Infura product tips and Web3 tutorials for developers building decentralized applications.

Infura Developer Tutorials

Tutorials

Getting Started With Infura

Comparing NFT Standards - ERC-721 vs ERC-721a vs ERC-1155

Wanna be in the same league as Damien Hirst, DC, and Niftys? Then build your NFT on the Palm network!

Infura and Palm NFT Studio are excited to announce that the Palm network is now available to all users of Infura. Nearly 430,000 developers now have access to create NFTs on the Palm network.

The Palm NFT Studio

The Palm network is an Ethereum sidechain dedicated to selling, purchasing, and trading NFTs that are associated with real-world and digital assets. Palm provides low gas costs and fast transaction finality. The network is over 99% more energy efficient than Proof of Work-based blockchain networks, providing artists with a more sustainable solution today, while Ethereum upgrades to Proof of Stake in the Merge, tentatively scheduled for release in July 2022.

The Palm NFT Studio has collaborated with Damien Hirst, DC, Niftys and Warner Brothers and other artists in minting NFTs on the Palm network.

How to mint an NFT on Palm

For users looking to mint NFT projects, the Palm network provides an effective solution to create NFTs that are able to be bridged back to Ethereum.

To access the Palm mainnet or testnet endpoints, here’s how to get started:

1. Login to your Infura account and select an existing Ethereum project (or create a new Ethereum project if you have not yet).

2. Click on the settings of the project to view the Palm API endpoints.

Using Palm mainet and testnet is free as part of all Infura plans, and requests sent to the API count as part of your existing plan. No action is required to activate Palm, and the availability of Palm testnet allows you to have an environment to begin developing and testing your NFT project before going live.

See our documentation on Palm for more information, as well as links to helpful tutorials.

Get in touch and start building

Interested in creating your own NFTs, but do not know where to start?

Learn more from Infura’s Director of Engineering, Kris Shinn, on minting NFTs using Infura and Replit, based on his EthDenver 2022 presentation.

Let us know how you are using NFTs in our Discord and feel free to ask questions in our community.

Just getting started with Infura? Our docs will get you building.

Ethereum Network Updates, Upgrades, Hard Forks

Infura - Ethereum Network Updates

Network

The Palm network is now available for all NFT creators on Infura

Last July, we published Ethereum JavaScript Libraries: web3.js vs. ethers.js (Part I) to provide “a comparison of web3.js and ethers.js, focusing on their similarities and differences, so that you can better understand the nuances of the libraries.”

To recap, we dove into the documentation of both libraries. We then used a modified version of a software evaluation framework Andres Canal established in his article “Using Quill,js to Build a WYSIWYG Editor for your Website” to guide the reader toward making the best decision in choosing which library to use.

The article was “written with the intent of educating developers to understand the tradeoffs between the two libraries to be able to make a decision of which library to use for their particular use case.”

How have the conclusions from that article changed over the course of the past year?

Table 1: Comparison of Conclusions about Web3.js and Ethers.js




July 2020
October 2021




Web3.js has a community of users and maintainers associated with the Ethereum Foundation. Web3.js has a good API reference and has been around since 2015. Many projects are using it and the library has been the go-to in many "Build Your First dApp" tutorials.
Web3.js is best for those who prefer the most-used Ethereum javascript library that is also maintained by a blockchain-focused firm. Web3.js's strengths are its tests and community. Documentation is currently modest in design and scope and can definitely be improved.


Ethers.js is excellent because it is a small, compact library with a large number of test cases. It has good “Getting Started” documentation, making it accessible to new users. Developers find ethers.js simple and intuitive and the library has gained in popularity over the past two years, seeing increased downloads and usage by projects.
Ethers.js is best for those who prefer lightweight, user-friendly libraries with robust testing and documentation. Ethers.js’s strengths are its documentation, tests and community. It could have more maintainers in order to create some redundancy in the development team - currently a single individual.







Let’s dive into the updated data and explain why these conclusions have changed.


1) How important is the popularity of the library?

To use Andres Canal’s definition,

“A popular project on Github means exactly what you expect it to. Namely, that there are lots of people questioning, contributing, and supporting it. This activity tends to result in more reliable software. You can check the popularity of any project by looking at how many stars it has, how many issues are opened, how many pull requests are pending, and how many contributors worked on the project.”

Web3.js has about 11,600 stars and ethers.js 3,100 as of Oct. 14, 2021. Web3.js is used by 112,849 repositories, compared with 65,519 for ethers.js.

Looking back to July 2020, Web3.js had nearly 8,800 stars while ethers.js had nearly 1,500. Web3.js was used by 51,300 repositories on GitHub, compared with 18,500 for ethers.js.

Web3.js remains the more popular library, but impressively for both libraries, their dependencies have at least doubled over this time period. Web3.js remains more popular, but the overall pie of dependent projects has grown from about 69,800 to 178,368. We will dive more into the dependencies in question 3.


2) How important is the maintenance of the library?

To quote ourselves from the last article,

“We want to use libraries that are updated frequently, so that bugs are squashed and new features are added. Though it is not perfect, looking at a month (or two) of commits, closed issues, active issues, and the number of maintainers over that time period can give a sense of the strengths and weaknesses of an open source project. This also reveals patterns of maintenance, giving some insight to the user of a library when updates and bug fixes are generally addressed, which can be used as a proxy metric for maintenance of the library.”

Using the same metrics from our last article, we see the following for both libraries.

Table 2: GitHub Monthly Pulse for Sept. 14, 2021 to Oct. 14, 2021




Library
Commits to Main Branch
Closed Issues
Active Issues
Maintainers




web3.js
20
60
107
10


ethers.js
2
17
39
1






GitHub Pulse now reveals some additional useful information about the libraries, which gives us even more context of why we are seeing what we are seeing.

For web3.js, we can see that a single individual is responsible for the majority of the commits, though there are three individuals that have 10 or more commits. This is something I see as a plus. This means there is redundancy built into the development team, in case someone were to stop contributing to the project. Web3.js has 121 commits to branches other than the main branch, as web3.js is transitioning off of the 1.x version as outlined in this blogpost the roadmap for web3.js (which we will reference again in the next section). This indicates a lot of work is going into the future releases.

For ethers.js, Richard Moore continues to be the sole maintainer of the library, but there are many more people contributing to the project by opening and closing issues on the GitHub repository, indicating that there is a community of individuals committed to the health of the library. By virtue of a single maintainer, it is not surprising to see fewer overall commits.


3) Who is responsible for the development of the library and how many projects use it?

ChainSafe is the lead maintainer of the web3.js library, having received a grant from the Ethereum Foundation to “continue the maintenance and further development” of web3.js in September 2020. ChainSafe’s outline of the future plans for web3.js reveals major changes coming, such as the deprecation of web3-bzz and web3-shh packages in version 3.x and an eventual move to Typescript in version 4.x. These are exciting additions, but the tradeoff is that ChainSafe will need time to realize these while maintaining the existing library.

