Commit Graph

15647 Commits

Author SHA1 Message Date
PastaPastaPasta
dd1245c2a9 Update dnsseed-policy.md (#2751)
* Remove irrelevant link in dnsseed-policy

* Link to dash-seeder in dnsseed-policy

instead of bitcoin-seeder
2019-03-08 21:21:20 +03:00
Alexander Block
64ae91268a
Merge pull request #2735 from codablock/pr_llmq_instantsend
Implement LLMQ based InstantSend
2019-03-08 11:32:44 +01:00
Alexander Block
06fc655595 Actually remove from finalInstantSendLocks in CInstantSendManager::RemoveFinalISLock 2019-03-07 21:15:09 +01:00
Alexander Block
041a1c26d0 Move safe TX checks into TestForBlock and TestPackageTransactions
Otherwise we'll miss checks for ancestors.
2019-03-07 21:15:09 +01:00
Alexander Block
4d3365ddb2 Completely disable InstantSend while filling mempool in autoix-mempool.py
Otherwise we overload Travis and tests start to randomly fail.
2019-03-07 21:15:09 +01:00
Alexander Block
fae33e03ae Let ProcessPendingReconstructedRecoveredSigs return void instead of bool
Return value is unused and the method actually never returned something.
2019-03-07 21:15:09 +01:00
Alexander Block
f8f867a6bb Sync blocks after generating in mine_quorum() 2019-03-07 21:15:09 +01:00
Alexander Block
3e60d2d2db Adjust LLMQ based InstantSend tests for new spork20 2019-03-07 21:15:09 +01:00
UdjinM6
41a71fe443 update autoix-mempool.py to test both "old" and "new" InstantSend (and fix CheckCanLock to respect mempool limits) 2019-03-07 21:15:09 +01:00
UdjinM6
843b6d7a95 update p2p-autoinstantsend.py to test both "old" and "new" InstantSend 2019-03-07 21:15:09 +01:00
UdjinM6
a8da11ac5d update p2p-instantsend.py to test both "old" and "new" InstantSend 2019-03-07 21:15:09 +01:00
UdjinM6
55152cb313 Move IS block filtering into ConnectBlock 2019-03-07 21:15:09 +01:00
Alexander Block
2299ee2836 Rename IXLOCK to ISLOCK and InstantX to InstantSend 2019-03-07 21:15:09 +01:00
Alexander Block
f5dcb00acf Introduce spork SPORK_20_INSTANTSEND_LLMQ_BASED to switch between new/old system 2019-03-07 21:15:09 +01:00
Alexander Block
280690792a Combine loops in CChainLocksHandler::NewPoWValidBlock 2019-03-07 21:15:09 +01:00
Alexander Block
9e9081110b Add override keywork to CDSNotificationInterface::NotifyChainLock 2019-03-07 21:15:09 +01:00
Alexander Block
e2f99f4ae1 Set llmqForInstantSend for mainnet and testnet 2019-03-07 21:15:09 +01:00
Alexander Block
5b8344e8f8 Use scheduleFromNow instead of schedule+boost::chrono 2019-03-07 21:15:09 +01:00
Alexander Block
f44f09ca04 Fix crash in BlockAssembler::addPackageTxs 2019-03-07 21:15:09 +01:00
Alexander Block
baf8b81c4a Fix no-wallet build 2019-03-07 21:15:09 +01:00
Alexander Block
2a7a5c6338 Only sign ChainLocks when all included TXs are "safe"
Safe means that the TX is either ixlocked or known since at least 10
minutes.

Also change miner code to only include safe TXs in block templates.
2019-03-07 21:15:09 +01:00
Alexander Block
96291e7a0f Cheaper/Faster bailout from TrySignChainTip when already signed before 2019-03-07 21:15:09 +01:00
Alexander Block
0a5e8eb862 Move ChainLock signing into TrySignChainTip and call it periodically
Later commits will introduce checks for "safe TXs" which might abort the
signing on first try, but succeed a few seconds later, so we periodically
retry to sign the tip.
2019-03-07 21:15:09 +01:00
Alexander Block
bd7edc8ae9 Track txids of new blocks and first-seen time of TXs in CChainLocksHandler 2019-03-07 21:15:09 +01:00
Alexander Block
7945192ff6 Enforce IX locks from the new system 2019-03-07 21:15:09 +01:00
Alexander Block
dc97835ec6 Disable explicit lock requests when the new IX system is active 2019-03-07 21:15:09 +01:00
Alexander Block
68cfdc9325 Also call ProcessTx from sendrawtransaction and RelayWalletTransaction 2019-03-07 21:15:09 +01:00
Alexander Block
1d2d370cd0 Whenever we check for locked TXs, also check for the new system having a lock 2019-03-07 21:15:09 +01:00
Alexander Block
3a6cc2cc17 Show "verified via LLMQ based InstantSend" in GUI TX status 2019-03-07 21:14:31 +01:00
Alexander Block
34a8b2997a Disable old IX code when the new system is active 2019-03-07 21:14:31 +01:00
Alexander Block
5ff4db0a05 Downgrade TXLOCKREQUEST to TX when new IX system is active
The new system does not require explicit lock requests, so we downgrade
TXLOCKREQUEST to TX and start propagating it instead of the original.
2019-03-07 21:14:31 +01:00
Alexander Block
1959f3e4a5 Handle incoming TXs by calling CInstantXManager::ProcessTx
This also includes handling of TXs that were previously orphanced
2019-03-07 21:14:31 +01:00
Alexander Block
83dbcc483f Implement CInstantSendManager and related P2P messages 2019-03-07 21:14:31 +01:00
Alexander Block
5bbc122749 Implement PushReconstructedRecoveredSig in CSigningManager
We can reconstruct recovered sigs from other P2P messages to avoid
re-validation of those. We will do this later in InstantSend code.
2019-03-07 21:14:31 +01:00
Alexander Block
2bbac8ff77 Introduce NotifyChainLock signal and invoke it when CLSIGs get processed 2019-03-07 21:14:31 +01:00
UdjinM6
e47af29d41
Fix "chainlock" in WalletTxToJSON (#2748)
* Fix "chainlock" in WalletTxToJSON

