interesting
/
go-ethereum
mirror of https://github.com/ethereum/go-ethereum.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

add logs page and update mining page

This commit is contained in:
Joe 2022-10-05 10:37:08 +01:00
parent e54b879469
commit 07137a2ec3
2 changed files with 46 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
src/pages/docs/fundamentals/logs.md
View File

@ -5,6 +5,9 @@ description: Geth's log messages explained
A Geth node continually reports messages to the console allowing users to monitor Geth's current status in real-time. The logs indicate when Geth is running normally and indicates when some attention is required. However, reading these logs can be difficult for new users. This page will help to interpret the log messages to better understand what Geth is doing.
Note that there are a large number of log messages covering a wide range of possible scenarios for a Geth node. This page will only address a subset of commonly seen messages. For more, see the [Geth Github](https://github.com/ethereum/go-ethereum), [Discord](https://discord.gg/WHNkYDsAKU) or search on [ethereum.stackexchange](https://ethereum.stackexchange.com/). Log messages are usually sufficiently self-descrining that they do not require additional explanation.
## Configuring log messages
Log messages are displayed to the console by default. The messages can be tuned to be more or less detailed by passing `--verbosity` and a value between 0 and 6 to Geth at startup:
@ -28,12 +31,12 @@ geth --verbosity 5 >> /path/eth.log 2>&1
### Startup
When Geth starts up it immediately reports a fairly long page of configuration details and status reports that allow the user to confirm Geth is on the right network and operating in its intended modes. The b asic structure of a log message is as follows:
When Geth starts up it immediately reports a fairly long page of configuration details and status reports that allow the user to confirm Geth is on the right network and operating in its intended modes. The basic structure of a log message is as follows:
```
MESSAGE_TYPE [MONTH-DAY][TIME] MESSAGE VALUE
MESSAGE_TYPE [MONTH-DAY][TIME] MESSAGE VALUE
```
Where `MESSAGE_TYPE` can be `INFO`, `WARN`, `ERROR` or `DEBUG`. These tags categorize log messages according to their purpose. `INFO` messages inform the user about Geth's current configuration and status. `WARN` messages are for alerting the user to details that may affect the way Geth is going to continue running. `ERROR` messages are for alerting the user to problems. `DEBUG` is for messages that are relevant to troubleshooting or for developers working on Geth.
Where `MESSAGE_TYPE` can be `INFO`, `WARN`, `ERROR` or `DEBUG`. These tags categorize log messages according to their purpose. `INFO` messages inform the user about Geth's current configuration and status. `WARN` messages are for alerting the user to details that affect the way Geth is running. `ERROR` messages are for alerting the user to problems. `DEBUG` is for messages that are relevant to troubleshooting or for developers working on Geth.
The messages displayed on startup break down as follows:
@ -53,7 +56,7 @@ INFO [10-04|10:20:52.129] Writing default main-net genesis block
INFO [10-04|10:20:52.372] Persisted trie from memory database nodes=12356 size=1.78MiB time=21.535537ms gcnodes=0 gcsize=0.00B gctime=0s livenodes=1 livesize=0.00B
```
This block of logs shows the user that the node is connecting to Ethereum Mainnet and some low level configuration details. The cache size is bumped to the Mainnet default (4096). The maximum peer count is the highest number of peers this node is allowed to connect to and can be used to control the bandwidth requirements of the node. Logs relating to `ethash` are out of date since Ethereum moved to proof-of-stake based consensus and can safely be ignored.
The logs above show the user that the node is connecting to Ethereum Mainnet and some low level configuration details. The cache size is bumped to the Mainnet default (4096). The maximum peer count is the highest number of peers this node is allowed to connect to and can be used to control the bandwidth requirements of the node. Logs relating to `ethash` are out of date since Ethereum moved to proof-of-stake based consensus and can safely be ignored.
```
@ -119,41 +122,69 @@ The default for Geth is to sync in snap mode. This requires a block header to be
Assuming Geth has a synced consensus client and some peers it will start importing headers, block bodies and receipts. The log messages for data downloading look as follows:
```
INFO [07-28|10:29:49.681] Block synchronisation started
INFO [07-28|10:29:50.427] Imported new block headers count=1 elapsed=253.434ms number=12,914,945 hash=ee1a08..9ce38a
INFO [07-28|10:30:00.224] Imported new block receipts count=64 elapsed=13.703s number=12,914,881 hash=fef964..d789fc age=18m5s size=7.69MiB
INFO [07-28|10:30:18.658] Imported new block headers count=1 elapsed=46.715ms number=12,914,946 hash=7b24c8..2d8006
INFO [07-28|10:30:21.665] Imported new state entries
```
For state data
For state sync, Geth reports when the state heal is in progress. This can takea long time. The log message includes values for the number of `accounts`, `slots`, `codes` and `nodes` that were downloaded in the current healing phase, and the pending field is the number of state entires waiting to be downloaded. The `pending` value is not necessarily the number of state entries remaining until the healing is finished. As the blockchain progresses the state trie is updated and therefore the data that needs to be downloaded to heal the trie can increase as well as decrease over time. Ultimately, the state should heal faster than the blockchain progresses so the node can get in sync. When the state healing is finished there is a post-sync snapshot generation phase. The node is not in sync until the state healing phase is over.
```
