Floor Drees

May 11, 2025 • 4 minutes

Building and maintaining an open source project takes a village. In a mini-series on this blog we would like to highlight the work of our maintainers, component owners, and members of the larger community.

Today’s superstar: Jeff Mealo. Jeff is a Senior Software Engineer at Gisual, where a CloudNativePG-powered outage-analytics platform turns hours of triage into minutes of insight. Any business that can’t afford downtime due to power outages, fiber cuts, or other faults can pinpoint the issue in minutes and dispatch the right technicians only when needed. Gisual doesn’t just use CloudNativePG to keep the lights on for themselves and their customers; they actively contribute engineering to ensure it shines brightly for the entire community.

It’s no wonder Jeff ended up in Tech, he has always had an insatiable curiosity to figure out how things work:

“We got our first home computer at 8, and the internet at 9. The internet was one of those things, and I found myself creating video game fan sites using HTML and Perl from ages 9-10. One game wasn’t enough and this exploded in scope to helping create the largest database of video game cheat codes using PHP, and my ambitions quickly exceeded what free hosting could achieve at the time. So I did what any 11 year old would do, start a computer salvage operation on eBay using old gear that folks in my community had donated to me to get a dedicated server and then sell enough web hosting to keep the lights on!”

He tells me that because he didn’t have any real life mentors, only strangers helping him on IRC, he was unaware that this was a career path. While working two jobs to try to pay for college, a classmate observed Jeff working the night shift at a convenience store and asked why he wasn’t doing anything with computers. Thanks to that guy we now get to work with Jeff!

Jeff first came across CloudNativePG when he was looking to achieve equivalent or better performance, availability, reliability than a managed service. CloudNativePG was the only operator that ticked all the boxes and it had a thriving open source community. “With CloudNativePG now under the CNCF umbrella, we feel like we made the right choice!”

While vetting CloudNativePG for production usage, Jeff did a lot of chaos engineering and really put it through its paces before deciding to migrate away from the safety of managed Postgres. “It met and exceeded our expectations, but we continued to have issues with volume snapshots and a couple of other edge cases that were very difficult to isolate and reproduce.” Even though prior to this, Jeff hadn’t done any Go development, with the help of some gophers on the Go slack, he was able to find the root cause of the issues with the operator: connection handling in the instance manager. If 3 backups failed in a row, it would no longer report its status properly which resulted in numerous failures (luckily, none of which caused an outage in themselves, just reduced redundancy).

Jeff has a wishlist for the project:

Support for other topologies (namely distributed Citus)
Out of box experience: a wizard that helps you configure things to get the best performance/availability/reliability/recovery time – there’s so much to consider and it’s rare to find an individual who possesses enough knowledge on all the moving parts to get things right the first time!
Polishing alerts and dashboards and gathering first hand experience and guidelines for production usage.

Anyone interested working on the above topics, find Jeff in the CloudNativePG channels on the CNCF Slack workspace, or on Bluesky! Jeff’s also active in the #postgresql channel on Libera, and while he’s still getting used to chatting on Slack and Discord for Open Source, he is a member of the gophers.slack.com, victoriametrics.slack.com, postgresteam.slack.com, and cloud-native.slack.com.