560 Commits (8fd8ba62b70659dadbb08e4dd3a72c06b9169854)

Author	SHA1	Message	Date
Sebastian Abondano	8e3d80a6db	Realtime indexer waits half a second before fetch ... Why: * For Sokol, on staging, we noticed many errors like this one: ``` 00:45:05.627 application=indexer [error] realtime indexer failed to validate for block 4668478: [%{code: -32602, data: %{"blockNumber" => "0x473C3E", "hash" => "0x532e069e200be3e372da9680872826bb08caef91"}, message: "Unknown block number"}, %{code: -32602, data: %{"blockNumber" => "0x473C3E", "hash" => "0x95426f2bc716022fcf1def006dbc4bb81f5b5164"}, message: "Unknown block number"}, %{code: -32602, data: %{"blockNumber" => "0x473C3E", "hash" => "0xe797a1da01eb0f951e0e400f9343de9d17a06bac"}, message: "Unknown block number"}]. Block will be retried by catchup indexer. ``` * Issue link: https://github.com/poanetwork/blockscout/issues/808 This change addresses the need by: * Editing `Indexer.Block.Realtime.Fetcher.fetch_and_import_block/2` for it to wait half a second before starting to fetch and import. The rationale here is that we're hoping that half a second will be enough for Parity to sync and not return an "Unknown block number" error.	6 years ago
Sebastian Abondano	14b916f9e8	Realtime.Fetcher async fetching and no skips ... Why: * For the realtime fetcher to keep up on ETH we want it to fetch blocks asynchronously. We noticed that the realtime fetcher was falling behind on ETH mainnet. Currently the realtime fetcher fetches and imports one block at a time. This commit changes that, so that whenever we're notified of a new block number, through websockets, we start a new process to fetch and import each block asynchronously. * Also, whenever the websocket subscription, through which we're notified of new block numbers, skips block numbers, we want to make sure the skipped blocks are also fetched and imported. * Issue link: n/a This change addresses the need by: * Adding a new `Task.Supervisor` called `Indexer.Block.Realtime.TaskSupervisor` under the `Indexer.Block.Realtime.Supervisor` supervisor. This supervisor was created to supervise fetching and importing processes started by the realtime fetcher every time a new block number is received through the websocket subscription. * Editing `Indexer.Block.Realtime.Fetcher` to start a new process, under the new supervisor mentioned above, to fetch the latest block per the websocket `newHeads` subscription. It also determines if any block numbers have been skipped and starts individual processes to fetch and import the skipped blocks also.	6 years ago
Igor Florian	52e861fa73	Include and Log token_balances error attribute	6 years ago
Lucas Narciso	d8b22f6988	Handle large strings when fetching tokens	6 years ago
Amanda Sposito	32ad1214b6	Increase timeout at the `Chain.import` at the 'Indexer.InternalTransaction.Fetcher'	6 years ago
Amanda Sposito	3244113be4	Increase timeout at the `Repo.transaction` level at the 'Indexer.TokenBalance.Fetcher'	6 years ago
Amanda Sposito	eea81817f7	Increase 'Indexer.TokenBalance.Fetcher' timeout	6 years ago
Amanda Sposito	b4a66ad7f0	Add treatment to TokenBalance.Fetcher when there is an error with the import data ... * The TokenBalanceFetcher was not considering empty lists and errors when trying to import the data into the database.	6 years ago
Alex Garibay	56299e308b	Add support for ERC-721 token transfer logs with addresses in `data`	6 years ago
Luke Imhoff	1a8bb6cbed	Restructure Indexer.BlockFetcher to match other Fetchers ... Move the "Fetcher" part below the "time" part of "Realtime" or "Catchup", so that the "Fetcher" and "Supervisor" are siblings like in the fetchers that don't have a "time" part.	6 years ago
Luke Imhoff	5b7872f030	Don't use closure for BufferedTask Task.Supervisor.async_nolink call ... Fixes #676 For some reason, which I haven't tracked down yet, a closure inside of `BufferedTask.spawn_next_batch/1` continually grows in size. When this closure is passed to `Task.Supervisor.async_nolink`, the `Task.Supervisor`, for #676, this was `Indexer.TaskSupervisor` and is now `Indexer.InternalTransaction.TaskSupervisor`, it also grows in memory size. I think this is related to the way that functions and closures are serialized when sent between processes, but I haven't figured out how to verify that. The change also cleared the MsgQ for `Indexer.TransactionFetcher`, probably because its not blocked by the call to `Indexer.InternalTransaction.Fetcher`, which was blocked due to something to do with the memory usage of the closures. The `Indexer.InternalTransaction.TaskSupervisor` consistently shows multiple tasks under it in `:observer`, which it did not previously either.	6 years ago