Ethers is the lead maintainer of the ethers.js library, which is Richard Moore’s project, so we can think of Richard Moore = Ethers. Ethers has been the lead maintainer of the project since the inception of it. This can be a risk for some projects, but in this case, it has made ethers.js very quick in adding features and capabilities. For example, ethers.js is fully TypeScript-ready with “definition files and full TypeScript source” according to the documentation.


Table 3: Dependent Repository and Package Counts, October 15, 2021




Library
Dependent Repositories
Dependent Packages




web3.js
113,018
4,685


ethers.js
66,950
3,505







Web3.js, likely due to the longer existence of the project, has more repositories and packages that are dependent upon it. When compared with metrics from our last article, ethers.js is closing the gap with web3.js when it comes to dependent packages.


4) How important is it that the library has tests?

Answer: Very important to any dApp developer worth their salt. Luckily for us, both libraries have gotten even better since our last article.

Ethers.js has a test suite API documented for the current v5 release on their GitHub repository. The documentation of the testing is very impressive. Check it out.

Web3.js has added test documentation since our last article. This documentation, based on my personal inspection, appears robust and clearly written.

I believe ethers.js has the lead, but web3.js has done a great job catching up in terms of its tests and their documentation, and I feel both have great coverage when it comes to tests. Tests are a strength for both libraries now.


5) How important is the number of downloads of the library?

As of Oct. 15, 2021, web3.js has 30,767,910 downloads and ethers.js has 24,284,490 downloads. Compare that with May 20, 2020, when we first looked at this statistic, when web3.js had 14,703,432 downloads and ethers.js had 8,500,336 downloads. The overall pie has grown, but the difference in downloads is surprisingly close: a difference in total downloads of 6,484,420 on Oct. 15, 2021 compared with a difference of 6,203,096 in late May of 2020.

When we look at weekly downloads, the two libraries are practically even: web3.js at 330,102 and ethers.js at 339,318. Weekly downloads of both libraries have nearly doubled since we last checked this statistic (May 20, 2020, for those following along): web3.js at 175,661 and ethers.js at 184,798.


6) How important is web performance?

Web performance is indirectly measured by comparing the size of the libraries, as we are assuming smaller libraries take less time to load in the browser.

Both libraries are small. Web3.js has decreased the size of the library to 7.28 MB unpacked (unpacked refers to the size of the library when published) from 10.6 MB when we last checked. Ethers.js touts its very small size, which has increased to 322 kb uncompressed from 284 kb uncompressed over the same time period. It’s unclear what occupies 322 kb within ethers.js. NPM shows ethers.js 10.2 MB unpacked, and I confirmed this when I downloaded to my personal machine and calculated the size of the ethers.js package using the Finder application on Macs. So we have to take ethers.js's word that it is small in size, which is a big plus, though I do need to do further research to confirm this claim is valid.


7) How important is the quality of the documentation?

Ethers.js is the clear winner when it comes to the quality of the documentation of the library and it stands out for three key reasons.

 1. Ethers.js has an extensive “Getting Started Guide,” which helps a user orient themselves to both the library and the vocabulary of Ethereum.
 2. Ethers.js has useful call outs and additional information provided as part of the documentation, giving the end user visual cues, convenient links, and other helpful user experience features that make reading and using the documentation an enjoyable experience.
 3. Ethers.js documentation has an extremely useful feature - a toggle to single page view, in which all documentation is provided on a single page. This makes using “Ctrl + f” in the documentation a breeze, and helps with finding the method you are looking for quickly.

Web3.js has not had many visible changes to the documentation since our last article and could use an update to make the user experience more pleasant.

An indirect additional comment on the documentation - ethers.org provides a very helpful “playground,” where you can try out the library in browser: https://playground.ethers.org/. This useful feature lets a user get started playing around with the library quickly and easily. While it could be argued that remix.ethereum.org provides a similar playground experience for web3.js, this is not the goal of remix, and remix support ethers.js as well.


8) How important is the overall usage of the library?

The last time we looked at the numbers, web3.js was the clear winner in overall usage, with ethers.js gaining users at a rate that appeared that it would only be time before ethers.js caught up to web3.js. The results today show that both have healthy ecosystems and a large number of users. My personal belief is that the overall growth of the ecosystem shows that both libraries have been adopted by a significant number of projects and users. So, while web3.js has more downloads, stars, and overall metrics, ethers.js is competitive to web3.js in its usage. It’s fair to say both have been widely adopted.


9) How important is the license?

This is one area that has not changed since we last reported. Web3.js still has the LGPLv3 license and ethers.js the MIT license. I am not a lawyer, and so this is the hardest area for me to pick a winner. It is worth reviewing Slava Todavchich’s article “Understanding open-source and free software licensing,” which we suggested in our July 2020 blogpost.

But wait... there is a new category!


10) How important is the community and community experience?

Yes, I am adding a new way to evaluate the two libraries, based on my experience in Developer Relations and Software Community. I am personally looking for robust channels where developers using the software can interact with one another, find information about updates and hotfixes, share their use cases and help each other debug their code. The best software projects provide this for their users and it is a differentiating factor.

Web3.js and ethers.js have both embraced Discord as a means to communicate directly with their community of developers. Web3.js lives within the ChainSafe Discord Server, in the channel #web3js-general. Web3.js has used this channel not just to interact with developers, but  as of September 2021, to promote a web3.js developer survey to help drive the direction of changes for version 4.x of web3.js. Ethers.js utilizes the EF JavaScript Team Discord Server, in the #ethers channel. Both are active channels with communication on a daily basis.

Both also use Twitter to communicate changes to their audiences, though both have relatively light follower counts. @web3_js (1,250 followers) and @ethersproject (1,279 followers) only tweet to announce updates and changes to the libraries, serving as another way for notifications about updates to get through to developers.

ChainSafe also uses their Medium account to post some updates about web3.js, though not a regular cadence.  

GitHub continues to be a source of discussion for communities of both libraries, though the shift to Discord by both reflects a growing trend in the Ethereum community, as well as in software communities in general, to commit to this tool for a mix of synchronous and asynchronous communication.

The verdict is that both take development and interaction with their communities seriously.

What’s the next step?

You should go ahead and test out the libraries themselves with Infura, and we created a tutorial for just this purpose: Ethereum JavaScript Libraries: web3.js vs ethers.js (Part II). The best way to see if my evaluation is valid is to kick the tires on both libraries.

