Build Error on one Client but not another

So I keep having my builds fail on the firefox test it seems and the only reason I find it weird is that if I transfer all the files to my laptop and do a build things go fine.

As far as I can tell everything is up to date with my npm, and firefox so Im kinda left scratching my head. Its kinda felt like I have never been able to get the test to validate 100% on my PC.

Here is the log:

LOG: 'Running Outline'
LOG: '%c validated', 'color: green'
LOG: 'Running Visibility'
LOG: '%c validated', 'color: green'
Firefox 70.0.0 (Windows 10.0.0): Executed 128 of 128 (5 FAILED) (13 mins 43.216 secs / 13 mins 41.451 secs)
[10:10:13] 'tests-validation-virtualscreen' errored after 14 min
[10:10:13] Error: 1
    at formatError (C:\Users\pryme\AppData\Roaming\npm\node_modules\gulp-cli\lib\versioned\^4.0.0\format-error.js:21:10)

    at Gulp.<anonymous> (C:\Users\pryme\AppData\Roaming\npm\node_modules\gulp-cli\lib\versioned\^4.0.0\log\events.js:33:15)
    at emitOne (events.js:121:20)
    at Gulp.emit (events.js:211:7)
    at Object.error (F:\gitHUB\Babylon.js\node_modules\undertaker\lib\helpers\createExtensions.js:61:10)
    at handler (F:\gitHUB\Babylon.js\node_modules\now-and-later\lib\mapSeries.js:47:14)
    at f (F:\gitHUB\Babylon.js\node_modules\once\once.js:25:25)
    at f (F:\gitHUB\Babylon.js\node_modules\once\once.js:25:25)
    at tryCatch (F:\gitHUB\Babylon.js\node_modules\async-done\index.js:24:15)
    at done (F:\gitHUB\Babylon.js\node_modules\async-done\index.js:40:12)
    at removeAllListeners (F:\gitHUB\Babylon.js\node_modules\karma\lib\server.js:370:9)
    at webServer.close (F:\gitHUB\Babylon.js\node_modules\karma\lib\server.js:381:13)
    at Server.close (net.js:1608:9)
    at Object.onceWrapper (events.js:313:30)
    at emitNone (events.js:111:20)
    at Server.emit (events.js:208:7)
[10:10:13] 'tests-all' errored after 18 min
[10:10:13] 'default' errored after 37 min
npm ERR! errno 1
npm ERR! babylonjs@0.0.0 build: `gulp --max-old-space-size=8192 --tsLintFix`
npm ERR! Exit status 1

might be a driver issue and rendering being different, as long as the server tests are consistent it is all good.

what would you recommend doing? Is there anything I can do really other then like a fresh windows install and fresh drivers and stuff?

Is the diff pretty big on the pictures ?

I’m not sure what you mean?

you can see the errors reported in the tests http://localhost:1338/tests/validation

when you run locally

then you can see the difference for each tests so either it is just a bit off -> driver diff or it is a lot different which could highlight a bug as well.

Ill look now

So all the tests validate when I do it with the local running… but when I run a build they fail. I’m assuming it might be a console error then?

Ouch even weirder!!!

hmmm, but once it gets to that point on the build it should be fully compiled and ready to go right?

So I could just run the tests in browser afterwards to validate on my own. Maybe when its booting up from the command line something funky is happening?

Yup it is really strange as the tests run the same page and collects the failures only :frowning:

maybe my client is not allocating enough ram or something dumb? Ive been really confused by this behavior for a while. :frowning:

Is there a timeout on the tests I can crank up?

My command line is stupid slow perhaps running it from a slail command line is causing the issue?

I had it working for a second then it went back to crashing every build, while the same tests there were not validating check out instantly when I manually run the tests.

Also it takes a large chunk of time per test on the build execution of browser, while the manual one while the local is running cranks each test out lickity split.