Error I get in zeebe broker:
2020-05-30 06:27:53.874 [Broker-0-Exporter-1] [Broker-0-zb-fs-workers-0] WARN io.zeebe.broker.exporter.elasticsearch - Failed to flush at least one bulk request ElasticsearchException[Elasticsearch exception [type=mapper_parsing_exception, reason=object mapping for [value.customHeaders.payload] tried to parse field [payload] as object, but found a concrete value]]
zeebe | 2020-05-30 06:27:53.874 [Broker-0-Exporter-1] [Broker-0-zb-fs-workers-0] WARN io.zeebe.broker.exporter.elasticsearch - Failed to flush bulk completely
zeebe | 2020-05-30 06:27:54.295 [GatewayLongPollingJobHandler] [Broker-0-zb-actors-1] DEBUG io.zeebe.gateway - Remove blocking request BrokerActivateJobsRequest{requestDto={"type":"lwm2mApiCall","worker":"b90523e4-68e4-4652-b476-7cdd52d80250","timeout":1200000,"maxJobsToActivate":132,"jobKeys":[],"jobs":[],"variables":[],"truncated":false}} for job type lwm2mApiCall after timeout of PT30S
I am using customHeaders and have a ‘payload’ custom header in several tasks, with different values, sometimes it’s a simple text string, sometimes it’s a string containing json
My guess is the elasticsearch is not handling this, as it is applying parsing on what should really be a simple string. This decision is left to elastic, which creates the mapping by guessing the types
Is there any way around it?
E.g. can zeebe exporter be configured with the desired mapping ?