TypeError: Cannot set property 'request' of null

Versions:

• node: 8.9.4
• apm-agent-nodejs: 1.0.0
• elasticsearch, apm-server: 6.2

the nodejs client throws (briefly after start when first sample is apparently processed) with

TypeError: Cannot set property 'request' of null
    at /APPDIR/node_modules/elastic-apm-node/lib/instrumentation/transaction.js:160:31
    at /APPDIR/node_modules/after-all-results/index.js:9:18
    at _combinedTickCallback (internal/process/next_tick.js:131:7)
    at process._tickDomainCallback (internal/process/next_tick.js:218:9)

corresponding entry in apm-server log: 2018-02-08T10:55:33.755Z INFO [request] beater/handlers.go:150 handled request {"request_id": "c64c3d8b-e467-4f79-91ce-0af2d185a214", "response_code": 202, "method": "POST", "URL": "/v1/errors", "content_length": 1410, "remote_address": "9.8.7.6", "user_agent": "elastic-apm-node/1.0.0 elastic-apm-http-client/5.1.0"}

content of elastic-apm-node.js:

module.exports = {
	active: process.env.NODE_ENV === 'production',
	serviceName: 'service-name',
	serverUrl: 'http://1.2.3.4:8200',
	flushInterval: 5,
	maxQueueSize: 40,
	stackTraceLimit: 100,
	transactionSampleRate: 0.1
}

gladly tell me what other information would be helpful 😃

[consense]

thanks - reducing stackTraceLimit to 5 only doubled usual cpu usage, which still seems high but than again leaves more than enough room to actually use APM for the next time without sampling. Will ping you in case I notice anything unusual. From a feature perspective both apm-server (inside kibana) as well as the agent are a joy to use - kudos to you and the rest of the elastic team!

[watson]

We just released version 1.0.1 with the fix for the sampling issue you discovered.

Regarding the CPU usage: Could I get you to send me an e-mail with some more data on your application, e.g RPM when you see the CPU increase and possibly also a list of your dependencies from the package.json? Then I’d love to see if I can figure out what’s going on: [e-mail]