Since switching to 5.4.1 we've had some (seemingly random) failures on some test-runs relating to largestassets.summary.
The error displayed is the following:
[2017-07-19 08:07:37] ERROR: Error: No data to send to graphite for message:
{
"uuid": "83a5a939-92e0-4a3a-9d69-cc768b712e4a",
"type": "largestassets.summary",
"timestamp": "2017-07-19T08:07:37+00:00",
"source": "assets",
"data": {
"plain": []
},
"group": "url.that.was.tested
[20170719-sitespeedio-error-log.txt](https://github.com/sitespeedio/sitespeed.io/files/1158767/20170719-sitespeedio-error-log.txt)
"
}
at Object.processMessage (/usr/src/app/lib/plugins/graphite/index.js:59:29)
at queue.process (/usr/src/app/lib/support/queueHandler.js:155:57)
From previous event:
at queue.process (/usr/src/app/lib/support/queueHandler.js:155:42)
at drainItem (/usr/src/app/node_modules/concurrent-queue/index.js:92:21)
at Immediate.drain (/usr/src/app/node_modules/concurrent-queue/index.js:73:84)
at runCallback (timers.js:672:20)
at tryOnImmediate (timers.js:645:5)
at processImmediate [as _immediateCallback] (timers.js:617:5)
A full log is attached, too [1]. Just re-running the failed one fixes it in all cases that happened so far (probably because the chance of failure itself is relatively low, we run ~30-50 tests per day).
[1][20170719-sitespeedio-error-log.txt](https://github.com/sitespeedio/sitespeed.io/files/1158768/20170719-sitespeedio-error-log.txt)
Thanks @julrich for reporting. @tobli or @beenanner do you see whats wrong? I can have a look later if you don't have time!
Possible related to the code waiting for the first of a type before classifying remaining assets? The initial HTML page doesn't show up as the largest of it's kind for testing against https://www.sitespeed.io
https://github.com/sitespeedio/sitespeed.io/blob/master/lib/plugins/assets/aggregator.js#L36
@beenanner you are right, seems like a bug, I think I've missed checking in something in the past, the constructor misses some parameters, let me fix that.
Released this today, I'll close it now but please re-open @julrich if you still see it in the logs after upgrading to 5.4.4.
Perfect, will give this a spin... and report back!
About 30-50 scans later, error gone 馃挴 :)
thanks @julrich for reporting back!