Blog Post

Prmagazine > News > News > Cloud veterans launch ConfigHub to fix ‘configuration hell’ | TechCrunch
Cloud veterans launch ConfigHub to fix ‘configuration hell’ | TechCrunch

Cloud veterans launch ConfigHub to fix ‘configuration hell’ | TechCrunch

Three cloud industry leaders have set up a new company with a mission to modernize software configuration data management.

Starting from invisibility today, there are $4 million in funds. Confighub It is the CEO’s handicraft Alexis Richardson,,,,, Founder of Cloud Local Container management platform weaving; CTO Brian GrantFormer Google Software Engineer and original principal architect of Kubernetes; and CPO Jesper Joergensenhold various product roles at Salesforce (including Heroku) Lead its voice, video and platform teams before joining Twilio.

Richardson told TechCrunch that their new company is setting out to “get people out of the config hell.”

If that’s just blurting out you, “Huh?” Let’s explain. Last July, CrowdStrike was released Awry configuration update to its Falcon sensor security software, resulting Widespread destruction and More than $5 billion Loss of Fortune 500 companies. Delta Air Lines says Losed $500 million Due to flight interruption, What to start A long legal battle to make up for its losses.

This event proves that software is now a critical infrastructure and is combined with a network of interdependent components connected to the API to create powerful systems and applications. However, if any part of the network is intentionally compromised, or intentionally or otherwise compromised, it can lower not only the entire house but the entire town.

This gives us the configuration data, which ensures that all these different parts work well with each other and that the software and the system understand how they should behave. But these configuration files can be a clumsy mess, and Confighub is working on solving this problem.

“The problem is that configuration data is spread everywhere – it has become a whole spill,” Richardson said.

Fix Devops

The way software is deployed and managed has changed a lot over the past 30 years. In Yore’s days, enterprise software was mostly deployed via CD-ROMs on local hardware, and configuration was limited to some text tiles that instruct the operating system or application to find what is needed to run correctly.

Today, it’s not that simple. Gargantuan configuration files with thousands of lines of code are required to make the software run in a large-scale environment that is usually dynamic.

Joanson said he learned a valuable lesson at Heroku (Salesforce-have Platform as a service): The code and configuration data are not the same and require very different approaches. Configuration data defines system settings and behaviors that cannot be discussed like software code. However, understanding this data (whether using Terraform or Kubernetes) is critical to avoid expensive unqualified conditions that can cause downtime or delay.

“We’ve made great strides in the ever-evolving collaborative coding practices with tools like Github,” said Joergensen. “But configuration of real-time infrastructure requires a different approach. It’s not just a bunch of files. Our goal is to bring the elegant application developer experience of Heroku pioneers to a variety of on-site production operations.”

Confighub therefore promises “unified configuration management through modern automated development workflows and compliance.” Instead of looking for the right configuration corresponding to a given error, on confighub, everything will be saved in a single database, making the configuration easy to find and with a real-time view of what the system is actually doing.

“So, if your customers don’t have access to the system, you should be able to see and experience what they see and experience,” Richardson added. “The configuration data in our system will be an expression of a real-time production system that not only recognizes the data, but also changes the data so that they can solve the customer’s problems immediately.”

The product will be available through the SaaS model, with initial focus on Kubernetes DevOps tools such as rudder,,,,, Argo,,,,, Flux,,,,, Terraform Its open source fork Opentofu. Richardson said that although he did not name it, Confighhub has worked with some “middle” corporate design partners.

The product is a few months away from its official launch, so today’s announcement is more about Confighub introducing the world and announcing its $4 million investment. Funding comes from some well-known VCs and Angel Investors, including Crane Venture Partners, Coding Ventures, Pear VCs and Poolside CEO Jason Warner.

AI Factor

While the CrowdStrike episode is one of the most expensive configuration clangers lately, it’s not an isolated incident.

“Configuration changes have long been called the main cause of cloud system failure,” Grant told TechCrunch.

For example, last year, the Australian Pension Fund Unisuper was attacked by shutdown After Google, this leaves approximately 500,000 members unable to access their accounts Special shape construction The Google Cloud VMware engine (GCVE) that caused Unisuper to be deleted.

And in January Configuration error On Github’s cloud, all GIT operations cannot be used for up to two hours.

The problem will worsen – AI will be more complicated as companies compete to embed AI into their software structures.

“AI completely overturns the way the world writes, manufactures and operates software, not just code writing,” Warner said in a statement. “The next few years are not just the atmosphere of small changes to the foundation of Greenland’s code.

Of course, companies can build their own solutions from a mixture of tools like Terraform, Kubernetes, Prometheus, and more. However, these setups and management can be too complex.

There are also incumbents with name, such as ServiceNow and Atlassian, which provide services to solve configuration problems. However, such companies were founded 20 years ago, before cloud computing actually took off. In addition, many of these companies have long entered the fields of human resources, CRM, team collaboration and project management.

“These tools are not suitable for modern dynamic, cloud, AI-powered containerized applications,” Richardson added. “If a simple misconfiguration means that AI will see their company data and publish it on the internet, you can’t anyone try to roll out AI.”

The third is cloud

Prior to Confighub, Richardson founded an enterprise-focused cloud messaging company rabbitThis is acquired by VMware subsidiary Springsource 2010. Then in 2014 He founded the knitting workit continues to raise more than $60 million from investors such as ACCEL, Amazon Web Services and Google Ventures. At Weaveworks, Richardson also developed Gitops Framework, now widely used in cloud-native and Kubernetes environments, is used to manage infrastructure and application deployment.

Formal knitting Stores closed in February last yearRichardson Write then. So in mid-2024, he started Confghub with Grant and Joergensen.

Grant and Richardson As a job project After Google donated it in 2015. Around 2020, Grant began giving up his various leadership roles on the Kubernetes project, and over time he developed an idea for a new business that might not match Google’s priorities. this meeting However, in the end, the weaving work last year was over.

The jump from one of the largest technology companies on the planet to startups seems to be a huge transition. But Grant worked at two startups before Google – for one, he was the lead architect of a small startup called Peakstream, Google obtained in 2007.

“I was the third engineer there – when we started, we didn’t even have a name, we didn’t have an office, we didn’t have a business plan,” Grant said. “So I’ve done a very early startup before.”

Although Google continues to grow exponentially in the company’s 17 years, Grant’s involvement with Kubernetes, the company’s participation Google opened in 2014which is also a bit similar to working in a startup company.

“Kubernetes started out as four engineers and a product manager; we didn’t even have a co-manager, and we all reported to different people,” Grant continued. “It’s a startup in itself.”

Meanwhile, Richardson met with Joergensen when he acquired RabbitMQ at VMware 10 years ago. After some of the world’s largest tech companies have held various senior positions, Joergensen has also chosen to enter the world of startups.

“You can do more than ever now – we’re jumping into the cracks of AI, and I probably don’t want to be anywhere else right now,” Joergensen said. “In a startup, we can see clearly what we can do.”

Source link

Leave a comment

Your email address will not be published. Required fields are marked *

star360feedback