 1 winner per block for the previous 4500 blocks (was 10 items per block for 6000 blocks)
- Send full sync for last 5 blocks and next 20 blocks
- Turn off mnb cleaning, there aren't that many broadcasts on the network and this could cause problems in some cases
- Fix budget submission collateral check
- use map.insert instead of [] (should be safer)
- debug output / comments / spaces / names
- fix few long strings / make translatable one more label in UI
- if(pindexPrev->nHeight + 4 < pindexBestHeader->nHeight || pindexPrev->nTime + 600 < GetTime()) return;
-- && allowed skipping in various situations, which caused blocks to be rejected because of lack of mnfinalbudget data
- Consensus system selects 1/10 of the oldest masternodes by payment, then selects payee by score from those. This fixes various race conditions when blocks are close together or inconsistant historical winner lists.
- Ask for up to 2 cycles of history
- Keep up to 5 cycles of history locally
- Cache saves masternode payment history
- On startup, the client will find the most recent block and calculate the amount of entries to ask for. The other peer will then send that amount of entries to save bandwidth on restarts.
- Replaced coinbase cache in favor of using the masternode payments voting system only
- Syncing masternode payments now supports up to the syncing the entire payment list
- Do not rely on local lastTimeSeen and requested fRequested anymore. Use last know (signed) ping instead and base all logic on that. Should reduce mn list difference between
nodes.
- Rework CActiveMasternode accordingly along with states, errorMessages, rpc etc.
- Clean some related code, move parts from public to private
- drop c_str in LogPrintf that were related to this functionality (todo: drop it for LogPrintf everywhere else)
- With nodes coming and going on the network, the network could come to different opinions about who should get paid next in line due to some nodes being flagged as failing a PoSe check. This will have to be fixed by introducing a blockchain based PoSe system, but that's out of the scope of this release. To fix the issues in the interrim, I'm removing PoSe checks for the time being.
- Masternode nLastPaid is removed and a new caching system that keeps the last 30 days of coinbase payees replaces it
- To deal with some significant attack vectors, the masternode donation feature was removed. The donation feature was added to support developement anyway, so this will be replaced by the budgeting code.
- This code should allow the network to come to consensus about who should be paid pretty effectively
- Fixed a bug causing the broadcasts to be ignored in some situations
- Added some logic to rebroadcast new masternodes, so the rest of the network finds out about unknown nodes quickly
- should help masternode payments to work properly
- Removed of reference node and replaced with decentralized quorums that pick the masternodes who get paid each block.
- Made a budgeting system, where masternodes can vote on individual budgets and the data is stored perminently on each clients computer
- Added new CMasterBroadcast/CmasternodePing objects, with all of the code required to check each new message. SHould be much easier to tell what's going on now.
- Added nLastPaid to masternode information for persistant storage of the masternode payment cycle
- Masternodes get flagged as just paid when they enter the masternode list and must wait a full cycle to be paid again
- Masternode RPC commands now show short vin string instead of ip address
- Ensures ports remain open and client are responsive to IX requests.
- Completely 100% decentralized. This farms out the work of checking the masternode network to the masternode network. 1% of the network is determistically selected to check another 1% of the network each block. It takes six separate checks to deactivate a node, thus making it tamper proof.
- Nodes are kept in the masternode list if they fail enough PoSe checks to deactivate. They will continue to be checked until the operator fixes them. However they will not be paid while they're failing checks.
- Send a few inputs/output each cycle
- Mix other users inputs/outputs into relay each cycle
- Send signatures as a tickle, also mixed with other users signatures
- Fixed a bug with validating the final transaction
- call ProcessMessageMasternodePayments on ProcessBlock (lost after moving mnodeman functionality)
- do not process extra functionality messages (DS, IX, spork) on initial download / reindex
vecMasternodes.size() == 0 is a nearly impossible condition due to the behavior
of dseep. Each time a node receives a ping where it's unaware of the masternode
it will ask for the dsee from it's peer.
- if(c % 5 == 0 && (RequestedMasterNodeList <= 2 || vecMasternodes.size()
+ if(c % 5 == 0 && RequestedMasterNodeList < 3){
LiteMode disables Darksend/Masternodes/InstantX for clients
who want speed and don't need access to these features. UI
for Darksend is also hidden while in this mode.
- Made masternodes/darksend compatible with regression testing mode (a local-only blockchain that doesn't require mining). Developers can now test multiple rounds in a few minutes without waiting on mining (much faster).
- Added dsee security verification to v11
- darkSendMasternodes -> vecMasternodes (must clearer)
