At https://docs.prylabs.network/docs/prysm-usage/parameters/#beacon-node-parameters it has "Define a mainchain web3 provider string endpoint. Can be either a IRC file strong or a WebSocket endpoint." The double typo led me to think IPC couldn't be used, please change to "IPC file string". May have confused others to as I was told in discord it had to be http.

https://www.reddit.com/r/ethereum/comments/fo1p2b/ethereum_on_arm_nethermind_and_hyperledger_besu/

Running: sudo systemctl start/restart/status prysm-beacon or sudo systemctl start/restart/status prysm-validator

To add flags to beacon/validator system files:
cd /etc/ethereum
nano prysm-beacon.conf / prysm-validator.conf

The docusaurus v2 docs seem pretty great at this point and seems like we might be able to take the leap. It includes great features such as a rich plugin ecosystem. Opening this issue for tracking.

New functionality has been added to Prysm to support loading flag values from a yaml file. prysmaticlabs/prysm#4878

The usage is straightforward, you may create a yaml file as a simple key value representation of the flags you wish to provide.

For example, if we usually run with --datadir=/data --enable-ssz-cache we could create a file like

datadir: "/data"
enable-ssz-cache: true

Then the user would provide the runtime flag to Prysm of --config-file=/path/to/file.yaml along with any other runtime flags they wish.

An important note is that flags via command line and via config file can both be present but the config file has the priority. In other words, if we set the flag --datadir=/data2 and datadir: "/data1" in the config.yaml file, Prysm would understand that it should write to /data1.

Please add this documentation to the website and any relevant markdown in Prysm. Thanks

It would be great to have a list of all validator phases and what they mean for the user.
e.g unknown, deposited, ..

Need to write a page that lays out the many stages in the lifecycle of a typical validator.

See prysmaticlabs/prysm#5362 for more context.

Please wait until Saturday (4/11) to merge a PR.
We are targeting a testnet restart over the weekend, but may be delayed.

The high level # headers on each page are currently title capitalised. I'd prefer to see them changed to sentence cap, like the lower level headers. These are located on the top of each .md document under the /docs/ directory

As mentioned prysmaticlabs/prysm#5663 and prysmaticlabs/prysm#5794

It would be good to make a note to users (especially those who run on cloud ISPs) to blacklist internal IPs so they don't get stopped by ISP.

maybe giving example with --p2p-blacklist 10.0.0.0/8,172.16.0.0/12,192.168.0.0/16,100.64.0.0/10,169.254.0.0/16

https://docs.prylabs.network/docs/prysm-usage/slasher/

Under: Running slasher with Docker on Windows

Was informed on discord I needed a --network="host" param added to get slasher started properly.

So:

docker run -it --network="host" -v c:/prysm:/data gcr.io/prysmaticlabs/prysm/slasher:latest

instead of

docker run -it -v c:/prysm:/data gcr.io/prysmaticlabs/prysm/slasher:latest

Otherwise I ran into a bunch of errors then the process died. Would hope someone else would test this to be sure but that fixed it for me. Here is the error output for without that param added while also running a beacon and multiple validators concurrently on the same win machine in docker:

docker run -it -v c:/prysm:/data gcr.io/prysmaticlabs/prysm/slasher:latest
[2020-05-08 04:35:59] WARN flags: Using default mainnet config
[2020-05-08 04:35:59] INFO node: Checking DB database-path=/home/.eth2
[2020-05-08 04:35:59] INFO RPC-API listening on port address=:5000
[2020-05-08 04:35:59] WARN You are using an insecure gRPC connection! Provide a certificate and key to connect securely[2020-05-08 04:35:59] WARN beaconclient: You are using an insecure gRPC connection to beacon chain! Please provide a certificate and key to use a secure connection
[2020-05-08 04:35:59] INFO beaconclient: Successfully started gRPC connection
[2020-05-08 04:35:59] ERROR beaconclient: Could not fetch sync status error=rpc error: code = Unavailable desc = connection error: desc = "transport: Error while dialing dial tcp 127.0.0.1:4000: connect: connection refused"
[2020-05-08 04:35:59] INFO beaconclient: Waiting for beacon node to be fully synced...
[2020-05-08 04:36:11] ERROR beaconclient: Could not fetch sync status error=rpc error: code = Unavailable desc = connection error: desc = "transport: Error while dialing dial tcp 127.0.0.1:4000: connect: connection refused"
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0xd4b17a]