Luke Imhoff	09051ceb8a	Separate usage of Task.Supervisor ... Have a separate subtree for each fetcher with a supervisor supervising the fetcher and a Task.Supervisor that is only used by the fetcher, so that looking at the Task.Supervisor in that subtree will show which fetcher is causing any memory, message, or blocking problems.	6 years ago
Luke Imhoff	9b3ed74507	Allow all genserver options to be passed to BufferedTasks ... Previously, only name could be passed, but now, by separating callback init options from genserver options, we can pass any options that GenServer supports, including :debug options, which is helpful for tracing performance issues.	6 years ago
Felipe Renan	6148658f61	Store tokens with invalid strings in Token indexer. ... It's a temp fix for https://github.com/poanetwork/blockscout/issues/669. For now, we are going to store those tokens that have names and/or symbols invalids considering: - the first 6 characters from the contract address hash for the name. - nil for the symbol.	6 years ago
Felipe Renan	e86c5c79f3	Handle timeout when fetching token balances from blockchain	6 years ago
Igor Florian	7077a41576	Change Address fetched_balance's columns to fetched_coin_balace	6 years ago
Igor Florian	acf14d6cf3	Rename Balance references to Address.CoinBalance	6 years ago
Igor Florian	8e0f8b2e92	Rename BalanceFetcher to CoinBalanceFetcher	6 years ago
Igor Florian	2a7c2feac2	Rename Balances module to CoinBalances	6 years ago
Felipe Renan	ad50a21c1e	Fix dialyzer	6 years ago
Igor Florian	53e7866016	Fetch TokenBalances value in the Catchup Indexer	6 years ago
Igor Florian	b56f8d4137	Fetch TokenBalances values in the Realtime Indexer	6 years ago
Igor Florian	c2ff17756c	Get TokensBalance in the Blockchain ... We're now able to get the balances from a Address given it's address_hash, block_number and token_contract_address_hash.	6 years ago
Luke Imhoff	c484bc2d25	Separate Indexer.Address.TokenBalances from Indexer.Balances ... TokenBalances need to extract 3 fields, while Balances only need 2, so they shouldn't be mixed together in the same module since the expected params in the set differs.	6 years ago
Luke Imhoff	fdd1090c45	Re-enable Ecto loggers ... Allow Ecto loggers to log to the application-specific log **file**, `log/<Mix.env>/ecto.log` but raise `:console`'s level to `:info`, so that Ecto output does not flood `iex` and make it unusable.	6 years ago
Alex Garibay	cade255e15	Add support for named addresses (#590)	6 years ago
Luke Imhoff	b7c0d188e3	Use handle_continue so that Realtime.init does not block on RPC call	6 years ago
Luke Imhoff	c8c53405fc	Subscribe to newHeads for realtime indexer ... Separate Catchup and Realtime indexer's fetching, so that Catchup continues to use the old polling, `EthereumJSONRPC.json_rpc` while Realtime subscribes with `EthereumJSONRPC.subscribe`. Because Realtime gets notified of new blocks through the subscription, it no longer has a Sequence, so Sequence has been removed from `%BlockFetcher{}`. There is also no block concurrency or block batch size for Realtime, so that was removed from `%BlockFetcher{}` too and moved to `%BlockFetcher.Catchup{}`. Because Realtime requires a websocket to be connected, there is now an `Indexer.BlockFetcher.Realtime.Supervisor` that supervises the `Indexer.BlockFetcher.Realtime.WebSocket` and `Indexer.BlockFetcher.Realtime`. `Indexer.BlockFetcher.Realtime` because it is waiting on notifications from is subscription becomes a `GenServer` insteado of a `Task`, so it can be supervised normally. BlockFetcher.Supervisor becomes a normal `Supervisor` of `Catchup.Supervisor` and `Realtime.Supervisor`. As Realtime is using `EthereumJSONRPC.subscribe`, `subscribe_named_arguments` has been added as a sibling of `json_rpc_named_arguments` in the `:indexer` `config`s.	6 years ago
Luke Imhoff	ae21b7e8b0	Separate sequence from BlockFetcher.t ... Move code dealing with the Sequence to the outer most layer, so that Realtime can stop using it when switching to WebSocket.	6 years ago
Igor Florian	b05196df0f	Create records for TokenBalances in the database ... The TokenBalances will be built in the BlockFetcher and will be stored with value and value_fetch_at as nil. When the TokenBalanceFetcher is plugued this info will be fetched and updated.	6 years ago
Igor Florian	57b6f7487e	Stop indexing Balances from TokenTransfers ... A TokenTransfer doesn't transfer Coin. So we don't need to store Chain.Balance for it since nothing will change in the amount of coin a Address has after this particular event.	6 years ago
