User feedback on HotJar about brew install doc:

https://besu.hyperledger.org/en/latest/HowTo/Get-Started/Install-Binaries/

When I follow the instructions for installing via brew, I get this error. Easy fix, but the docs are wrong:

Error: Formulae found in multiple taps:

• hyperledger/besu/besu
• pegasyseng/pantheon/besu

Please use the fully-qualified name (e.g. hyperledger/besu/besu) to refer to the formula.

Doc trace_transaction API

PR hyperledger/besu#441

Add the recommended size of the Java VM to the docs:
Min 4GB for private. 16GB for mainnet, but really depends on your usage, network setup, etc.

Besu will use the most memory while syncing and then drop back. Use monitoring to see what your node is doing.

@shemnon @joshuafernandes @EdJoJob are the input/reviewers for this.

PR: hyperledger/besu#410

I am blockchain developer. I want to know the besu project if you have some production use case . thanks.

PR: hyperledger/besu#261

Replace the placeholder TLS image

Doc multi-tenancy

PRs:
hyperledger/besu#196

hyperledger/besu#264 hyperledger/besu#196

hyperledger/besu#266

hyperledger/besu#267

hyperledger/besu#272

The following figure caption contains reference to the old project name, Pantheon. This should be changed to Besu. The figure can be found here: http://besu.hyperledger.org/en/latest/Concepts/Permissioning/Permissioning-Overview/

The public transaction gas usage is calculated in the following way:

Gas usage = TX_BASE_COST + (TX_DATA_ZERO * NO_OF_ZEROS) + (TX_DATA_NON_ZERO * NO_OF_NON_ZEROS)

where,

TX_BASE_COST = 21000
TX_DATA_ZERO = 4
NO_OF_ZEROS = number of "00" in the hex payload of the transaction
TX_DATA_NON_ZERO = 68 pre-istanbul / 16 post-istanbul (as in istanbul hard fork)
NO_OF_NON_ZEROS = number of non-"00" in the hex payload of the transaction i.e. number of pairs in the hex that are not "00"

The PMT has 32 bytes of non-zero hex. It therefore consumes:

Gas usage of PMT = 21000 + (4 * 0) + (68 * 32) = 21000 + 2176 = 23176

Post-Instanbul the PMT will consume:

Gas usage of PMT post-istanbul = 21000 + (4 * 0) + (16 * 32) = 21000 + 512 = 21512

Invoking the Privacy Precompiled Contract consumes 0 gas.

More info: https://github.com/ethereum/EIPs/blob/master/EIPS/eip-2028.md

Comments from the original JI https://jira.hyperledger.org/browse/BESU-75:

Madeline Murray - Need to confirm where the extra 2992 is coming from 
 
Madeline Murray - Document where extra gas is coming from.

Also update in Besu and web3js, and web3j
  
Ivaylo Kirilov September 16, 20the changes required in Besu are documentation related only
 
Ivaylo Kirilov September 17, 2019, 12:14 AM
 
fixed in web3j - hyperledger/web3j#1029

We need a wiki article explaining our general release cadence philosophy, CI/CD driven numbering scheme, and rough guidance on quarterly big releases.

From Slack:

i.e. our numbering is ..
We bump milestone when a release is big enough (such as full mainnet compliance). We do a quarterly release where we upgrade all dependencies with a RC release, and bi-weekly all come in CD style. We don’t do multiple branches.

The eea.js library was renamed to web3.js-eea. Update doc throughout

Correct the spelling error on markdown YAML file (mkdocs.yml) at line no. 160 from Vadidating to Validating.

See for instance blocks on https://besu.hyperledger.org/en/latest/Tutorials/Private-Network/Create-Private-Network/

User feedback:
No documented process on how to perform a hardfork and what constitutes a hardfork

PR: hyperledger/besu#440

Users would set these options if they want a longer period of historic data than the defaults for pruning.

See RocketChat thread - https://chat.hyperledger.org/channel/besu?msg=w9Ri86pKg6k6ZMqJW

PR: hyperledger/besu#427

Include in changelog

Markdownlint

Some rules are ignored because markdownlint is not able to handle our extended markdown syntax and more specifically the code blocks.

But we have also other issues like not being able to exclude auto generated files from the check or easily extend the tool.

For this and because this tool is Ruby that we don't really have already in our tool suite preventing us to fix and provide PRs, moving to the Node,js version of the tool would be wise: https://github.com/DavidAnson/markdownlint

Markdown Link Checker

Update dependency to benefit from the latest version on https://github.com/tcort/markdown-link-check for which I am now a maintainer

User feedback about configuration starting a miner and opening RPC.

https://besu.hyperledger.org/en/latest/HowTo/Get-Started/Starting-node/

The configuration example on this page includes two things that seem ill-advised. There should, at least, be a warning to the user about (a) that this configuration starts a miner, and (b) that the RPC is 'wide open'. In the past, people said about both geth and parity that opening one's RPC in this way was asking for trouble. Maybe you shouldn't make that the default without at least a warning to the reader.

