Protocol timeout during Target.setAutoAttach

[BR0kEN-]

  | 2021-11-17T23:42:01.681+04:00 | /crawler/node_modules/lighthouse/lighthouse-core/gather/driver.js:341
  | 2021-11-17T23:42:01.681+04:00 | asyncTimeout = setTimeout(reject, timeout, new LHError(LHError.errors.PROTOCOL_TIMEOUT, {
  | 2021-11-17T23:42:01.681+04:00 | ^
  | 2021-11-17T23:42:01.681+04:00 | LighthouseError [LHError]: PROTOCOL_TIMEOUT
  | 2021-11-17T23:42:01.681+04:00 | at /crawler/node_modules/lighthouse/lighthouse-core/gather/driver.js:341:50
  | 2021-11-17T23:42:01.681+04:00 | at new Promise (<anonymous>)
  | 2021-11-17T23:42:01.682+04:00 | at Driver.sendCommandToSession (/crawler/node_modules/lighthouse/lighthouse-core/gather/driver.js:339:28)
  | 2021-11-17T23:42:01.682+04:00 | at Driver.sendCommand (/crawler/node_modules/lighthouse/lighthouse-core/gather/driver.js:362:17)
  | 2021-11-17T23:42:01.682+04:00 | at EventEmitter.TargetManager._onFrameNavigated (/crawler/node_modules/lighthouse/lighthouse-core/gather/driver/target-manager.js:38:27)
  | 2021-11-17T23:42:01.682+04:00 | at EventEmitter.emit (node:events:402:35)
  | 2021-11-17T23:42:01.682+04:00 | at EventEmitter.emit (node:domain:475:12)
  | 2021-11-17T23:42:01.682+04:00 | at Driver._handleProtocolEvent (/crawler/node_modules/lighthouse/lighthouse-core/gather/driver.js:285:24)
  | 2021-11-17T23:42:01.682+04:00 | at EventEmitter.emit (node:events:402:35)
  | 2021-11-17T23:42:01.682+04:00 | at EventEmitter.emit (node:domain:475:12) {
  | 2021-11-17T23:42:01.682+04:00 | code: 'PROTOCOL_TIMEOUT',
  | 2021-11-17T23:42:01.682+04:00 | friendlyMessage: {
  | 2021-11-17T23:42:01.682+04:00 | i18nId: 'lighthouse-core/lib/lh-error.js \| protocolTimeout',
  | 2021-11-17T23:42:01.682+04:00 | values: {
  | 2021-11-17T23:42:01.682+04:00 | errorCode: 'PROTOCOL_TIMEOUT',
  | 2021-11-17T23:42:01.682+04:00 | protocolMethod: 'Target.setAutoAttach'
  | 2021-11-17T23:42:01.682+04:00 | },
  | 2021-11-17T23:42:01.682+04:00 | formattedDefault: 'Waiting for DevTools protocol response has exceeded the allotted time. (Method: Target.setAutoAttach)'
  | 2021-11-17T23:42:01.682+04:00 | },
  | 2021-11-17T23:42:01.682+04:00 | lhrRuntimeError: true,
  | 2021-11-17T23:42:01.682+04:00 | protocolMethod: 'Target.setAutoAttach'
  | 2021-11-17T23:42:01.682+04:00 | }

The error happens randomly and without any steps that I can repeat to get it again. The approximate occurrence is 1 per 100 runs.

The Lighthouse is run in the AWS Fargate container (no parallel processes/threads). The custom code has the try-catch block however from the stack trace it seems the error eludes because thrown within the event listener, thus async.

The problem here is not about the errors but about the inability to catch them and fail the Node process gracefully. Currently, I have to run Lighthouse in a subprocess to catch errors.

Lighthouse version: 9.0.0

[ali-habibzadeh]

Same issue here