Startup Shoreline Inc. is exiting stealth mode today with the launch of a new incident automation platform that it says will help ease the burden on site reliability engineers and improve application and service availability.

Shoreline founder and Chief Executive Anurag Gupta is a former database and analytics services engineer at Amazon Web Services Inc., and said he had a front-row seat to the issues faced by SREs as they strive to ensure high availability of production systems at scale.

The problem is primarily one of volume. As applications based on microservices scale and fleets grow, operators are forced to spend many hours staring at monitor screens, grinding their way through ticket queues without ever reaching the end.

Gupta told SiliconANGLE in an interview that he likens production operations to the old Chinese torture of a “death by a thousand cuts.”

“There’s no silver bullet since no single issue dominates,” he said. “At AWS, each week, my teams would look at the prior week’s tickets and look for what we could extinguish through automation. It could be something as simple as bouncing a stuck process or as sophisticated as predicting a disk failure. Most issues are commonplace and have well-understood repairs.”

The key to eliminating this burden is automating those repairs, Gupta said, and that’s what Shoreline’s new platform is all about. SREs can quickly and easily create bots that can automatically take action when an alarm is raised, he said, to restore or improve system availability faster.

Shoreline’s platform gives engineers a way to build automations quickly that can respond to frequently occurring issues and repair them without manual intervention. Engineers gain fleetwide visibility into their applications’ and systems’ resources and metrics, and can take action on any issues that are flagged using a Linux command or a shell script, all from the same interface.

Gupta explained that what Shoreline does is different from other incident response platforms such as PagerDuty and StackPulse, which he said are really just platforms for messaging incident automation, providing basic “if this, then that integration between alarms and runbooks” that still leave a human in the loop. In other words, it’s still up to a human to implement the fix.

“That’s insufficient,” he said. “Shoreline provides real-time, fleet-wide interactive debugging and repair with a fluent integration of resources, metrics, and Linux commands – the nouns, adjectives, and verbs of operations. After you fix an issue the first time, you can rapidly automate using the debug session to create alarms, actions and bots.”

Though it’s exiting stealth today, Shoreline said its platform is already being used by engineers to solve all manner of issues, ranging from rotating expiring certificates to collecting diagnostics and purging old Docker images.

Redmonk Principal Analyst Stephen O’Grady said he sees a lot of potential in Shoreline’s platform. Most enterprises understand that manual remediation processes simply don’t scale, leading to unacceptable downtime and burned-out developers and engineering staff. So they’re looking for a solution.

“The only remedy is to introduce automation to diagnose and repair incidents without requiring manual intervention,” O’Grady said. “This is what Shoreline does.”

Image: Shoreline