In conclusion, instead of declaring a winner, it seems that Ethereum itself is benefitting from having two leading JavaScript front-end libraries. More libraries being built that can utilize the features of web3.js and ethers.js will push them to be even better (like web3-react, for example). In updating this article, I have been struck by the growth of downloads and dependencies of both libraries, making me of the opinion that you can choose either and feel comfortable with the decision. I hope this illuminated some ways you can evaluate both to choose what works best for you.

Get started building on Infura for free today.

Ethereum JavaScript Libraries: Web3.js vs Ethers.js: An Update

As the clock ticks down to the Merge, the Teku and Infura teams have continued refining their collaboration. In case you missed it, these two teams have created an eye-poppingly fast Ethereum 2.0 client sync experience. Since then, they have continued refining their integration and made some important discoveries and improvements.


Attestations with Teku

One of those discoveries was that validators linked to a remote beacon node could have significant performance issues: specifically, a lot of missed attestations. Obviously, this is a pretty crucial issue. It turned out to be not just a problem with a remote beacon node, but specifically a remote beacon node behind load balancers, including Infura’s load balancers.

If you’re interested in the blow-by-blow, check out the Github issue here.

The solution the team came to was to implement an option to disable early attestation publishing when the block is received:

--validators-early-attestations-enabled[=<FALSE>]

Find the full Teku documentation on this issue here.

This option allowed validators connecting to Infura beacon nodes to have performance equivalent to validators connected to a local beacon node, which in and of itself is impressive. It would appear that this is a fix not just for nodes connecting through Infura, but for any context where Teku validators are using load-balanced beacon nodes.

The second innovation the team has made surrounding attestations is batch attestations. This consists of sending all attestations produced for a given slot to the beacon node in a single batch. In the context of Infura, this allows running a lot more validators without hitting rate limits.


The future of history

Another feature that the team has been working on, which isn’t quite ready yet for a full release, is an innovative way to store historical state without having a hard drive the size of the Library of Alexandria.

With increasing adoption of decentralized technology, there is more and more demand, and even novel use cases, for historical chain data. The most top-of-mind for many in the space is tax compliance. Institutions and even individuals are increasingly in need of solutions for being able to query, quickly and cheaply, historical chain data.

Like most great things in this space, the solution being developed uses Merkle tries. Without getting too deep into the woods (pun intended), as it’s not finished yet, the technique essentially takes the difference between slots and stores the branch updates. The applicability for this is tremendous, so watch this space for more updates.


Safety through Diversity

As we approach the reality of Ethereum 2.0, more and more voices are calling out a topic of vital importance: client diversity. The Ethereum Proof of Work consensus mechanism is, at this point, venerable and battle-tested, at least in blockchain terms, and is resistant to 51% of attacks. As the network transitions to Proof of Stake, the threat remains that a single Ethereum client will be running on a majority of validator nodes. This opens the door not only for a malicious attack, but also for single points of failure bringing down the entire network.

For more explanation, check out this very readable primer that should bring a lot of issues into focus. There are so many reasons to make the move to a Proof-of-Stake consensus model, and the network is on track to do so. It would be a stunning failure of vision and quite simply, wildly unsafe to move Ethereum to a PoS network that has a majority of any Ethereum client.

Even if we were to assume that all clients were going to act in the best interests of the network as a whole, which in and of itself is a leap most aren’t willing to take, the Ethereum network is a sprawling global entity with external dependencies. No client should be able to be a single point of failure for the entire network.

As we build the Beacon Chain and validators up and out, client diversity must be included as an essential consideration.

Learn more about Infura’s Eth2 API here and sign up for a free account.

Teku + Infura to 2.0: How to Handle Attestations

SoK: Validating Bridges as a Scaling Solution for Blockchains

So, why would you need a hybrid infrastructure model alongside Infura?

Let’s think about how your database is set up. You probably have your own database, which is great. What happens if that database goes down? Would your app continue to run seamlessly, delivering the experience you want your customers to get? Or, would your database shutdown cause outages and crashes in your app?

If the dataset you’re hosting is a public dataset, you have a new paradigm for making your infrastructure much more resilient and reliable by integrating Infura into your overall strategy. Infura implements a reverse proxy pattern internally and can be a highly reliable and performant node in your infrastructure stack, allowing you the ability to have a portion (or all) of your traffic routed to Infura.

You can choose to keep your database as your primary database with Infura as a backup, or vice versa. In this way, Infura will act as an always-on backup for your database, providing a level of flexibility previously not available with a single database running.

You also have the opportunity to use Infura as a stats comparison tool to make sure your performance is up to par. Alternatively, you may need to handle more load (request traffic) than your infrastructure can currently handle. You can route the overflow to an Infura project ID to handle spikes in traffic so your service stays up and running.

There are many opportunities for you to use Infura alongside your own in-house node infrastructure to improve your user experience. Let’s dive a little deeper on how we accomplish this.


What does it mean when Infura implements a reverse proxy pattern?

Let’s review what a normal proxy is first, and then we’ll flip it and reverse it, a la Missy Elliott, to talk about the reverse proxy. A proxy server (or web proxy or forward proxy) is a server located in front of a group of client machines (computers). When those computers make requests, the proxy server intercepts those requests and communicates with the target web servers on behalf of those clients. Basically, a proxy is a fancy word for a request/response middleman.

If you’re a visual learner, here’s a diagram from Cloudflare that shows just where that forward proxy sits:


What is a reverse proxy?

A reverse proxy is a server that typically sits on the edge of your infrastructure. It is responsible for receiving requests from web clients and subsequently routing them to one or more backend web servers.

Here’s another great diagram from Cloudflare to show how the reverse proxy fits in to the whole flow:




One of the main reasons to use a reverse proxy server is load balancing, ensuring the flow of traffic between clients and servers remains smooth. This is one of the main advantages of using both your own node and Infura. In addition, it allows you to cache content and provides web acceleration for improved performance. Lastly, reverse proxies allow for more efficient, secure SSL encryption and protection from DDoS attacks and related security issues.

As we previously mentioned, Infura can be used as a reverse proxy to assist with ensuring the flow of traffic for your clients and servers stays as smooth as possible, making for a more enjoyable user experience.


Some examples of a reverse proxy in action

Let’s first look at how we can set up a very simple reverse proxy using Caddy.

We will use our localhost and a backend HTTP service running on 127.0.0.1:9000 to make sure we’re getting our reverse proxy working correctly.

First, download Caddy. Once that’s downloaded, check out their Getting Started guide here. Or, follow along to learn how to set up a simple reverse proxy!