INFO [07-28|10:30:21.965] State heal in progress accounts=169,633@7.48MiB slots=57314@4.17MiB codes=4895@38.14MiB nodes=43,293,196@11.70GiB pending=112,626
INFO [07-28|10:30:21.969] State heal in progress accounts=169,633@7.48MiB slots=57314@4.17MiB codes=4895@38.14MiB nodes=43,293,196@11.70GiB pending=112,626
INFO [09-06|01:31:59.885] Rebuilding state snapshot
INFO [09-06|01:31:59.910] Resuming state snapshot generation root=bc64d4..fc1edd accounts=0 slots=0 storage=0.00B dangling=0 elapsed=18.838ms
```
There are other log messages that are commonly seen during syncing. For example:
```sh
WARN [09-28|11:06:01.363] Snapshot extension registration failed
```
This warning is nothing to worry about - it is reporting a configuration mismatch between the node and a peer. It does not mean syncing is stalling or failing, it simply results in the peer being dropped and replaced.
```sh
# consensus client has identified a new head to use as a sync target - account for this in state sync
INFO [10-03|15:34:01.336] Forkchoice requested sync to new head number=15,670,037 hash=84d4ec..4c0e2b
```
The message above indicates that the fork choice algorithm, which is run by the consensus client, has identified a new target Geth should sync up to. This redirects the sync to prevent syncing to an outdated target and is a natural part of syncing a live blockchain.
## Common Error messages
## Transaction logs
Transactions submitted over local IPC, Websockets or HTTP connections are reported in the console logs. For example, a simple ETH transaction appears in the console logs as follows:
```sh
INFO [09-06|01:31:59.910] Submitted transaction hash=0x2893b70483bf1791b550e5a93763058b0abf7c6d9e6201e07212dbc64d4764532 from: 0xFB48587362536C606d6e89f717Fsd229673246e6 nonce: 43 recipient: 0x7C60662d63536e89f717F9673sd22246F6eB4858 value: 100,000,000,000,000,000
```
Other user actions have similar log messages that are displayed to the console.
## Common warnings
There are many warnings that can be emitted by Geth as part of its normal operation. However, some are asked about especially frequently on the [Geth Github](https://github.com/ethereum/go-ethereum) and [Discord](https://discord.gg/WHNkYDsAKU) channel.
```sh
# the following indicates a request timeout, may indicate IO speed is low.
WARN [10-03|18:00:40.413] Unexpected trienode heal packet peer=9f0e8fbf reqid=6,915,308,639,612,522,441
```
The above is often seen and misinterpreted as a problem with snap sync. In reality, it indicates a request timeout that may be because I/O speed is low. It is usually not an issue, but if this message is seen very often over prolonged periods of time it might be rooted in a local connectivity or hardware issue.
```sh
# no consensus client
WARN [10-03|13:10:26.441] Post-merge network, but no beacon client seen. Please launch one to follow the chain!
```
The above message is emitted when Geth is run without a consensus client on a post-merge proof-of-stake network. Since Ethereum moved to proof-of-stake Geth alone is not enough to follow the chain because the consensus logic is now implemented by a separate piece of software called a consensus client. This log message is displayed when the consensus client is missing. Read more about this on our [consensus clients](/docs/interface/consensus-clients.md) page.
```sh
# consensus client is not working, most likely it is not in sync yet - swapping cl to checkpoint sync is fastest
WARN [10-03 |13:10:26.499] Beacon client online, but never received consensus updates. Please ensure your beacon client is operational to follow the chain!
```
```
The message above indicates that a consensus client is present but not working correctly. The most likely reason for this is that the client is not yet in sync. Waiting for the consensus client to sync should solve the issue.
## Summary
There are a wide range of log messages that are emitted while Geth is running. The level of detail in the logs can be configured using the `verbosity` flag at startup. This page has outlined some of the common messages users can expect to see when Geth is run with default verbosity, without attempting to be comprehensive. For more, please see the [Geth Github](https://github.com/ethereum/go-ethereum) and [Discord](https://discord.gg/WHNkYDsAKU).

4
src/pages/docs/fundamentals/mining.md
View File

@ -2,6 +2,7 @@
title: Proof-of-work mining with Ethash
description: Introduction to proof-of-work mining with Geth
---
{% include note.html content=" Proof-of-work mining is no longer used to secure Ethereum Mainnet. The information below is included because the Ethash code is still part of Geth and it could be used to create a private proof-of-work network or testnet." %}
Blockchains grow when individual nodes create valid blocks and distribute them to their peers who check the blocks and add them to their own local databases.
@ -9,7 +10,8 @@ Nodes that add blocks are rewarded with ether payouts. On Ethereum Mainnet, the
Ethereum wasn't always secured this way. Originally, a proof-of-work based consensus mechanism was used instead. Under proof-of-work, block producers are not selected randomly in each slot. Instead they compete for the right to add a block. The node that is fastest to compute a certain value that can only be found using brute force calculations is the one that gets to add a block. Only if a node can demonstrate that they have calculated this value, and therefore expended energy, will their block be accepted by other nodes. This process of creating blocks and securing them using proof-of-work is known as "mining".
Much more information about mining, including details about the specific algorithm ("Ethash") used by Ethereum nodes is available on [ethereum.org](http://ethereum.org).
Much more information about mining, including details about the specific algorithm
("Ethash") used by Ethereum nodes is available on [ethereum.org](https://ethereum.org/en/developers/docs/consensus-mechanisms/pow/mining-algorithms/ethash).
## CPU vs GPU