Dshackle can connect to multiple independent APIs ("upstreams") and provides an unified API on top of it.
Supported upstream protocols:
-
JSON RPC
-
Websockets
-
gRPC (i.e. can connect to another Dshackle)
Those protocols can be configures with additional security, TLS and authentication.
-
Most of Ethereum nodes support WebSocket connection, in addition to the JSON RPC. If it’s available on your node, it’s suggested to configure both JSON RPC and WebSocket connection
-
Bitcoind needs to be configured to index/track addresses that you’re going to request. Make sure you configure it to index your addresses with
importaddress
. If you request balance for an address that is not indexed then it returns 0 balance. -
To track all transactions you need to setup index for transactions, which is disabled by default. Run it with
-reindex
option, or settxindex=1
in the config.
version: v1
cluster:
defaults:
- blockchains:
- ethereum
min-peers: 10
- blockchains:
- kovan
min-peers: 2
upstreams:
- id: us-nodes
blockchain: auto
priority: 10
connection:
dshackle:
url: https://35.226.252.117
tls:
ca: ca.crt
certificate: client.crt
key: client.p8.key
- id: infura-eth
blockchain: ethereum
priority: 5
role: fallback
labels:
provider: infura
disable-validation: true
connection:
ethereum:
rpc:
url: "https://mainnet.infura.io/v3/${INFURA_USER}"
basic-auth:
username: ${INFURA_USER}
password: ${INFURA_PASSWD}
ws:
url: "wss://mainnet.infura.io/ws/v3/${INFURA_USER}"
basic-auth:
username: ${INFURA_USER}
password: ${INFURA_PASSWD}
There are two main segments for upstreams configuration:
-
upstreams - a list of API to connect to, with all configuration specific for upstream and chain
-
and default options as common configuration options applied to all nodes in that group
In the example above we have:
-
default configuration for Ethereum Mainnet which accepts upstream as valid when it not in fast synchronization mode and has at least 10 peers.For Kovan Testnet nodes the requirements are much relieved
-
as upstreams it has 2 configurations
-
balancer connects to another Dshackle/another machine by using gRPC protocol
-
accepts (i.e. proxies) any blockchain available on that remote
-
verifies TLS certificate of the server
-
uses client certificate for authentication, i.e. remote server is accepting only clients authenticated by a certificate
-
-
connects to Infura provided Ethereum Mainnet
-
as a fallback upstream, which means that it’s used only if
us-nodes
fails -
configuration is using placeholders for
${INFURA_USER}
and${INFURA_PASSWD}
which will be replaced with corresponding environment variables values -
uses Basic Authentication to authenticate requests on Infura
-
label
[provider: infura]
is set for that particular upstream, which can be selected during a request.For example for some requests you may want to use nodes with that label only, i.e. "send that tx to infura nodes only", or "read only from archive node, with label [archive: true]" -
upstream validation (peers, sync status, etc) is disabled for that particular upstream
-
-
upstream
us-nodes
has a higher priority thaninfura-eth
so if there is a different block on the same height (i.e., a fork) then Dshackle chooses block fromus-nodes
. Note that it is specific only for Proof-of-Stake blockchains, such as Ethereum.
-
When two different upstream produce a different block it’s critical to route the request traffic to an upstream on the right fork.
With Proof-of-Work blockchains Dshackle chooses the block with more works produced, and that’s the general consensus for PoW blockchains.
For Proof-of-Stake this is impossible and instead of Total Work you have to configure a priority of upstream, i.e., the process is more manual. In a situation when Dshackle sees a fork it chooses the upstreams with the block on the side with the highest priority.
priority
:- id: eth-one
blockchain: ethereum
priority: 10
- id: eth-two
blockchain: ethereum
priority: 20
- id: eth-three
blockchain: ethereum
priority: 15
In this case upstream eth-two
has the highest priority, and it’s blocks are considered as true in case of a fork.
If the upstream is down for a some reason, then blocks from eth-three
are used instead.
Note
|
Priority is a configuration for a Fork decision, and it doesn’t change the priority of the traffic. Except the situation when Dshackle sees a fork between different upstreams; in this case all traffic is routed only to upstreams on the right side of the fork. |
Note
|
Configuration applies only for Proof-of-Stack and for Proof-of-Work blockchains priority options is not used. |
By default, the Dshackle connects to each upstream in a Round-Robin basis, i.e. sequentially one by one. If you need more gradual control over the order of which upstream is used and when you can assign following roles:
-
primary
(default role if nothing specified) -
secondary
-
fallback
Where primary
and secondary
are considered here a standard upstreams, and fallback
is used on failure of standard upstreams.