At this point, you may be running your own node, but since it takes too long to sync one for this article, we can substitute with a simple NodeJS stub to check that our reverse proxy is working correctly. I simply named mine main.js.


const http = require('http');

const hostname = '127.0.0.1';
const port = 9000;

const server = http.createServer((req, res) => {
   res.statusCode = 200;
   res.setHeader('Content-Type', 'text/plain');
   res.end('Reverse proxy success!\n');
});

server.listen(port, hostname, () => {
   console.log(`Server running at http://${hostname}:${port}/`);
});



This sets up a backend HTTP service running on 127.0.0.1:9000 mentioned earlier and will display a “Reverse proxy success!” message if our proxy is working correctly. To start this, run `node main.js` in a different terminal window and leave that connection open!

Next, we will want to create a Caddyfile (literally create a new text file in whatever directory you’re working in and name it Caddyfile). Once you have that set up, we will set up our simple reverse proxy itself inside the Caddyfile with just two lines:

localhost 

reverse_proxy 127.0.0.1:9000



In this instance, reverse_proxy is where we want to get our information from (the original site), and localhost is where we are redirecting to using our reverse proxy.

Now, from the same directory as your Caddyfile in a new terminal window from your `node main.js`, run:  

caddy run



This will start the caddy server and will allow your reverse proxy to work! Now, you’ll notice when you run this that you don’t see the “Reverse proxy success!” message popping up.

That’s because this just started the server. You now need to make a request to “https://localhost” and you’ll see the message. To do this, open another terminal window (that makes three now, for those of us keeping track), and simply run:

curl https://localhost



Running this command will give you the “Reverse proxy success!” message we were looking for, and now you know you’ve successfully set up a simple reverse proxy.


Enter Infura!

Now that we’ve gotten a basic reverse proxy up and running, let’s try one using an Infura endpoint. We will keep our overall setup of using a Caddyfile the same, but we won’t need the backend HTTP service running in the background, as that will be switched to an Infura endpoint.

We’re going to change up how our Caddyfile looks:

localhost 

reverse_proxy https://rinkeby.infura.io {
   header_up Host {/v3/<YOUR_PROJECT_ID>}
}



You will notice that we’ve added in a “header_up Host” inside the reverse_proxy section. This allows us to include our Project ID to both the Rinkeby endpoint and the localhost endpoint we are proxying to.

To test this out, we will need to send a request, so I used web3_clientVersion as an example:

curl https://localhost/v3/<YOUR_PROJECT_ID> \
-X POST \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","method":"web3_clientVersion","params": [],"id":1}'



When we run this, we get the following output:

{"jsonrpc":"2.0","id":1,"result":"Geth/v1.10.8-omnibus-aef5bfb3/linux-amd64/go1.16.7"}



Cool! But how do I know my reverse proxy is working? Well, if you switch out `localhost` in our curl command to `rinkeby.infura.io`, you get the same output, which shows that our reverse proxy is running properly.


How do you use Infura as a backup for your own node?

We’re going to set things up initially the same as we did in the previous example but with two upstreams, one to our localhost and the other to the Infura node.

Next, we’re going to set up load balancing. Now, what’s load balancing? Load balancing acts like a traffic cop sitting in front of your application servers, routing client requests across all the servers that can fulfill those requests quickly and efficiently.

So how do we set that up?

Since we’re using Caddy, we can refer to their docs for the basics of how to use more than one upstream. When we look at lb_policy, there are many ways to determine which upstream Caddy will use first. Caddy will default to using the first listed upstream and will move down the list if the servers stop responding. In our example, we want to have localhost:9000 from our main.js file be the first upstream, meaning we want our reverse proxy to send the data through our localhost first, so we will identify that by putting it first in our list of upstreams (we will leave our Infura Rinkeby endpoint in as the second upstream):

localhost

reverse_proxy localhost:9000 https://rinkeby.infura.io {
   header_up Host {/v3/66172ed429624788ac264604755ba42c}
}



Since we are using the localhost:9000 port, we want to get that server back up and running. In another terminal window, start the Node server with node main.js.

To make sure everything is set up correctly, let’s run our request through by running the same request from earlier but with the :9000 port included, expecting to get the “Reverse proxy success!” message:

curl http://localhost:3000/v3/<YOUR_PROJECT_ID> \
-X POST \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","method":"web3_clientVersion","params": [],"id":1}'



and we get the following output:

Reverse proxy success!



Excellent! Now, how do we incorporate the Infura endpoint to be a fallback?

You’ll notice we left the Infura endpoint in our Caddyfile where it originally stood and we asked our proxy to load balance using the first upstream (our localhost), which was how we got that “Reverse proxy success!” message.

Now that we confirmed that we set everything up correctly on our end, we’re going to kill the localhost server, mimicking what would happen if your private node went down. Once we do that, our requests should automatically start routing to the Infura node and we should get something similar to the below output:

{"jsonrpc":"2.0","id":1,"result":"Geth/v1.10.8-omnibus-aef5bfb3/linux-amd64/go1.16.7"}* Closing connection 0



But wait! We don’t get that! Instead, we get an error message.

This is because Caddy only allows us to use either two HTTP or two HTTPS endpoints - in our case, localhost is HTTP and our Infura endpoint is HTTPS.

Luckily, there is a tool we can use to make this work with localhost that will create a TLS certificate and sign it with a CA trusted locally by your device and browser. That tool is called mkcert, and it’s incredibly easy to install by following these steps.

Keep in mind if you’re doing this with your own private HTTPS node, you likely will not need this additional step.

If you are following our tutorial step-by-step (using localhost), be sure to follow all the steps, including updating your main.js file to the below to include the HTTPS requirement and the options constant.

const https = require('https');
const fs = require('fs')
const options = {
   key: fs.readFileSync('<PATH_TO_CERTIFICATE-KEY_FILE>.pem'),
   cert: fs.readFileSync('<PATH_TO_CERTIFICATE_FILE>.pem')
};

const hostname = '127.0.0.1';
const port = 9000;

const server = https.createServer(options, function(req, res) {
   res.statusCode = 200;
   res.setHeader('Content-Type', 'text/plain');
   setTimeout(() => {res.end('Reverse proxy success!\n');}, 1000)
});

server.listen(port, hostname, () => {
   console.log(`Server running at https://${hostname}:${port}/`);
});



When you run the mkcert localhost command from the tutorial, the certificate key file and certificate file used in the code above will be created.

Now that we have our localhost running with HTTPS, we can continue! Since we are running everything with HTTPS, we need to update our Caddyfile to reflect those changes.

localhost:3000 

reverse_proxy https://localhost:9000 https://rinkeby.infura.io {
   header_up Host {/v3/66172ed429624788ac264604755ba42c}
}



Now that we have all our files in order, let’s make sure our Node.js server is running and send that request again.

curl http://localhost:3000/v3/<YOUR_PROJECT_ID> \
-X POST \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","method":"web3_clientVersion","params": [],"id":1}'



What do we get?

Reverse proxy success!



