Error: 13 INTERNAL when publishing multiple messages on the same workflow instance

Camunda 8 Topics Discussion & Questions
1

Hello,

When publishing a bunch of messages to zeebe, I get Internal error response in the client.
I use Node Client to start workflows, for 100 createWorkflowInstance I get this error 60 times :

(node:19) UnhandledPromiseRejectionWarning: Error: 13 INTERNAL: Unexpected error occurred during the request processing
    at Object.exports.createStatusError (/usr/src/app/node_modules/grpc/src/common.js:91:15)
    at Object.onReceiveStatus (/usr/src/app/node_modules/grpc/src/client_interceptors.js:1209:28)
    at InterceptingListener._callNext (/usr/src/app/node_modules/grpc/src/client_interceptors.js:568:42)
    at InterceptingListener.onReceiveStatus (/usr/src/app/node_modules/grpc/src/client_interceptors.js:618:8)
    at Object.onReceiveStatus (/usr/src/app/node_modules/zeebe-node/dist/lib/GrpcClient.js:145:36)
    at InterceptingListener._callNext (/usr/src/app/node_modules/grpc/src/client_interceptors.js:568:42)
    at InterceptingListener.onReceiveStatus (/usr/src/app/node_modules/grpc/src/client_interceptors.js:618:8)
    at callback (/usr/src/app/node_modules/grpc/src/client_interceptors.js:847:24)

Also tried directly with zbctl.darwin command:
zbctl.darwin --insecure create instance chatbot --variables '{"userInput": "hello", "correlationKey":"1234", "chatFinished": true}'
duplicated 100 times and i get the same error
Unexpected error occurred during the request processing for half of my requests

zbctl status

Cluster size: 3
Partitions count: 3
Replication factor: 3
Gateway version: 0.24.2
Brokers:
  Broker 0 - cwa-dev-zeebe-0.cwa-dev-zeebe.development.svc.cluster.local:26501
    Version: 0.24.2
    Partition 1 : Follower
    Partition 2 : Follower
    Partition 3 : Leader
  Broker 1 - cwa-dev-zeebe-1.cwa-dev-zeebe.development.svc.cluster.local:26501
    Version: 0.24.2
    Partition 1 : Leader
    Partition 2 : Leader
    Partition 3 : Follower
  Broker 2 - cwa-dev-zeebe-2.cwa-dev-zeebe.development.svc.cluster.local:26501
    Version: 0.24.2
    Partition 1 : Follower
    Partition 2 : Follower
    Partition 3 : Follower

I found this issue https://github.com/zeebe-io/zeebe/issues/3675 and tried to add the requestTimeout parameter but I still have the problem and for my use case I dont want to wait for the result.

Thanks for your help

1 Like
2

This looks like it should opened as an issue against the broker.

3

Yes you’re right
For information we solved the problem by adding gateway instances

4