Read the errors your PHP site is already logging.

Every PHP installation writes errors somewhere, whether or not anyone configured it to. Fatals, warnings, notices and deprecations land in the same flat file in the order they happened, each with a timestamp, a severity, a file and a line at the front, and a stack trace underneath it where there is one.

The trouble is the shape of the thing. A busy site writes tens of thousands of lines a week — the same handful of problems repeated and interleaved, in a file you have to SSH in to read. So nobody reads it, and a fatal that fires for one customer in fifty goes unnoticed until that customer telephones.

It reads the same file and turns it into errors instead of lines. Repeats collapse into a single entry with a count against it. Each entry keeps its severity, its file and line, its message and its full stack trace, so the cause sits one click from the symptom rather than a grep away.

Filters run across all of it: by severity, by file, by line, or by a tag you apply yourself. The feed updates as the file grows, so an error that fires while you are watching appears while you are watching.

Each error carries a fix status. Closing one keeps a record of who closed it and when, which over a few months becomes an honest answer to how long your team takes to turn a reported error into a resolved one.

Alerts go out by email or into a Slack channel for the severities you choose, so the critical ones do not sit waiting for somebody to open the dashboard.

The alternative is a package inside your application, which means a dependency, a deploy and one more thing that can itself fail at three in the morning. The log file is already there and already being written. Reading it needs read access and nothing else.

It also means the tool works on the application nobody dares deploy. The ten-year-old PHP site whose original developer is unreachable is exactly the one whose errors nobody can see, and exactly the one you cannot safely add a package to.

The free plan covers one site: the live feed, the filters, the stack traces and the fix status, with no card and no expiry. What it limits is report downloads.

Team access, longer log history and anything specific to the way you work are built to order. Tell us what you need and we will quote it. There is no price list, because those requests have never twice been the same.

We maintain other people’s PHP applications, and every one of those handovers started the same way: sitting on an SSH session reading a log file to find out what the site had been doing wrong for years. We built this to stop doing that by hand, and it is now the first thing we connect when we take a maintenance engagement on.