pragmaxim
pragmaxim
Tools like explorer are getting blockIds by height and then corresponding blocks request-by-request which is quite slow and parallelization can either overload a node or be blocked by firewall. See...
So that distribution of transactions in blocks is more equal.
When our node receives `SyncV2` message from an older node, we should check difficulty of its extra headers. Steps : 1. find common header 2. get difficulty of common header...
As I believe that all the people that complain about syncing are those with weak CPU -> huge message queues -> OutOfMemory error ... because these 2 methods are not...
``` applyBlock(block) shouldBe 'success ``` This is not a good assertion as it has side effect that may fail : ``` nodeViewHolderRef ! LocallyGeneratedModifier(fullBlock.header) ``` And following assertions then fail....
Step 1 of 2, testing `CleanupWorker` actor ... later switching to `WrappedArray` will follow Closes https://github.com/ergoplatform/ergo/pull/1559
Follow-up of https://github.com/ergoplatform/ergo/pull/1559 ... again, tests first, then impl ... better to have it efficient in case of traffic increase or ddos Closes https://github.com/ergoplatform/ergo/issues/1556
This is a follow-up of https://github.com/ergoplatform/ergo/pull/1549 where we switched implementation of underlying `txs` in blocks, the same should be done for `CandidateGenerator` and consequently `CleanupWorker` https://github.com/ergoplatform/ergo/issues/1556. There was a few...
There is a lot of untested logic in `CleanupWorker` so it is not safe to for instance Switch from `List` to `WrappedArray` with `cfor` and such. Fixed at : https://github.com/ergoplatform/ergo/pull/1559...
`wallet/payment/send` ... currently it is done statically : https://github.com/ergoplatform/ergo/blob/52eca96214dcb9a145935979e32d1cee99e69da1/src/main/resources/application.conf#L249 @kushti We had a feature request for this, I'm thinking whether the fee is intentionally set to a fixed value or...