Perfect! That’s exactly what we want to get, as we have asked our reverse proxy to go to https://localhost:9000 first.

Now, what happens if we kill the Node server and send our request again? Remember last time we did this, we got an error message. This time we should be getting the result from our Infura Rinkeby node:

{"jsonrpc":"2.0","id":1,"result":"Geth/v1.10.8-omnibus-aef5bfb3/linux-amd64/go1.16.7"}



Success!

Our reverse proxy is, in essence, ignoring the localhost node, as it’s not functioning, and is defaulting to the Infura Rinkeby node!

Super easy, right? Now you know how to use Infura as a backup node to keep your services up and running smoothly!


Where can I learn more about reverse proxies?

Take a look at our client story from MakerDAO. They currently use Infura for a multitude of reasons, including using us as a backup node solution to ensure all developers can use the MakerDAO toolkit!

Another great example is the Polygon PoS network. We initially began working with Polygon to share some traffic load for their network, allowing users to access the Polygon networks via Infura endpoints and effectively ensuring their network remains fast and functional, regardless of traffic spikes they may see. We have since increased our partnership so we are now handling 100% of their network traffic.

Check out this blog post highlighting multiple companies across all verticals using a hybrid infrastructure model of their own nodes plus Infura. Get some ideas on how we can help you accomplish your needs, from verifying transactions to providing backup database services!

If you would like to run an Infura node in addition to your own, sign up for a free trial today.

How to Run Your Own Node AND an Infura Node

When enterprise developers build dapps on Ethereum, they often have to consider the entire tech stack and development process. The list of considerations range from application features such as Web2 interfaces to deployment, which is usually on a permissioned network hosted by a preferred cloud provider. Since most full-stack software engineers build off of SDKs and API calls, it’s important to establish the value of the application with consideration of the deployment network.



Get the best of both Web2 and Web3 worlds

Many enterprise blockchain business cases build on the principles of traditional Web2 development. Web2 focuses on operating with a fluid user experience with attention to detail on scalable and responsive code on multiple devices. A valuable decentralized application (dapp) in Web3 redesigns the economics and industry roles while building on complementary existing systems -- instead of fully replacing those systems with analogous Web3 features.

Say you would like to reinvent a syndicated loans issuance platform such as IPREO. This effort would require significant IT resources to spin up a minimum viable product from the ground up if you’re recreating features like tranche issuance calculations or calendar date exceptions. Instead, consider a path to lesser resistance: blockchain modules that connect these digital assets to new liquidity networks. The functionality is a re-issuance of an existing financial instrument to ABC network with a listing on XYZ automatic market maker rather than the full end-to-end dashboard.

To connect to pockets of liquidity, an enterprise developer must deploy with the same smart contract language that is compatible with the Ethereum Virtual Machine (EVM) and look to bridge the assets to networks with the desired investor and issuer markets.


Why build on public Ethereum?

Until now, publishing enterprise dapps on Ethereum requires significant system integration, executive buy-in and high set-up costs. Recently, we have seen more public blockchains working with enterprises to cover non-functional technical needs in order to meet minimum requirements of security audits and performance. However, the relatively tribal nature of the blockchain developer subculture means that mainstream enterprises are still conservative when it comes to building on Ethereum.

If you are an enterprise developer, consider the following advantages of building on public Ethereum vs. permissioned networks:

1. Faster setup and lower total cost of ownership

2. Immutability and longevity of the data stored with thousands of independent nodes

3. Composability and low barrier of entry to decentralized finance (DeFi)




1. Faster set-up and lower total cost of ownership

This is a standard private blockchain journey for a consortium:

After you develop your dapp, a lot of time is focused on node deployment adhering to enterprise software standards. On the business side, you are also trying to build a consortium to agree on the benefits and adoption before the tech is tested.

The consortium model usually chooses a separate tech vendor that does not have single ownership from one consortium member to prevent conflicts of interest. The tech vendor determining deployment and long-term support likely runs the Ethereum nodes on behalf of participants. This will get the product to market faster and focus on the business benefits.

The last step involves time and licenses for adding more nodes and mutualizing infrastructure costs across participants. Compared with private networks that remove tokens from their operation, public networks require ownership of native tokens to fund transactions. This gas cost pays for the infrastructure -- third parties running full nodes and running operations for mining/validating. In my time building enterprise projects, we have never seen this happen.

The public journey is much more straightforward. After you develop your dapp, the deployment is a choice between public networks with updated configurations on your Truffle setup. The added complexity is the gas costs for funding and running your transactions on this network.

To simplify connectivity to EVM-compatible networks, you can use Infura’s Ethereum API to connect your app instantly with a single line of code. Simplify your transaction management with Infura Transactions (ITX) using prepaid accounts without holding ETH on your balance sheets. ITX is a transaction relayer that manages your transactions to prevent dropped or stuck transactions.

You will also likely want to work with a team such as Diligence that can audit your contracts and monitor them when they’re live.

Public Ethereum has an overall lower total cost of ownership when you compare the operational cost of running a system on mainnet rather than focusing on the ownership and maintenance of the network itself.

A consortium may have more control in the private chain and face less security and treasury reviews at first. However, your infrastructure overhead and negotiations on consortium network matters tend to outweigh the convenience and added features of connecting to mainnet.


2. Immutability and Longevity

The Ethereum mainnet is secured by the interaction of thousands of independent nodes run by individuals and miners throughout the world. Private chains typically have a small number of nodes controlled by one or a few organizations. Those nodes can be tightly controlled, but only a few need to be compromised in order to rewrite the chain or commit fraudulent transactions.

At ConsenSys, we’ve recommended public networks to enterprise developers for their resilience of replicated data as you are able to recover more quickly with a larger network of more than 15,000 independent nodes. The more peers you have, the easier it is to find a seed and fully recover your recorded transactions. In consensus decisions, the more blockchain validators there are to verify transactions, the more difficult it is to coordinate a majority decision that may lead to an attack.

In a permissioned network, the security is governed by the node with the least security. Your network is only as secure as the weakest link. Any added nodes to the network may also increase the surface area of attack. Most institutions will recommend a base software security test in order to join as a full node owner. The node owner will have to pass standard penetration testing to prevent any hacking on the node in case it is a bad actor to the entire network.


3. Composability and Low Barrier to Entry on DeFi


The Ethereum network powers a vibrant ecosystem of third party applications and technology including DeFi and NFTs. A highly composable and interoperable system allows you to connect to DeFi to build financial products that enable and combine multiple apps to match specific user needs. The Ethereum mainnet offers enterprise developers the ability to gain access to market liquidity and provide innovative ways for users to participate in the exchange, trading or growth of that liquidity.