I.e. the Dshackle always starts with making requests to standard upstreams.
If all of them failed, if responses are inconsistent (ex. for eth_getTransactionCount
), or when it needs to broadcast to a wider network (sendrawtransaction
), then upstreams with role fallback
cames to use.
The internal request order is (goes to next only if all upstreams on current step a not available or failed):
-
tries with primary upstreams
-
tries with secondary upstream
-
… delay (100ms at first, increased each iteration)
-
tries with primary upstreams
-
tries with secondary upstream
-
tries with fallback upstreams
-
… go to step 3
Steps 3-6 are repeated until a valid response received, or a timeout for the original request is reached.
In general:
- you set role secondary
for upstream in another cluster/datacenter - you set role fallback
for an external upstream which may be provided by a third party, and you want to use it as a last resort
Options (default or as part of upstream config):
Option | Default | Description |
---|---|---|
|
false |
if |
|
3 |
specify minimum amount of connected peers, Dshackle will not use upstream with less than specified number |
|
60 |
timeout in seconds after which request to the upstream will be discarded (and may be retried on an another upstream) |
|
|
specify if this node should be used to fetch balance for an address |
Dshackle currently supports
-
rpc
a standard Ethereum JSON RPC -
ws
websocket connection (supposed to be used in addition torpc
connection) -
dshackle
connects to another Dshackle instance
-
getbestblockhash
-
getblock
-
getblocknumber
-
getblockcount
-
gettransaction
-
getrawmempool
-
getrawtransaction
-
gettxout
-
getreceivedbyaddress
-
listunspent
-
sendrawtransaction
-
getmemorypool
-
getconnectioncount
-
getnetworkinfo
-
eth_gasPrice
-
eth_call
-
eth_estimateGas
-
eth_getBlockTransactionCountByHash
-
eth_getUncleCountByBlockHash
-
eth_getBlockByHash
-
eth_getTransactionByHash
-
eth_getTransactionByBlockHashAndIndex
-
eth_getStorageAt
-
eth_getCode
-
eth_getUncleByBlockHashAndIndex
-
eth_getTransactionCount
-
eth_blockNumber
-
eth_getBalance
-
eth_sendRawTransaction
-
eth_getBlockTransactionCountByNumber
-
eth_getUncleCountByBlockNumber
-
eth_getBlockByNumber
-
eth_getTransactionByBlockNumberAndIndex
-
eth_getTransactionReceipt
-
eth_getUncleByBlockNumberAndIndex
-
eth_feeHistory
-
eth_getLogs
-
net_version
-
net_peerCount
-
net_listening
-
web3_clientVersion
-
eth_protocolVersion
-
eth_syncing
-
eth_coinbase
-
eth_mining
-
eth_hashrate
-
eth_accounts
It’s possible to enable additional methods that are available on upstream, or disable an existing method.
For that purpose there is methods
configuration:
upstreams:
- id: my-node
blockchain: ethereum
labels:
archive: true
methods:
enabled:
- name: trace_transaction
disabled:
- name: eth_getBlockByNumber
Such configuration option allows executing method trace_transaction
and also disables eth_getBlockByNumber
on that particular upstream.
If a client requests to execute method trace_transaction
then it will be scheduled to that upstream (or any upstream with such method enabled).
Note
|
It’s especially useful when used together with upstream labels.If an archive upstream has label archive: true it’s possible to specify that the client wants to execute method trace_transaction only on an archive node(s), which has complete historical data for tracing.
|
You can overwrite existing methods or add new ones using a static response:
upstreams:
- id: my-node
blockchain: ethereum
methods:
enabled:
- name: net_version
static: "\"100000\""
- name: eth_chainId
static: "0x186a0"
- name: eth_custom_array
static: '["custom_array_response"]'
- name: eth_custom_bool
static: "false"
All connection types can use TLS secured connection, with optional client certificate authentication:
-
ca
path to certificate required from remote server -
optional
certificate
andkey
for client authentication.
Note
|
Please note that key must be encoded with PKCS 8
|