dfuse-eosio
dfuse-eosio copied to clipboard
dfuse-io
dfuse for EOSIO
make mindreader keep a few last blocks (ex: 120 last blocks) so it can send them when burst is requested followup from closed issue #169
When loading statedb shards into the kv store, you only get a report every time a new file is loaded. If the files are pretty big, then you don't get...
The format of /healthz is not consistent across components. For example: eosws: ``` {"errors":[],"hub":{"head_block_num":146217680,"head_block_time":"2020-10-09T14:30:44.5Z","time_latency_sec":0},"healthy":{"hub":true}} ``` statedb: ``` ready ``` eosq: ``` ok ``` Of course it is reasonable that different...
When the connection to etcd is broken and etcd is replaced by a different instance, dfuse search fails to update its connection and stays broken. Also the health reports as...
Some queries take a long time because they search a lot of indexes, even though the results returned are very small. Would like to be able to specify the amount...
If the statedb-injector is running live and for some reason gets stuck injecting new k/v state, it will continue to pre-process arriving blocks. pre-processing is generally a good thing. But...
dfuse prometheus metrics should be prefixed with dfuse_* Currently they have no prefix so looking at the list of available metrics in the prometheus make it hard to see which...
When the buffer is too small, mindreader logs a lot like this: ``` (mindreader-stdin) reading from console logs (mindreader/mindreader.go:245){"error": "bufio.Scanner: token too long"} ``` This happens 100s of times per...
The `maximumSupply` and `totalSupply` values returned through tokenmeta are showing up the same, even when they should be different according to the `stat` table. Example: https://bloks.io/account/eosdtsttoken?loadContract=true&tab=Tables&table=stat&account=eosdtsttoken&scope=EOSDT&limit=100 https://mainnet.eos.dfuse.io/graphiql/?query=cXVlcnkgewogIHRva2VucyAodG9rZW5TeW1ib2xzOiJFT1NEVCIpIHsgCiAgICBibG9ja1JlZiB7CiAgICAgIGlkCiAgICAgIG51bWJlcgogICAgfQogICAgcGFnZUluZm8gewogICAgICBzdGFydEN1cnNvcgogICAgICBlbmRDdXJzb3IKICAgIH0KICAgIGVkZ2VzIHsKICAgICAgY3Vyc29yCiAgICAgIG5vZGUgewogICAgICAgIHN5bWJvbAogICAgICAgIGNvbnRyYWN0CiAgICAgICAgaG9sZGVycwogICAgICAgIHRvdGFsU3VwcGx5CiAgICAgICAgbWF4aW11bVN1cHBseQogICAgICAgIAogICAgICB9CiAgICB9CiAgfQp9 @jubeless
If you are running mindreader with the `mindreader-oneblock-suffix` option, and also running multiple mindreaders at once (for redundancy), then you end up with a lot of one-block files. Currently the...