Luke Imhoff	ad7e15ca3b	Handle :ignore from EthereumJSONRPC.fetch_internal_transactions ... `c:EthereumJSONRPC.fetch_internal_transactions/2` can return `:ignore` as it does for `EthereumJSONRPC.Geth` because it doesn't currently support internal transactions, but `Indexer.BlockFetcher.Realtime.internal_transactions/2` did not handle `:ignore` and forwarded it along instead. Switch the `with` to a `case` so we don't forward unexpected returns in the future. Make `:ignore` return empty `:internal_transaction_params`.	6 years ago
Alex Garibay	f81d20fc24	Add support for multiple return types in contracts (#562)	6 years ago
Andrew Cravenho	3e8c5b84d6	Changes from POA Explorer to BlockScout	6 years ago
Andrew Cravenho	28e485a5c1	Replace poanetwork/poa-explorer urls with poanetwork/blockscout	6 years ago
Luke Imhoff	bf201f1a52	Elixir 1.7.2 / Erlang 21.0.4	6 years ago
Luke Imhoff	24c2d8e24e	Use nil value to indicate unfetched Balance ... Balances will be created when a given (address_hash, block_number) is first seen, but with a `nil` value. The BalanceFetcher will stream this value is `NULL` rows on restart to initialize the queue.	6 years ago
Alex Garibay	8b16f81b73	Catalog token transfers and tokens during indexing (#484) ... * Fetch token information asynchronously during indexing * Include token transfers imports in main importer * Extract token transfers when importing blocks while indexing * Move ecto to explorer deps * Update References to the Indexer namespace * Add token insertions with main importer * Import token address/type during block indexing	6 years ago
Luke Imhoff	06475f1275	Don't broadcast from catchup ... Fixes rebase picking wrong version with broadcasting turned on for catchup index, which we don't want	6 years ago
Luke Imhoff	81ba3a5b7b	Only use address_hash_to_block_number if block_number is missing	6 years ago
Luke Imhoff	f2884992a7	Fetch realtime internal transactions and balances synchronously ... Fixes #366 Queuing the internal transactions and balances to be fetched asynchronously for realtime blocks lead to upto 15 minute delay between when a block appeared and the internal transaction was drained from the queue. It is a better user experience to not see the realtime blocks at all than to see them incomplete, so fetch the internal transactions and then the balances before importing the whole tree of data instead of using the catchup asynchronous queues.	6 years ago
Luke Imhoff	12f70945da	Switch Chain.import options to maps ... 1. Allows for easier composing with put_in than Keyword.t 2. Allows more precise specs as individal options can be marked as optional or required while the Keyword.t pattern makes it seem like everything is optional.	6 years ago
Luke Imhoff	68229420b4	Move Receipts functions out of BlockFetcher	6 years ago
Luke Imhoff	11735b1289	Separate Catch, Realtime, Supervisor, and BlockFetcher states ... BlockFetcher.Catchup and BlockFetcher.Realtime get their own structs, which use BlockFetcher struct. This better emphasizes that Catchup and Realtime are both block fetchers. The BlockFetcher.Catchup and BlockFetcher.Realtime structs move into Supervisor's struct.	6 years ago
Luke Imhoff	2333cc6cf5	Separate supervision and fetching parts of BlockFetcher ... Move supervision aspect to BlockFetcher.Supervisor and leave only the actual fetching code in BlockFetcher.	6 years ago
Luke Imhoff	e0fed23fad	Fix catchup_block_number not being set ... catchup_block_number can't be set in state because it is set in a separate task, so have the Task return it instead of storing it in a state field.	6 years ago
Luke Imhoff	807f5e6d7c	BlockFetcher.realtime_task -> BlockFetcher.Realtime.task	6 years ago
Luke Imhoff	2afa934225	BlockFetcher.catchup_task -> BlockFetcher.Catchup.task ... Move `catchup_task` out of `BlockFetcher`, so it is easier to spot catchup vs realtime vs common code.	6 years ago
Luke Imhoff	2f3af51741	Unified Explorer.Chain.import ... * Merge `Explorer.Chain.import_blocks` and `Explorer.Chain.import_internal_transactions` into `Explorer.Chain.import`. * Move body of `Explorer.Chain.import` and its private helper functions to `Explorer.Chain.Import` as an independent context that `Explorer.Chain` can use. `Explorer.Chain.Import` must be under `Explorer.Chain` because `Explorer.Chain.Import` still needs direct access to the `Explorer.Chain` schemas.	6 years ago
jimmay5469	e5e695331f	Subscribe to only realtime-indexer events for channel broadcasts	6 years ago