Plans differ by features,
not by volume

Every error your applications report is an event. Many error trackers limit how many you can send per month. We don't.

The same occurrence reported twice - for example, logged through ILogger and also sent by hand - arrives once and isn't counted as a repeat.

The first occurrence arrives at once, in full. Repeats of the same error are counted, and the count is written into that message: 10 minutes later, then after 30 minutes, 1 hour, and every 6 hours while it continues. Pushover can't edit messages, so there the count arrives as summaries.

Choose which errors go to which channel - say, Critical ones to Pushover, Staging errors to a separate Discord channel, errors from the Payments module to the payments team's chat. Without rules, every channel of a project gets every error. Rules guide

Two keys per project, like in Azure. Replace a key without losing errors: move your apps to the other key, then regenerate the old one.

Direct calls, ILogger, Serilog, NLog and log4net: the BugsRadar package and a package for each logger on NuGet. .NET guide

Uncaught errors, direct calls, Express, winston and pino: the bugsradar package on npm, for Node.js 18 and later. Node.js guide

The logging module, loguru, uncaught exceptions, Django, Flask and FastAPI: the bugsradar package on PyPI, for Python 3.9 and later. Python guide

One HTTP request from curl or PowerShell: SQL Server Agent jobs, backups, CI/CD pipelines, cron, Task Scheduler and systemd. Scripts guide

Serilog.Sinks.Seq and other Seq clients send CLEF to BugsRadar with no BugsRadar package: the errors in it go to your channels. CLEF guide

Who did what and when: projects created or deleted, keys regenerated, channels turned off, people invited or removed. It records actions in your account, never your errors.