Persistent storage of Orion known nodes has been added to enable HA in Consensys/orion#332 and doc'd for Orion in Consensys/doc.orion#38

PR: hyperledger/besu#351

Providing a new RPC method admin_reloadPlugin to dynamically (without the need to restart Besu) reload plugin config (this required a new reloadConfiguration method in Besu plugin api as it’s available for all plugins) see hyperledger/besu#261

Doc automatic log bloom caching

User feedback:
The documentation does not specifically say what the commands for ETH,ADMIN,API,CLIQUE etc do for RPC, this is more of a nice thing as one can figure it out but its not explicitly stated I believe

https://besu.hyperledger.org/en/stable/Concepts/NetworkID-And-ChainID/ explains how to set network ID and chain ID but doesn’t explain exactly what they are.

Requirements:
Add content to network ID and chain ID page explaining what they are and the implications of them being the same/different. eg, two networks with the same network IDs and different chain IDs or two networks with different network IDs and the same chain IDs.

From Hotjar feedback:
This all sounds like the network ID and chain ID are not conceptually the same but in practice they are always the same. So, CAN they be different? Not clear. Would be nice to make it explicit.

User feedback:
The permissioning on-chain smart contracts documentation is lacking and is unclear if its last release (sept, 2019) is the current one that should be used .

To override the default rules provided by markdownlint we need a style file and update the --style=all option in the mdl command in circleci to use this new style.

https://consensys.slack.com/archives/GCEHWHAQ2/p1581509902077000

PR: hyperledger/besu#297

Requirement: mkdocs-redirects==1.0.0
Config in mkdocs.yml plugins section, see https://github.com/datarobot/mkdocs-redirects

The How To section currently has an example of creating a signed tx using Web3.js.

We should add a Web3j example as well.

Requirements: 

API doc for fixed and open APIs.  Need to confirm if this is publishing the JavaDoc or something else 
Plugin HowTo tutorial for simpler plugins 
Talk to Danno Ferrin to confirm requirements/where to start. 

See RocketChat: https://chat.hyperledger.org/channel/besu-contributors?msg=XESD4c7nBH8kBQJMT

Doc trace_block API

PR hyperledger/besu#449

Doc reorg compatible privacy

Include that this is an early access feature in 1.4.

PR: hyperledger/besu#384

User feedback:
IMO there should be a dedicated page on this, as this is a crucial step. There should be documentation on what 'mixhash', 'coinbase', 'chainid' specify on a dedicated page

• An Index of currently used ChainID's is a must

Doc NAT options

PRs:

hyperledger/besu#190

hyperledger/besu#298

Bounty: Consensys/protocol-BountiedWork#2

New CLI options:
--nat-external-ip-usage-enabled

New NAT methods -> DOCKER, KUBERNETES

Update UPnP content: https://besu.hyperledger.org/en/stable/HowTo/Find-and-Connect/Using-UPnP/

See https://chat.hyperledger.org/channel/besu?msg=6fbnzWoKcedFM9PUR

PR: hyperledger/besu#343

Add tips to https://besu.hyperledger.org/en/stable/HowTo/Configure/Consensus-Protocols/IBFT/ and https://besu.hyperledger.org/en/stable/HowTo/Configure/Consensus-Protocols/Clique/ indicating you getSignerMetrics can be used to identify validators that aren't participating.

In docs/HowTo/Find-and-Connect/Managing-Peers.md#limiting-remote-connections, more details required about why limiting remote connections is unnecessary. Perhaps when you should and when you should not limit remote connections.

Some discussion around the desired return values from JSON-RPC methods miner_start and miner_stop came up in the following PR:
hyperledger/besu#104 (comment)

As of #104, miner_start will return true if the node is already mining or mining was successfully started. And miner_stop will return true if the node was already idle or mining was stopped.

An alternative would be to return true if the mining status was actually updated (similar to the behavior of other methods like admin_addPeer / admin_removePeer which return true if the node was successfully modified).

We should decide on the desired behavior and update our documentation to clarify the meaning of the returned boolean: https://besu.hyperledger.org/en/latest/Reference/API-Methods/#miner_start

We will keep the current functionality, but we should update the documentation to be more explicit.

In short
miner_start: returns true if mining was started or if the node was already mining
miner_stop: returns true if mining was stopped or if the node was not mining

PR: hyperledger/besu#250

moved from https://jira.hyperledger.org/browse/BESU-87
Make the Postman file available in the doc website directly to prevent dependency with Postman hosting.

PR for JSON-RPC HTTP: hyperledger/besu#271

If you have pruning or fast sync enabled, you have a full node.
Pruning and fast sync must be disabled to get an archive node.

Full node - the full world state is available for the chain head (and generally a bit before that).
Archive node - the full world state is available for every block
Light node - parts of the world state are requested on demand, but no full world state is kept. Currently Besu doesn't have a light node mode.

Include approx size required for mainnet nodes for full and archive nodes in system requirements.

Create links to the TLS content between the products.

Add known bug caution to pruning CLI option. Reference changelog.

Doc tracing APIs