* Add `AssertLockHeld(cs_main); // for mapBlockIndex`

* move assert higher
2019-03-07 16:29:37 +03:00
Alexander Block
e21d8d6b95 Fix error message for invalid voting addresses (#2747) 2019-03-07 16:29:24 +03:00
Alexander Block
80891ee6f7 Make -masternodeblsprivkey mandatory when -masternode is given (#2745) 2019-03-06 22:45:53 +03:00
Alexander Block
521d4ae08f Implement 2-stage commit for CEvoDB to avoid inconsistencies after crashes (#2744)
* Let Commit() return void

The boolean return value will loose its meaning in the next commit

* Implement 2-stage commits for CDBTransaction and CScopedDBTransaction

CDBTransaction is changed to allow CDBBatch, CDBWrapper and other
CDBTransactions as parent instead of just CDBWrapper. This in turn allows
to implement multi-staged commits in CEvoDB.

We now have the "current transaction" which is started and ended (commit
or rollback) for each call to Connect-/DisconnectBlock. When the current
transaction is committed, it moves its contents into the "root transaction"
instead of directly writing to CDBWrapper.

CommitRootTransaction() then handles the final commitment to CDBWrapper. It
is called at the same time when the chainstate is flushed to disk, which
guarantees consistency between chainstate and CEvoDB.

* Allow to efficiently move values into parent transactions to avoid copies

When CDBTransaction<CDBTransaction<...>>::Commit() is called, we can avoid
copying values from this transaction to the parent transaction and instead
pass values by rvalue and let the contents be moved.

* Revert "Force FlushStateToDisk on ConnectTip/DisconnectTip while not in IBD (#2560)"

This reverts commit 6dfceaba5a.
2019-03-06 22:45:39 +03:00
UdjinM6
e63cdadc97
Add a button/context menu items to show QR codes for addresses (#2741)
* Add a button/context menu item to show qr codes for addresses in address book

* Add a context menu item to show qr code for addresses in transaction list
2019-03-06 22:45:22 +03:00
UdjinM6
b6177740c2 Add collateraladdress into masternode/protx list rpc output (#2740) 2019-03-06 08:01:03 +01:00
UdjinM6
b5466e20a9 Add "chainlock" field to more rpcs (#2743) 2019-03-06 08:00:47 +01:00
Alexander Block
8dd9349224
Don't be too harsh for invalid CLSIGs (#2742)
The local node might be the bad one actually as it might not have catched
up with the chain. In that case, LLMQs might be different for the sending
and receiving node.
2019-03-06 08:00:21 +01:00
Alexander Block
a34fb6d6f7 Fix banning when local node doesn't have the vvec (#2739)
When ProcessMessageBatchedSigShares returns false, it's interpreted as
if an invalid/malicious message was received, causing a ban. So, we should
return "!ban" instead of just "ban".
2019-03-05 14:04:04 +03:00
Alexander Block
4a495c6b4d
Only include selected TX types into CMerkleBlock (#2737)
It was reported on iOS that CMerkleBlock sometimes included the dummy
quorum commitments introduced with v13, which led to banning of nodes as
these were not supported/expected there.

We should in general only include TXs here that are of interest for SPV
nodes, so we should maintain the list of allowed TX types.
2019-03-04 07:52:14 +01:00
Alexander Block
f971da8318 Stop g_connman first before deleting it (#2734) 2019-03-01 18:30:30 +03:00
Alexander Block
9eb0ca7040 Ignore sig share inv messages when we don't have the quorum vvec (#2733)
* Ignore sig share inv messages when we don't have the quorum vvec

* Update src/llmq/quorums_signing_shares.cpp

Co-Authored-By: codablock <ablock84@gmail.com>
2019-03-01 18:30:11 +03:00
fish-en
080b59a573 Backport bitcoin#14385: qt: avoid system harfbuzz and bz2 (#2732)
We may eventually want to break out harfbuzz and build it in depends, but
for now just ensure that runtime dependencies don't depend on whether or not
harfbuzz was present on the builder.
2019-03-01 18:29:48 +03:00
Alexander Block
2041186f4d
On timeout, print members proTxHashes from members which did not send a share (#2731)
* On timeout, print members proTxHashes from members which did not send a share

* Move inactive quorums check above timeout checks

This allows to reuse things in the next commit

* Avoid locking cs_main through GetQuorum by using a pre-filled map

* Use find() instead of [] to access quorums map
2019-03-01 08:21:42 +01:00
Alexander Block
ea90296b6d
Actually start the timers for sig share and recSig verification (#2730)
Was wondering why verification was always 0ms...this explains it :)
2019-03-01 08:21:28 +01:00