How to handle error while running lighthouse for multiple URLs

I am running lighthouse on multiple URLs, out of that couple of URLs fails due to some error like timeout error or 500 status code, 403 …etc So lighthouse exits with code 0 and uploading fails.

Is there any way we can upload partial success reports to LHCI server?

Below is the error we get while running lighthouse

18:00:00  Running Lighthouse 1 time(s) on https://somedomain.com/1
18:01:30  Run #1...done.
18:01:43  Running Lighthouse 1 time(s) on https://somedomain.com/2
18:02:26  Run #1...done.
18:02:49  Running Lighthouse 1 time(s) on https://somedomain.com/3
18:03:28  Run #1...failed!
18:03:28  Error: Lighthouse failed with exit code 1
18:03:28      at ChildProcess.<anonymous> (/node_modules/@lhci/cli/src/collect/lighthouse-runner.js:103:21)
18:03:28      at ChildProcess.emit (events.js:314:20)
18:03:28      at Process.ChildProcess._handle.onexit (internal/child_process.js:276:12)
18:03:28  {
18:03:28    "userAgent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.75 Safari/537.36",
18:03:28    "environment": {
18:03:28      "networkUserAgent": "",
18:03:28      "hostUserAgent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.75 Safari/537.36",
18:03:28      "benchmarkIndex": 16
18:03:28    },
18:03:28    "lighthouseVersion": "5.5.0",
18:03:28    "fetchTime": "2021-01-12T12:33:12.727Z",
18:03:28    "requestedUrl": "https://somedomain.com/3",
18:03:28    "finalUrl": "https://somedomain.com/3",
18:03:28    "runWarnings": [
18:03:28      "Lighthouse was unable to reliably load the page you requested. Make sure you are testing the correct URL and that the server is properly responding to all requests. (Status code: 500)"
18:03:28    ],
18:03:28    "runtimeError": {
18:03:28      "code": "ERRORED_DOCUMENT_REQUEST",
18:03:28      "message": "Lighthouse was unable to reliably load the page you requested. Make sure you are testing the correct URL and that the server is properly responding to all requests. (Status code: 500)"
18:03:28    },

[patrickhulce]

You can’t achieve the specific behavior you want just from the configuration file. autorun is meant for the majority of common use cases but when you’d like any custom logic like what you’re talking about here, that’s what the individual commands for each step are there for 😃

[patrickhulce]

We have many proposed hooks along the lines of what I think you’re asking for, but none I can foresee that would handle arbitrary catching of fatal errors such as this one in a way that allows LHCI to continue as normal.

If you have a proposal for a specific hook into LHCI you’d like to see and how it would work, please feel free to add details in #438 😃

(FWIW, this situation alone wouldn’t be enough to justify it. As I currently understand it, what you want to do is already easily accomplished with ~3 lines of a bash script whereas the compelling examples in #438 are currently impossible)