diff --git a/pages/builders/chain-operators/tools/_meta.json b/pages/builders/chain-operators/tools/_meta.json index 908ae60d2..4c5077834 100644 --- a/pages/builders/chain-operators/tools/_meta.json +++ b/pages/builders/chain-operators/tools/_meta.json @@ -1,4 +1,5 @@ { + "chain-monitoring": "Chain Monitoring", "op-challenger": "Configure Challenger For Your Chain", "op-conductor": "Conductor", "op-deployer": "Deployer", diff --git a/pages/builders/chain-operators/tools/chain-monitoring.mdx b/pages/builders/chain-operators/tools/chain-monitoring.mdx new file mode 100644 index 000000000..c97183443 --- /dev/null +++ b/pages/builders/chain-operators/tools/chain-monitoring.mdx @@ -0,0 +1,128 @@ +--- +title: Chain Monitoring Options +lang: en-US +description: Learn about onchain and offchain monitoring options for your OP Stack chain. +--- + +import { Callout } from 'nextra/components' + +# Chain Monitoring Options + +This explainer covers the basics of onchain and offchain monitoring options for your OP Stack chain. Onchain monitoring services allow chain operators to monitor the overall system and onchain events. +Offchain monitoring lets chain operators to monitor the operation and behavior of nodes and other offchain components. + +## Onchain Monitoring Services + +Onchain monitoring services provide insights into the overall system, helping chain operators track and monitor on-chain events. Some examples of onchain monitoring services include `monitorism` and `dispute-mon`. + +### `monitorism` + +Monitorism is a tooling suite that supports monitoring and active remediation actions for the OP Stack chain. Monitorism uses monitors as passive security providing automated monitoring for the OP Stack. They are used to monitor the OP stack and alert on specific events that could be a sign of a security incident. + +Currently. the list of monitors includes: + +Security integrity monitors: These are monitors necessary for making sure Bridges between L2 and L1 are safe and work as expected. These monitors are divided in two subgroups: + +* Pre-Faultproof Chain Monitors: + * Fault Monitor: checks for changes in output roots posted to the L2OutputOracle contract. When a change is detected, it reconstructs the output root from a trusted L2 source and looks for a match.. + * Withdrawals Monitor: checks for new withdrawals that have been proven to the OptimismPortal contract. Each withdrawal is checked against the `L2ToL1MessagePasser` contract. +* Faultproof chain monitors: + * Faultproof Withdrawal: The Faultproof Withdrawal component monitors `ProvenWithdrawals` events on the `OptimismPortal` contract and performs checks to detect any violations of invariant conditions on the chain. If a violation is detected, the issue is logged, and a Prometheus metric is set for the event. This component is designed to work exclusively with chains that are already utilizing the Fault Proofs system. This is a new version of the deprecated `chain-mon`, `faultproof-wd-mon`. For detailed information on how the component works and the algorithms used, please refer to the component README. + +Security monitors: Those tools monitor other aspects of several contracts used in optimism: + +* Global Events Monitor: made for taking YAML rules as configuration and monitoring the events that are emitted on the chain. +* Liveness Expiration Monitor: monitors the liveness expiration on Safes. +* Balances Monitor: emits a metric reporting the balances for the configured accounts. +* Multisig Monitor: The multisig monitor reports the paused status of the OptimismPortal contract. "If set, reports the latest nonce of the configured Safe address and the latest presigned nonce stored in One Password.. The latest presigned nonce is identified by looking for items in the configured vault that follow a `ready-.json` name. The highest nonce of this item name format is reported. +* Drippie Monitor: tracks the execution and executability of drips within a Drippie contract. +* Secrets Monitor: takes a Drippie contract as a parameter and monitors for any drips within that contract that use the `CheckSecrets` dripcheck contract. `CheckSecrets` is a dripcheck that allows a drip to begin once a specific secret has been revealed (after a delay period) and cancels the drip if a second secret is revealed. Monitoring these secrets is important, as their revelation may indicate that the secret storage platform has been compromised and someone is attempting to exfiltrate the ETH controlled by the drip. + +For more information on these monitors and how to use them, [check out the repo](https://github.com/ethereum-optimism/monitorism?tab=readme-ov-file#monitorism). + +### `dispute-mon` + +Chain operators should consider running `op-dispute-mon`. It's an essential security monitoring service that tracks game statuses, providing visibility over the last 28 days. + +`dispute-mon` is set up and built the same way as `op-challenger`. This means that you can run it the same way (run `make op-dispute-mon` in the directory). + +A basic configuration option would look like this: + +``` +OP_DISPUTE_MON_LOG_FORMAT=logfmt +OP_DISPUTE_MON_METRICS_ENABLED=true +OP_DISPUTE_MON_METRICS_ADDR=0.0.0.0 +OP_DISPUTE_MON_METRICS_PORT=7300 + +OP_DISPUTE_MON_L1_ETH_RPC=.. +OP_DISPUTE_MON_ROLLUP_RPC=.. +OP_DISPUTE_MON_GAME_FACTORY_ADDRESS=.. + +OP_DISPUTE_MON_HONEST_ACTORS=.. +``` + +`OP_DISPUTE_MON_HONEST_ACTORS` is a CSV (no spaces) list of addresses that are used for the honest `op-challenger` instances. + +Additional flags: + +* `OP_DISPUTE_MON_GAME_WINDOW`: This is the window of time to report on games. It should leave a buffer beyond the max game duration for bond claiming. If Fault Proof game parameters are not changes (e.g. MAX\_CLOCK\_DURATION), it is recommended to leave this as the default. +* `OP_DISPUTE_MON_MONITOR_INTERVAL`: The interval at which to check for new games. Defaults to 30 seconds currently. +* `OP_DISPUTE_MON_MAX_CONCURRENCY`: The max thread count. Defaults to 5 currently. + +You can find more info on `op-dispute-mon` on [the repo](https://github.com/ethereum-optimism/optimism/tree/develop/op-dispute-mon). + +Chain operators can easily create their grafana dashboard for Dispute Monitor using the following json file: [Download the Dispute Monitor JSON](/grafana/dispute-monitor-1718214549035.json). + +## Offchain Component Monitoring + +Offchain monitoring allows chain operators to monitor the operation and behavior of nodes and other offchain components. Some of the more common components that you'll likely want to monitor include `op-node`, `op-geth`, `op-proposer`, `op-batcher`, and `op-challenger`. +The general steps for enabling offchain monitoring is pretty consistent for all the OP components: + +1. Expose the monitoring port by enabling the `-—metrics.enabled` flag +2. Customize the metrics port and address via the `--metrics.port` and `--metrics.addr` flags, respectively +3. Use [Prometheus](https://prometheus.io/) to scrape data from the metrics port +4. Save the data in `influxdb` +5. Share the data with [Grafana](https://grafana.com/) to build your custom dashboard + +### `op-node` + +`op-node` metrics and monitoring is detailed in the [Node Metrics and Monitoring](/builders/node-operators/management/metrics) guide. To enable metrics, pass the `--metrics.enabled` flag to `op-node` and follow the steps above for customization options. +See [this curated list](/builders/node-operators/management/metrics#important-metrics) for important metrics to track specifically for `op-node`. + +### `op-geth` + +To enable metrics, pass the `--metrics.enabled` flag to the op-geth. You can customize the metrics port and address via the `--metrics.port` and `--metrics.addr` flags, respectively. + +### `op-proposer` + +To enable metrics, pass the `--metrics.enabled` flag to the op-proposer. You can customize the metrics port and address via the `--metrics.port` and `--metrics.addr` flags, respectively. + +You can find more information about these flags in our [Proposer configuration doc](https://docs.optimism.io/builders/chain-operators/configuration/proposer#metricsenabled). + +### `op-batcher` + +To enable metrics, pass the `--metrics.enabled` flag to the op-batcher. You can customize the metrics port and address via the `--metrics.port` and `--metrics.addr` flags, respectively. + +You can find more information about these flags in our [Batcher configuration doc](https://docs.optimism.io/builders/chain-operators/configuration/batcher#metricsenabled). + +### `op-challenger` + +The `op-challenger` operates as the *honest actor* in the fault dispute system and defends the chain by securing the `OptimismPortal` and ensuring the game always resolves to the correct state of the chain. +For verifying the legitimacy of claims, `op-challenger` relies on a synced, trusted rollup node as well as a trace provider (e.g., [Cannon](/stack/protocol/fault-proofs/cannon)). See the [OP-Challenger Explainer](/stack/protocol/fault-proofs/challenger) for more information on this service. + +To enable metrics, pass the `--metrics.enabled` flag to `op-challenger` and follow the steps above for customization options. + +``` + --metrics.addr value (default: "0.0.0.0") ($OP_CHALLENGER_METRICS_ADDR) + Metrics listening address + + --metrics.enabled (default: false) ($OP_CHALLENGER_METRICS_ENABLED) + Enable the metrics server + + --metrics.port value (default: 7300) ($OP_CHALLENGER_METRICS_PORT) + Metrics listening port +``` + +## Next Steps + +* If you encounter difficulties at any stage of this process, please reach out to [developer support](https://github.com/ethereum-optimism/developers/discussions). diff --git a/words.txt b/words.txt index 53e4d2a99..025833397 100644 --- a/words.txt +++ b/words.txt @@ -82,6 +82,8 @@ Discv discv DIVU Drand +dripcheck +Drippie Eigen ENABLEDEPRECATEDPERSONAL enabledeprecatedpersonal @@ -95,12 +97,15 @@ ETHSTATS ethstats EVMTIMEOUT evmtimeout +executability +exfiltrate EXITWHENSYNCED exitwhensynced EXTRADATA extradata Farcaster farcaster +Faultproof FDLIMIT fdlimit featureset @@ -187,6 +192,7 @@ minsuggestedpriorityfee Mintable Mintplex MIPSEVM +Monitorism Moralis Mordor mountpoint @@ -342,7 +348,6 @@ syncmode SYNCTARGET synctarget syscalls -Terraform therealbytes Thirdweb threadcreate