When we say composable, it’s much more than just an open banking API. It’s a set of open source live protocol contracts that can be called and extended. We’ve seen pure forks of open source code like with Uniswap to get sushiswap. The market has seen countless copies of DeFi models such as Compound and Aave, and it doesn’t look like things are slowing down anytime soon.

There’s more and more real value happening in DeFi leading us to believe that cryptocurrencies are not a fad. It is here to stay.

Watch Clemens’ presentation “Pitching Public Ethereum to Enterprises” at EdCon 2021.

Interested in building on Ethereum? Contact us to get support on starting your enterprise project.

3 things enterprise developers should consider when deploying on Ethereum

Are you excited by blockchain but don’t have a great deal of experience coding? It can be easy to fall down a rabbit hole of technical tutorials that make assumptions regarding your initial development knowledge. A particularly important topic to grasp is basic security within your code.

As you begin to build out your very own dApp, you will have to include information such as an Infura API key, along with a mnemonic.

If you are new to coding, this is the equivalent of giving a stranger your social security number and keys to your home. Once this information is public, there is no definitive way to ensure it hasn’t been captured by a person or web crawler, which could result in malicious behavior.

So, how can we address this?

Thankfully, we can use dotenv along with git to protect our sensitive information within our project when / if we push to github.

Dotenv is a zero-dependency module that loads environment variables from a .env file into process.env, storing configuration in the environment separate from the code. It is often a convention for environment variables to be stored in a .env file or other mechanisms based on the server you are using.

Note: This article will focus on building out a project using Truffle and Infura.

In our project, we will need to run the following command:

npm i dotenv


Once the process in our terminal has been completed, we will want to make sure we have the dependency successfully installed. Head over to your package.json file and check to ensure that dotenv has been added under dependencies.

Next, at the root of our project, let’s create a .env file:

Here is where you will add all of the sensitive information that should not be shared.

We will make sure we can access our .env information in our truffle.config file. At the top of the truffle.config file, add the following:

require("dotenv").config


This will tell your project to load anything that is in the .env into an environment variable, granting you access to it.

Now, instead of having to put our Infura api key or mnemonic in an exposed file, we can import that information from our .env file.

When it comes to gaining access to our project key we are going to use, sign up to Infura and create a project. In your dashboard, you should see something similar to this:

The following code is an example between the usage of .env and exposed information within truffle.config:

const HDWalletProvider = require('@truffle/hdwallet-provider');

const infuraURL = 'https://rinkeby.infura.io/v3/<project id here>'

const mnemonic = "Your mnemonic here"


rinkeby: {

provider: () => new HDWalletProvider (mnemonic, infuraURL),

network_id: 4,

gas: 5500000,

confirmations: 2,

timeoutBlocks: 200,

skipDryRun: true

},


Imagine if you pushed this up to github. Anyone would have the ability to open your truffle.config file could take advantage of your exposed mnemonic and project id from Infura.

Let’s take a look at that same code and implement better security practices.

Notice how we are calling upon the information without having to directly add to our file.

One important detail, as of now even with an .env file: if you push this project to github, it will show the .env file, which would defeat the purpose!

Take a look for yourself.

This is where .gitignore comes into play. We will need to create a .gitignore file, so we can push this project up to github.

Without getting into git, a quick summary could be git is version control software that keeps track of all of your files and the versions of them. What .gitignore does is it tells git not to keep track of this file along with everything it contains.

If you want to dive deeper into git along with other foundational material, have a look at our Basic Training, a free course that helps ensure you have the fundamental tools when developing.

At our root (similar to the level our .env file lives), we will create a .gitignore. Inside of that file, we will add .env along with any other files you would rather not be pushed to github.

We can now push our project up to github and notice that our .env file is no longer exposed. Our sensitive information is now much more secure.

If you are experimenting with this, make sure you add your .env file to your .gitignore before pushing.

If you are running into issues, remember gitignore ignores only untracked files. If your files were committed in the past, they are now tracked by git. To ignore them, you first need to delete said files, git rm them, commit and then ignore.

Otherwise, you can delete the .env file and run

Git rm .env --cached


Here is a helpful link discussing the removal of sensitive data from a repo if it was previously committed and discusses how to clean the git history. Savvy attackers don't just look at the most recent repository info, they'll look at the history of it as well.

Note: If you come across it in documentation, a question you might have is
“What is env. example”

An env. example file, on the contrary to this tutorial, is actually included in the repository, it is used as an example file for you to know what KEY=VALUE pairs you need for your project. Most often, it is used to copy the .env file and then change the values.

Example of execution:
Enter to the project root directory using your terminal.

Check if .env the file exists or not.

If the file does not exist then create a file named .env.

Now execute the command

cp .env.local .env.


So the .env.sample file has all the ENV declarations defined but without their values. It helps other developers establish/set up the project without having to reverse-engineer the codebase.

So after git clone, you usually execute the command `cp .env.sample .env` and fill in the values with the Infura credentials as we saw in the example above.

Note: When you create a new repo using github, if you scroll down you should see the option to include a .gitignore file so you don’t have to create one yourself in the future.

Hopefully, you found this article on basic security practices helpful as you begin to develop your very own projects. Feel free to join our Discord if you want to continue the conversation, share projects, ideas, and join the community!

Start here with a free Infura account >

How to Use .env to Enhance Basic Security Within Your DApp

Optimistic Ethereum is a rollup scaling solution that allows users to submit transactions on the Ethereum network and get them executed faster for a much lower gas cost. From the perspectives of users and developers, this is a "parallel universe" called Layer 2, or L2, where users typically have the same addresses as they do on the Ethereum blockchain (also called Layer 1 or L1). There are gateways that allow messages and assets to be transferred between the universes.

The Optimism network API largely uses the same JSON-RPC specification as Ethereum and Infura supports all the necessary API methods for building a fully functional application on the Optimistic Ethereum network.


Why are transactions cheaper on Optimism?

On L1 Ethereum, thousands of nodes run your transaction to verify that the proposed block that includes it is valid. This process is expensive and has to be paid for. Unlike on L1 Ethereum, Optimistic roll ups provide faster and cheaper transactions because it allows users to perform most of their tasks off-chain in a "layer 2" protocol. On Optimistic Ethereum, the smart contract has only two tasks: deposits/withdrawals and verifying proofs of transactions on chain. This makes the process more efficient and a lot cheaper.


Why is this safe?

The great thing about decentralization is you can trust that transactions are executed correctly because there are thousands of computers executing them. Honest miners would reject such a transaction. With rollups, you don't have that level of redundancy.

The answer, in the case of optimistic rollups, is incentives. Every node in the network has to put up a stake before it can propose a new block with the execution results of a transaction. Every other node can check up on those results, and if it thinks they are incorrect, dispute them. The dispute is resolved on L1, so it is as trustworthy as anything else in Ethereum. Whichever node lies, either proposer or disputer, loses its stake and most of it goes to the other node involved.