goroutine 79 [running]:
github.com/prysmaticlabs/prysm/slasher/beaconclient.(*Service).querySyncStatus(0xc00057a6c0, 0x11b4800, 0xc0005d0500)
slasher/beaconclient/chain_data.go:67 +0x5ba
github.com/prysmaticlabs/prysm/slasher/beaconclient.(*Service).Start(0xc00057a6c0)
slasher/beaconclient/service.go:180 +0x4df
created by github.com/prysmaticlabs/prysm/shared.(*ServiceRegistry).StartAll
shared/service_registry.go:46 +0x23e

Considering 95% of the formatting stuff in the documentation is super secret knowledge that exists only in my head, there definitely needs to be a styleguide containing every format detail in order to retain consistency across future contributor PRs and reduce the amount of markdown review on my part. This will be complimented by a documentation contribution guide, procedures for efficient writing, running checks, etc.

Needs to be made clearer that the linked guide is still available and supported (but given its third party can still not be fully recommended for production)

If a user is unfamiliar with grpc, it is not immediately clear that they will have to use

--grpc-gateway-port

to enable our HTTP Rest API in the beacon node. We should make this clear in our documentation on how it works.

Currently there is no way for the average user to know on how to backup their db , and where the backup db will be kept. This would be a useful information to include in our docs portal.

Hi team,

This is a tracking issue to keep tabs of the compiled lists of necessary changes after each future monthly documentation sweep. If you're completely lost right now, reach out to Preston. :)

The idea is to take 15 mins to a half hour once a month to read the manual cover-to-cover as group and collab to figure out what needs changing/removing/adding and discuss recent repeat questions to add onto the FAQ or elaborate on. At the end of each month, I'll upload them into checklist form here to cover progress.

DAY OF MONTH: 3rd Tuesday
TIME OF DAY: 5 PM UTC

DATE OF NEXT MEETING: May 18 2020

URL from the About section in https://github.com/prysmaticlabs/documentation to docs.prylabs.network returns not found 404.

Should redirect to https://docs.prylabs.network/docs/getting-started ?

The purpose of this is to ensure that page descriptions generated on search engines and social media are relevant and lack any markdown. This is probably going to be a pain in the wazoo, but I see it going down 1 of 2 ways:

1. Hardcoding the meta descriptions directly into the Markdown files. As HTML/CSS can be passed through the markdown and docusaurus will generate it, simply adding the below to each page:
   <meta name="description" content="bananas are great">

2. Use js and fill the content field with the existing page specific variables included with docusaurus, and putting all of the other contents in a js or json file. This may be more elaborate than it's worth frankly, but may make maintenance easier to have them all in a single file.

Non-urgent, will tackle when I can, though if anyone feels like they can take this on -- go for it. 🙂

Update number of shards in ethereum-2.0.md from 1024 to 64

Feature request: add documentation page with a full, in-depth, rationale around the design choice around bazel build tooling and why it is critical to the success of Prysm.

Users may see WARNING: This key is not certified with a trusted signature! when installing Prysm via the shell script. This issue is to add documentation to add Prylabs signing key as a trusted key.

Hi all,

The :latest docker tag is risky as it can have bugs from latest master that are not yet proven stable. Instead, we should be recommending the latest stable release. Maybe we can create a :stable docker tag which utilizes the latest release for Prysm for both the beacon node and the validator.

In the Activating a Validator portion of the documentation, the following links in "Step 1" are currently broken:

• https://docs.prylabs.network/docs/install/mac/mac
• https://docs.prylabs.network/docs/install/mac/activating-a-validator/docker
• https://docs.prylabs.network/docs/install/mac/activating-a-validator/bazel

We should have a dedicated page that is thorough regarding Go guidelines for Prysm developers, along with testing standards for anyone contributing to the repository.

