DevLogBus exists because local debugging can turn into a stupid little scavenger hunt.
One service is logging to stdout. Another is using structured logs. A CLI command prints something useful once and vanishes. The browser console has the other half of the story. Linux has the systemd journal. Somewhere in that mess is the cause and effect you need, but your eyes are bouncing between five terminals and a browser as if that is a reasonable way to live.
DevLogBus gives that work one local stream.
DevLogBus is a real-time log viewer for full-stack development and active troubleshooting. It runs a local daemon, accepts records from tools and applications, keeps a bounded in-memory replay buffer, and exposes browser and terminal viewers that can show sources side by side or merged into one timeline.
It can collect records from:
slogdevlogbus CLIjournald through devlogbus-journal-bridgeDevLogBus is for developers who need to see local cause and effect across more than one process, runtime, or machine boundary.
It fits:
DevLogBus is not trying to be a hosted observability platform.
It does not provide:
Use a real production observability stack when that is the job. DevLogBus is for the local work before that, where you want signal now and do not want to wire a whole cathedral just to see why a button click made a worker complain.
DevLogBus is intentionally local, boring, and direct.
That last point is not accidental. DevLogBus gives you the tools and the choice. If you choose the fast path, you own the tradeoff.
In short, piss on the electric fence if you want. Just don’t act surprised when physics files a bug report on your ass.