Worker job fail command reduces retries to 0 even when indicating 3

[patozgg]

Issue tracker is ONLY used for reporting bugs. New features and questions should be discussed on our slack channel.

Expected Behavior

I am using the following command in my worker (https://github.com/camunda-community-hub/zeebe-client-node-js#connection-behaviour)

	job.fail({
		errorMessage: 'Triggering a retry with a thirty second back-off',
		retryBackOff: 30000,
		retries: 3,
	})

which as per the documentation the retries should stay at 3 as I am providing a value. However, after running such code the retries go down to 0 and an incident is created.

Current Behavior

Retries get reduced to zero and an incident is created. Instead I would expect the retries to stay at 3.

Possible Solution

Steps to Reproduce

1. Have Camunda 8 environment running
2. Setup worker with the job fail command
3. Setup BPMN with service task
4. Run worker and process

Context (Environment)

I am using version 8.2.3 of https://github.com/camunda-community-hub/zeebe-client-node-js#connection-behaviour and version 8.2.5 of Camunda 8.

Detailed Description

Possible Implementation

[jwulf]

Thanks for the report.

I have a fix for this. I need to extend the test coverage, and I should be able to release a fix for this tomorrow.

[jwulf]

Version 8.2.3 is published with this fix in it: https://www.npmjs.com/package/zeebe-node