We should use our docs portal for all our READMEs and key information for Prysm. Currently, we have instructions for running our slasher here but we should create a dedicated item in the sidebar of the docs portal.

The footer on mobile is not aligning properly, as seen by the picture below. It is managed by the footer.js file and requires basic HTML/CSS knowledge. If anyone wants to tackle a quick fix, would be greatly appreciated!

We recently published recommended system requirements. These would be helpful to have somewhere in the docs pages.

https://medium.com/prysmatic-labs/introducing-topaz-testnet-8e8a4e00a700

docs/install/arm.md, docs/install/linux.md and docs/install/mac.md all link to:

https://docs.prylabs.network/docs/install/prysm-usage/parameters

But this link gives a 404.

Currently we publish only ports 4000 , 13000 , whereas now we use 12000 for discovery. Docker instructions created should reflect this.

These instructions should be part of the recommended setup.

Soon, Prysmatic Labs will start discouraging use of their hosted ETH1 nodes so the documentation should reflect a user path for operating a ETH1 node.

Suggested format / paths:

• Run your own ETH1 goerli node (RECOMMENDED)
• Connect to a hosted node, provide a link to infura and provide the flags to connect to Prysm's nodes.

The recommended path is to reduce the centralization risk and mimic a more realistic scenario. Prysm will continue to provide a public ETH1 node for ease of use but users should strongly consider running their own ETH1 node if they are running a beacon node and validator.

Add key pair generation instructions for openssl and usage alongside grpc

Would be great to improve our FAQ with important networking related questions that pop-up in our discord such as:

1. To run a beacon node, do I need a static IP?
2. If yes, does a static IPv6 also work? How can I use IPv6 with beacon nodes?
3. Is it possible run multiple beacon nodes on same server if we link to different port and static IPs?

Most links are misaligned, resulting in broken links. I'll provide a few examples.

On https://docs.prylabs.network/docs/how-prysm-works/architecture-overview/:

• "this section" is: https://docs.prylabs.network/docs/how-prysm-works/introduction/ethereum-2.0 should be: https://docs.prylabs.network/docs/introduction/on-ethereum-2/
• "proof-of-stake validators is: https://docs.prylabs.network/docs/glossaries/terminology#proof-of-stake-pos should be: https://docs.prylabs.network/docs/terminology/#validator
• Others links on the page are also not working

On https://docs.prylabs.network/docs/how-prysm-works/beacon-node/:

• "proof-of-stake" is: https://docs.prylabs.network/docs/glossaries/terminology#proof-of-stake-pos should be: https://docs.prylabs.network/docs/terminology/#proof-of-stake-pos
• "proposal" is: https://docs.prylabs.network/docs/glossaries/terminology#propose should be: https://docs.prylabs.network/docs/terminology/#proposal-propose-a-idproposea
• "attested", "validators" and others are broken too.

Other pages are probably broken too.

even though bazel does fetch most of the dependencies, on a fresh ubuntu 20.04 machine, I had to manually install the following packages before I was able to run bazel build

apt install libgmp-dev libtinfo5

Yesterday, we ran a survey specifically targeting the Prysm community and contributors to let us know how they feel about the project. We received some high quality submissions and had a section where users could tell us how they feel about the docs. Here are the paraphrased, actionable responses:

• Instructions on how to run your own eth1 node and point prysm to it
• An explainer regarding how key management works: i.e. where are the keys stored, what format are they in. If I want to take out my withdrawal private key and put it in cold storage, how would I do that? IDEA: perhaps we can have a dedicated sidebar section to validator accounts / keys with various pages?
• We need more clarity on hardware. Not just tech specs but form factor. Are server comps going to be best because they are built to run 24/7 or will a NUC have just as much uptime/longevity? Perhaps some best practices for operating and maintaining hardware while running nodes (How to transfer your validator(s) to a new SSD if you have a hardware issue or just need to upgrade for storage)
• Info on how to do a voluntary exit
• Grafana info is important, we need the links to be fixed

These items can be split into separate tracking issues if needed to resolve. cc: @har00ga

This is a tracking issue to add information regarding visualizing a local node's metrics via a Grafana dashboard. We should create a page called Visualizing Node Metrics that suggests users to use the following dashboard: https://www.dropbox.com/s/9mqsvm1s0jmzu98/Prysm%20Dashboard-1587566218430.json?dl=0. Nodes run prometheus metrics by default on localhost:8080/metrics, so we'd need instructions on how to setup grafana at a basic level, import that dashboard, and hook it up to the node's prometheus endpoint.

