This avoids the wasted time and consensus confusion causes by syncing and then validation failing. This is significant after the algo has run, as many signers will become invalid.
The validation is currently set to a feature trigger of height 0, although this will likely be set to a future block, in case there are any cases in the chain's history where this validation may fail (e.g. transfer privs?)
When users buy QORT ("Alice"-side), most of the API time is spent computing mempow for the MESSAGE sent to Bob's AT.
This is the final stage startResponse() and after Alice's P2SH is already broadcast.
To speed this up, the MESSAGE part is moved into its own thread allowing startResponse() to return sooner, improving the user experience.
Caveats:
If MESSAGE importAsUnconfirmed() somehow fails the the buy won't complete and Alice will have to wait for P2SH refund.
If Alice shuts down her node while MESSAGE mempow is being computed then it's possible the shutdown will be blocked until mempow is complete.
Currently only implemented in LitecoinACCTv3TradeBot as this is only proof-of-concept.
Tested with multiple buys in the same block.
This is a hopeful fix for extra memory usage since mempow activated, due to adding a lot of load to the garbage collector. It only applies to accounts verified from the import queue; the optimization hasn't been applied to block processing. But verifying online accounts when processing blocks is rare and generally would only last a short amount of time.
This allows testnets to more easily coexist on the same machines that are running a mainnet instance, and still tests the mempow computation and verification in a non-resource-intensive way.
We no longer need all the code complexity, now that 24 hours have passed since activation. We don't validate online accounts beyond 12 hours, and the data is trimmed after 24 hours.
- Added "singleNodeTestnet" setting, allowing for fast and consecutive block minting, and no requirement for a minimum number of peers.
- Added "recoveryModeTimeout" setting (previously hardcoded in Synchronizer).
- Updated testnets documentation to include new settings and a quick start guide.
- Added "generic" minting account that can be used in testnets (not functional on mainnet), to simplify the process for new devs.
We expect a "Couldn't build a to-be-minted block" log on every startup due to trying to mint before having any accounts. This one has moved from error to info level because error logs can be quite intrusive when using an IDE.
This allows one message to reference another, e.g. for replies, edits, and reactions. We can't use the existing reference field as this is used for encryption and generally points to the user's lastReference at the time of signing.
"chatReference" is based on the "nameReference" field used in various name transactions, for similar purposes.
This needs a feature trigger timestamp to activate, and that same timestamp will need to be used in the UI since that is responsible for building the chat transactions.
The BLOCK_SUMMARIES message type would differentiate between an empty response and a missing/invalid response. However, in V2, a response with empty summaries would throw a BufferUnderflowException and be treated by the caller as a null message.
This caused problems when trying to find a common block with peers that have diverged by more than 8 blocks. With V1 the caller would know to search back further (e.g. 16 blocks) but in V2 it was treated as "no response" and so the caller would give up instead of increasing the look-back threshold.
This fix will identify BLOCK_SUMMARIES_V2 messages with no content, and return an empty array of block summaries instead of a null message.
Should be enough to recover any stuck nodes, as long as they haven't diverged more than 240 blocks from the main chain.
