How to read your log and check for errors
The log is where Autoview tells you what it did with your command. A command came in, it was parsed, an order went to the exchange, the exchange answered. Every step lands in the log. When a signal fires and nothing seems to happen, the log is the first place you look, and usually the only place you need to.
Where you find it depends on which Autoview you run. The technique for using it is the same on both.
The two surfaces, and where each keeps its log
Autoview comes in two forms, and each has its own log.
- The webhook platform runs on our servers. It accepts a POST from any sender: TradingView, TrendSpider, a Python script, your own cron job. It executes around the clock, with no browser open. Its log lives in your account site, alongside your accounts and webhooks.
- The Chrome extension runs in your browser. It is wired directly to TradingView alerts, and it only executes while your computer and Chrome are on. Its log lives in the extension popup.
Most people should be on the webhook platform. It keeps running when your laptop is closed, and the log it writes is the server's own record of every webhook it received. The extension log is the right one to read only if your signals still flow through the browser extension. See how the two compare on the platforms page.
Reading the extension log
Click the Autoview icon in your Chrome toolbar to open the popup. In the left sidebar, select Log. That is the running feed of everything the extension has handled.
The feed carries four kinds of message, and you can show or hide each one. By default all four are on. Click a type to toggle it off, click again to bring it back.
- Log is the raw record of commands as they arrive.
- Info is the detail behind a command, including what a dry run would have done.
- Warning flags something worth your attention that did not stop the trade.
- Error is the loud one. Something failed, and this line tells you what.
Hiding the noisy types is the point of the filters. Turn off Log, Info and Warning, leave Error showing, and a long quiet feed collapses to just the lines that broke. That is the fastest way to find a single failure in a busy session.
Export your log
Sometimes you need the log out of the extension: to attach to a support ticket, to diff against your alert history, or just to read a long session somewhere more comfortable than a popup. The extension has an export built in.
Open the extension's options page and select Log in the sidebar. Next to the log controls is an Export Log dropdown with a window for how far back to reach: Last minute, Last hour, Last day, Last week, Last month, or Everything. Pick one and the extension writes every log entry in that window to a CSV file and hands it to Chrome as a download.
Two behaviors worth knowing before you wonder where your file went. If the window you picked has no activity in it, no file is created; an empty export simply doesn't happen. And Everything means everything, so on a busy setup that file can run to megabytes. Where the file lands is Chrome's call: if Chrome is set to ask where to save each download you'll get a save dialog, otherwise check your downloads folder.
When you're writing to support about a log entry, an export of the relevant window is the next best thing to a screenshot, and often better. It carries the full context around the line that confused you.
Reading the webhook platform log
Log in to your account and open the log there. It is the server's record, so it shows what arrived and what Autoview did with it whether or not any browser was ever open. A webhook that fired at 3am while you slept is in this log.
Read it the same way you read the extension feed. Find the entry for the signal you sent, follow it from the command Autoview parsed through to the exchange's response. If the exchange rejected the order, the rejection text in the log comes straight from the exchange, not from us. That distinction matters when you are working out whose rule you tripped.
Debug with d=1 before you risk real money
Here is the technique that the filters were built for. Add d=1 to any command and Autoview marks it disabled: it parses the command, writes out what it would have done, and sends nothing to the exchange. No order is placed. It is a dry run.
The result lands in the log. A disabled command shows up under the Log filter, and the detail of what it would have executed shows up under Info. So the move is: send your command with d=1, then read the Info line to confirm Autoview understood you the way you meant. Get a clean dry run before you ever send the live one.
This is also your best tool when a live command behaves oddly. Put d=1 back on it and resend. The dry-run output shows you what Autoview thinks you asked for, side by side with what you expected. The gap between those two is almost always the bug.
One note on dynamic values. A TradingView alert can write a live number into a command with a token like p={{close}}, where TradingView substitutes the candle's close at fire time. The {{close}} token is TradingView's, not Autoview's. Any sender supplies its own value its own way, and Autoview just receives the number. When you dry-run a command that used one of these, the log shows the value that actually arrived, which is how you catch a sender that sent the wrong thing.
When the log doesn't make sense
Some lines are self-explanatory. Order placed, order cancelled, balance read. Others are an exchange's own error text relayed verbatim, and those can be terse. A rejection for a quantity below the pair's minimum reads differently on every exchange, because every exchange wrote its own.
If a line stumps you, capture it and ask. A screenshot of the log entry, with the surrounding lines, is the single most useful thing you can hand support. It carries the command, the parsed result, and the exchange's reply in one frame. Reach the team at support@autoview.com.
Read the log first, dry-run with d=1 second, then send live. That order keeps mistakes cheap.