Of course, the whole point is to avoid running transactions in L1. This system achieves that by making it pointless to cheat. Any proposer who attempts to cheat is going to be caught by the other nodes, lose its stake, and not get the invalid transaction result accepted as valid. A disputer can force a single transaction to go into L1 arbitration as a denial of service attack, but the stake is high enough that it wouldn't be worth doing.


How does Infura work with Optimism?

Infura exposes endpoints for Optimistic Ethereum (the production network) and Optimistic Kovan (the test network). You use them the same way you use any other network, by selecting them from the menu and using the specified endpoint.

There are a few small differences between building on Layer 1 Ethereum and Optimism:

1. You must use the JSON-RPC API as we do not support Websockets yet.
2. Search filters are not yet supported


How to Get Started on Optimism...



The main difference between Optimism and regular L1 Ethereum (other than the cost) is the access point URL.

| Network          | RPC URL |


… If you are using the Truffle development environment

Infura users can add the free Truffle Optimism add-on directly on their Infura dashboard. The Truffle Optimism box also contains the configuration definitions required to use Infura. Currently, you will need to manually add the gas costs of transactions when you use Truffle to run deployments or tests on Optimistic Kovan or the main Optimistic Ethereum network.


… If you are using Hardhat

To use Optimistic Ethereum or Kovan, edit hardhat.config.js's modules.export.networks to add these definitions:

"optimistic-kovan": {
    url: 'https://optimism-kovan.infura.io/v3/<Infura Project ID>',
    accounts: { mnemonic: <your account mnemonic goes here> },
    gasPrice: 15000000,
    ovm: true // This sets the network as using the ovm and ensure
                      // contract will be compiled against that.
},
"optimistic-ethereum": {
  url: 'https://optimism-mainnet.infura.io/v3/<Infura Project ID>',
  accounts: { mnemonic: <your account mnemonic goes here> },
  gasPrice: 15000000,
  ovm: true // This sets the network as using the ovm and ensure
                    // contract will be compiled against that.
}



Then you simply run:

npx hardhat --network optimistic-<kovan or ethereum> <command>



… If you are using Remix

When you use the Remix Optimism plugin to develop for Optimism, network connections are created through the wallet. If you use MetaMask, you can select the network Custom RPC and fill in the details:

What is Optimistic Ethereum?

Since its founding over a year ago, Infura has had a mandate to support the Ethereum community, to help it grow. We’re pleased to announce that as part of this commitment, we are now running a signer (aka authority) and a bootnode on the Rinkeby test network.

Rinkeby is a proof-of-authority (PoA) network using a protocol called Clique, and it addresses important problems associated with proof-of-work (PoW) consensus on a test network like Ropsten. When PoW consensus is used on a network with a small amount of hashpower, an attacker can easily mine a large percentage of blocks, exclude legitimate transactions from being mined, modify the gas limit, and clog the network with his own spammy transactions. This has all happened on Ropsten, disrupting developers’ ability to build and test their software.

On Rinkeby, blocks are only minted by trusted signers, typically members of the community who are good citizens and can be relied on to help the network run smoothly. You can see the blocks minted by Infura here.

Infura also runs a Rinkeby bootnode. When a new node joins the network and wants a copy of the blockchain, a bootnode helps it to find other peers and sync the chain. The next release of geth will use Infura’s bootnode, although this is typically transparent to geth users.

The Rinkeby testnet has been running well for several months and is a great alternative to Ropsten.

Thanks to Péter Szilágyi and the go-ethereum team for entrusting us with the responsibility to run the signer and the bootnode.



Infura’s signer and bootnode on Rinkeby

The Infura team continues to grow and scale along with Ethereum itself. As a core infrastructure provider to the public Ethereum networks, the demands on our infrastructure over the last several months have been sizable. Fueled in large part by ERC20 token sale activity and ENS registrations, the main Ethereum network is handling about 400,000 transactions per day. These transactions all arrive into the Ethereum network through a node that relays it eventually to a miner who packages it in a block. Infura is one such gateway that these transactions relay through and find their way into the chain.

Beyond these 400,000 “write” transactions per day, the Ethereum network supporting decentralized application handles many more “read” requests in the form of JSON RPC API calls. Just as with traditional web development, the number of read requests are often orders of magnitude larger than the write requests. Infura is routinely servicing over 600 million of these “read” requests per day. We continue to work on back-end infrastructure to ensure these requests are served lightning fast and reliably.


Not all API Calls are Equal

One of the challenges and learnings we found quickly with our public-facing endpoints was that not all API method calls are equal. For example, eth_blockNumber is pretty lightweight and due to its nature, provides an opportunity to cache given it only changes about every 15.. I mean, 21 seconds (see Ethereum Ice Age).

Now let’s consider eth_getLogs and it’s impact on an Ethereum client. The “getLogs” method has become a very common tool for the Dapp developer to query previous blocks and transactions for data needed to determine current actions. While it has shown to be extremely useful to Dapp developers, the getLogs method is extremely heavy on an Ethereum client and its underlying chaindata database structure.

Imagine a call to search for any arbitrary array of data from block zero to the current block (currently ~4.1 million blocks). In this case, a call to eth_getLogs would initiate a sequential scan of the entire database, bringing that particular node to a crawl for several minutes. It isn’t that difficult to initiate a Denial-of-service attack against an Ethereum node using this technique.

Using Ferryman, the Infura service layer, we are able to automagically route specific method calls to fit-for-purpose client backends. The eth_getLogs method is a perfect use case for bespoke method routing with Ferryman.

When an inbound request comes in making a getLogs request, Ferryman isolates these to a pool of dedicated backend clients running in full archive mode and with specific resource capabilities needed to service the requests. This approach has the primary benefit of preventing our entire service from being negatively impacted due to heavy getLogs requests. A side benefit being that it helps us to optimize costs as we can size our resources more for-purpose.

The Infura team is hard at work identifying use cases such as this to optimize and scale in order to deliver the best possible experience to Dapp developers and users. We remain committed to our strategy to meet the infrastructure demands needed to support the ongoing growth of the Ethereum world computer.

Updated statistics 8/18/17.



Engineering

Scaling Infura: Not All API Calls are Equal

Once up a time, Ethereum blocks had plenty of room. Usually they had a few or even zero transactions, and once in a while you’d see a block with over a hundred. If you sent a transaction, it was usually included in the next block.

In the token launch era, however, things have changed. Now we see thousands of transactions flooding the network around the start of a launch. I say “around the start” because some people send their contributions even before the start block arrives — a suboptimal strategy.