Been requested multiple times over the past few mos but somehow keeps falling through the filed issue cracks: docs need a guide explaining how to safely migrate validator keys from one machine to another and how to properly store backups

i.e., for bazel:

bazel build --config=release //beacon-chain

Changes in site topology or titles can result in broken links throughout the docs. Include a automatic test for broken links using: https://github.com/stevenvachon/broken-link-checker

In order to get incoming connections from peers, port 13000 should be open from router/firewall to host.

If running on docker, present instructions say

docker run -it -v $HOME/prysm/beacon:/data -p 4000:4000
gcr.io/prysmaticlabs/prysm/beacon-chain:latest
--datadir=/data

Please add -p 13000:13000 to enable docker host to forward 13000 to container appropriately.

A user in our discord pointed out it would be helpful for users to know they can run their docker processes in the background using the -d flag on docker for detached mode. It should NOT be the recommended default but rather something we can briefly mention if they prefer in a yellow tip box.

For context: prysmaticlabs/prysm#6382

We should provide documentation around how to set up Prysm for log rotation using something like logrotate(8). Users may find that their logs are taking a considerable amount of disk space.

Whatever is directly below the highest level header on each page is what gets parsed by google, discord and just about everywhere else. Somehow there's become a build up of pages that feature notices and images instead of contextual information here, this needs to be changed by simply switching the order and moving any notices or images below the conceptual introduction paragraph on each page.

Hi @har00ga this is an issue to add some system requirements to installing Prysm in our docs portal. Currently, we tell people to visit our medium post https://medium.com/prysmatic-labs/introducing-topaz-testnet-8e8a4e00a700 but would be best to have the reqs at the top of each install page

I wanted to find any documentation on exiting a validator to share with someone on Discord.

Went to https://docs.prylabs.network/docs/getting-started/

Entered "exit" into search in the upper right corner and got the following:

Clicked on outlined link in screenshot, and got a Page Not Found error.

--datadir is missing and can be useful in case someone wants to save the files on an external HD

https://docs.prylabs.network/docs/prysm-usage/parameters/#slasher-parameters

As of prysmaticlabs/prysm#4893, we now provide debug images for the beacon-chain and validator docker images.

Many advanced users have asked for basic tooling that comes with the Busybox base image. Specifying docker images with a -debug suffix will now include that tooling.

In recent conversations with users, they are confused on some terminology in ETH2. I'd like to see some more documentation around the data structures in ETH2, how they work in the life cycle of a validator, and what it all means. Ideally we could have great diagrams to explain these objects. The Prylabs team has many talk slides with visuals of some of these objects and would be happy to share.

Starting points:

• Beacon Block
• Attestations
• Aggregation (this was the most confusing)
• Proposer slashing
• Attester slashing
• Voluntary exits
• Withdrawals

The new links to the grafana dashboards do not seem to work.

24be1f8

As we approach the launch of mainnet, the risks involved with running a validator node are starting to become real. A practical guide with actionable risk-prioritised best practices and recommendations would be an excellent resource that may save a number of people from having a very bad day by losing their hard earned ETH.

This guide should cover much more then just "how to not get hacked". Issues as simple as losing funds due to crashed hard drives, forgotten passwords, or accidentally deleted keys are arguably more common then getting pwned.

Please add your ideas for topics below. This guide will likely be an on-going effort requiring input from multiple experts before it reaches a level of maturity with good coverage of how to manage risks spanning confidentiality integrity and availability.

Add documentation for new state mgmt and how it's used. Similar to:
https://lighthouse-book.sigmaprime.io/advanced_database.html