Unable to Deploy Diagram using Desktop Modeller to Self Managed Development Environment

Camunda 8 Topics Discussion & Questions
1

I’ve cloned GitHub - camunda/camunda-platform: Links to Camunda Platform 8 resources, releases, and local development config and started everything up with docker compose up -d. All my services look healthy in Docker Desktop and I’m able to login to the various services using the default credentials.

I’ve created a basic workflow in Camunda Desktop Modeller and am trying to deploy it, following these instructions: Deploy diagram | Camunda 8 Docs

Configuration details:

When I do so, I get: Expected to execute command on partition 1, but either it does not exist, or the gateway is not yet aware of it [ deploy-error ]

Google isn’t giving me anything that is particularly useful. Has anyone else counted this? I’m on an M2 Macbook Air. Keycloak appears to be running fine. Thx.

2

Hi Justin!

I had exactly the same issue on a k8s cluster with Camunda self-managed deployed into it. Did you already tried to restart zeebe and the zeebe-gateway? If not, you can try that and check the logs if everything starts up correctly.

Regards,

Maarten

3

Thank-you for responding Maarten. I was starting to worry that I’m the only one with this error :confused: I don’t know if the error I found in the zeebe logs is relevant:

io.camunda.zeebe.util.actor - Uncaught exception in 'Startup' in phase 'STARTED'. Continuing with next job.
io.atomix.raft.storage.StorageException: Failed to open metastore
...
Caused by: java.nio.file.AccessDeniedException: /usr/local/zeebe/data/raft-partition/partitions/1/raft-partition-partition-1.meta

As for zeebe-gateway, I’m not sure. If that a separate container from zeebe? I don’t see it listed separately in docker-compose.yaml. I see quite a few references to the word GATEWAY in docker-compose.yaml but that’s all.

Restarting zeebe doesn’t make any difference.

4

Hi Justin,

I think we it is just the end result we share ( a non working zeebe cluster) but the underlying problems are different. I also had disk problems (full disk) so maybe that was my root cause. When I google around a bit I find some similar issues as you have. People describe that they upgraded or mixed up zeebe versions or use snapshot versions of zeebe. Maybe this is a pointer which can help you solving the issue? Maybe clean docker volumes? Clean docker images/container?

Good luck!

Regards,

Maarten

5

Thanks for the suggestions Maarten. I’ll poke around and try some more. This is 100% blocking me from starting to play with Camunda 8. I hope I can figure it out.

6

I haven’t made much headway yet but I have learned about the command line tooling:

zbctl deploy TwoStepProcess.bpmn --insecure

results in: Error: rpc error: code = Unavailable desc = Expected to execute command on partition 1, but either it does not exist, or the gateway is not yet aware of it which is the same error I’m getting in the Camunda Modeller.

zbctl status --insecure

outputs:

Cluster size: 1
Partitions count: 1
Replication factor: 1
Gateway version: 8.3.4
Brokers:
  Broker 0 - 172.23.0.5:26501
    Version: 8.3.4

I’m not sure what the the expected output of zbctl status is supposed to be. If I had to guess, this looks OK to me.

7

well, not completely ok, it should be looking like this

Cluster size: 1
Partitions count: 1
Replication factor: 1
Gateway version: 8.3.4
Brokers:
Broker 0 - camunda-zeebe-0.camunda-zeebe.camunda.svc:26501
Version: 8.3.4
Partition 1 : Leader, Healthy

I should try to clean everything and start it again and check for error message while starting zeebe.

9

I finally found something that worked: docker volume prune --all