The large number of pending transactions is testing the limits of Geth’s and Parity’s mempool settings. The mempool (the term is borrowed from Bitcoin) is the set of data structures inside an Ethereum client to store pending transactions before they are mined (Geth calls it the “transaction pool”; Parity calls it the “transaction queue”).

Typically, a transaction received by an Ethereum client is held in the mempool and also forwarded to other clients in the network. It eventually reaches a miner, who adds it to a new block. When a client later receives a block that includes a transaction already in its mempool, it removes it from its mempool, freeing up space.

In Geth 1.6.5, the basic config for the mempool is in a Go structure called TxPoolConfig. One of the interesting fields in that structure is GlobalSlots ("Maximum number of executable transaction slots for all accounts"). In the default config (DefaultTxPoolConfig), GlobalSlots is set to 4,096. Once that limit is reached, Geth starts freeing up space by evicting some of the transactions in its mempool according to a fairly complex set of rules.

In the current token launch environment, 4,096 slots is tragically low. This may be why token launch contributors sometimes see their transactions disappear after sending them to a client (via the JSON-RPC method eth_sendRawTransaction), even if the client's response (a transaction hash) appears to indicate success. The transactions are simply purged from the mempool while the network is overloaded. They may never reach a miner.

Geth 1.6.2 mercifully added several CLI options to tweak the mempool config: they are the ones with the --txpool prefix. Use --txpool.globalslots value for the global slots setting. However, it's doubtful that many clients are using this setting today. That needs to change, pro bono publico.

With tens of thousands of pending transactions sometimes floating around the network, you can set this setting quite high. Make sure to test with it to make sure your machine finds it agreeable.

Similarly, Parity 1.6.8 has a setting called --tx-queue-size LIMIT. The default is 1,024. Crank it.

At INFURA, we're working on a feature that will help your transactions get mined irrespective of the mempool situation. We'll announce it soon, so keep an eye on our Twitter @infura_io .

Thanks to INFURA intern Eric Tu for helping with research for this post.



When there are too many pending transactions

On October 18th at around 1:20pm UTC, the Ethereum network arrived at consensus for block 2,463,000. For all network participants using the latest client releases of the go-ethereum client, the Parity client and the Ethereum(J) client, the EIP150 1b/1c HardFork went into effect.

With this hard fork, the ongoing attacks that have been thwarting the network for a month appear to have come to an end. At Infura, we fully support the fork and look forward to returning to a stable network where we can develop and build solutions without this particular distraction. While we certainly don’t think this will be the last, for now at least, it appears to have restored order to the Ethereum mainnet.

The markets seem to agree that the fork is a positive influence as the price of ETH jumped as much as 7% on the day.

So, in a post-fork world, as always if you are running your own Ethereum node, you will want to be sure you are participating on the blockchain that you intend to be. If you are running a node locally, and you had NOT upgraded to one of the latest clients by block 2,463,000, you are not on the main Ethereum chain. Even if you do upgrade the client you would need to re-sync the entire chain so that you are certain you followed the right chain at the fork in the road. We experienced this on the testnet for Infura (morden.infura.io), so if you have any questions about what you might be seeing on your individual node please reach out and we can share our experience.

The Ethereum clients you need to be running so that you are on the right side of the fork (assuming you believe the fork is the right side) are:

go-ethereum v1.4.18

Parity 1.3.8

These two teams are unbelievable in their ability to deliver and we applaud their hard work. The go-ethereum team has named this latest release ‘Note 7’ showing that even after a month of sleepless nights they still haven’t lost their sense of humor and seem to be having fun along this wild ride.

So while these teams did all the heavy lifting, Infura stood by on the ready to deliver our service to the community. Lots of eyes were on the fork as it approached and went by without any issues. Our new friends at CodeTract wrote a countdown timer that included Infura as one of their sources of truth. Incoming requests to Infura spiked to an all-time peak over 2,000 request/sec. Almost all of which were JSON-RPC calls to the eth_blockNumber method as everyone anxiously watched the blocks come and go around the fork.

Now, as the dust settles and we prepare for the second EIP158 State Clearing HardFork we look forward to a bit of stability allowing us to focus on our roadmap of building out Infura features within Ferryman including smart routing to specific clients, caching, early-warning monitoring and auto-routing to healthy nodes and rich analytics.





The Ethereum Hard Fork and Infura

On September 19, in the early morning hours in Shanghai, the Ethereum community was about to wake up in excitement for the annual developer conference, Devcon2. Instead, we woke up to find the Ethereum network under attack. More specifically, an attacker had found a specific issue with the go-ethereum client, allowing specific transactions to be sent causing the go-ethereum client to hang and not be able to process blocks.

The go-ethereum team quickly responded and issued a patch within a few hours, and Devcon2 went on with only a minor delay in start time.


Infura: “For the People”

At the Infura project, we have the mission of providing secure, stable, fault tolerant and scalable Ethereum and IPFS nodes for the Ethereum community. Our experience in Shanghai was also highly impacted by the attack and subsequent activity. We woke up to a red alert that the Ethereum network was under attack. Our nodes were not broadcasting transactions and were falling behind. Further complicating things was the restricted internet access we had in China. We initially spent some time walking the streets of Shanghai looking for a reliable Wi-fi connection to assess the situation. As we were walking, Joseph Lubin, founder of ConsenSys, commented with a grin, “It’s hard standing up an entire ecosystem from scratch!” We eventually landed in the lobby of the Astor Place hotel which is a short walk from the Hyatt on the Bund where the conference was taking place.

As it became clear that the issue at the time was only impacting the go-ethereum client, we quickly made the decision to route all incoming traffic in Infura to our Parity nodes. This allowed us to quickly get service restored and any application that relied on us for their connection into the Ethereum blockchain was back online and able to behave predictably.

A month later, we have seen that that hiccup on the morning of Devcon2 was just the first in a series of attacks that has been ongoing and has impacted both go-ethereum and Parity. The following timeline summarizes the attacks over the last month.

The Infura team has been actively monitoring the health of the Ethereum network and contributing data back into the development teams as the attacks progress. We are actively developing an early warning detection within our “Ferryman” service layer that will help us to quickly identify attacks that hit particular clients and eventually automatically route incoming requests to healthy Ethereum nodes. We will be supporting the protocol hard fork, originally proposed by Gavin Wood and outlined in this EIP to put this particular attack vector behind us. It is our firm belief that these attacks make our network stronger.

It is our goal to provide stable and reliable core infrastructure so that those who choose to use our service can focus on developing and deploying decentralized applications that behave predictably rather than worrying about maintaining their own infrastructure.





Recent Posts (page 34)

Infura’s signer and bootnode on Rinkeby

Scaling Infura: Not All API Calls are Equal

When there are too many pending transactions

The Ethereum Hard Fork and Infura

Toward Reliable Ethereum Infrastructure