Skip to article
XTester Help
XTester Launcher

Automatic Restart

Learn how Launcher reports failures and restores robot operation.

Reviewed: Applies to: 1.0.0-preview.1

The crash notification

If the robot ends unexpectedly, the launcher shows a notification and tells Telegram, if Telegram is chosen in the launch profile. The Why? button opens the journal on the line with the reason.

Application screenshotXTester 1.0.0-preview.1
After the test process was terminated, Launcher recorded exit code 137 and began the first of three restart attempts. Select Why? to open the Journal.Interface screenshot · XTester 1.0.0-preview.1

A crash does not cancel the robot's orders or close its position: they stay on the exchange, and the next start continues with them.

Automatic restart

After a crash the launcher starts the robot again at once, in the same mode, Paper or LIVE. There are at most 3 restarts in 10 minutes; the journal shows which attempt it is.

• If the robot crashes again and again, after 3 restarts in 10 minutes the launcher stops restarting it.

• If the start needs a person (entering keys, confirming an exchange check, a test order), there is no restart: the robot stays stopped and the reason is in the journal.

• If the robot could not start at all, there is no restart: it would repeat the same error.

In all these cases the orders and the position are left untouched.

The robot did not answer or the connection was lost

If the robot did not answer and was stopped by force, or the connection to it was lost, the status bar shows Error with the reason. For a LIVE robot the journal opens by itself; check the position and orders on the exchange before the next start.

What to Do After a Notification

Choose Why? and read the journal. Distinguish a crash during execution, preparation refusal and lost connection. Repeating cannot fix the same installation failure or supply a required human decision.

Do not create a parallel instance solely because contact was lost. For LIVE, establish the existing account/order/position state through normal recovery. Process failure is not a close-position command.

After the retry limit, resolve the cause and deliberately start with the intended profile. Retain attempt number, time and original error for support; restarting alone does not prove the cause was fixed.