StorageException: Failed to acquire storage lock; ensure each Raft server is configured with a distinct storage directory

Camunda 8 Topics Discussion & Questions
1

Hello,

I am trying to upgrade our Zeebe from v0.26.1 to v1.1.0 (tried v1.0.0 as well) but I ran into “StorageException: Failed to acquire storage lock; ensure each Raft server is configured with a distinct storage directory” error and now our brokers won’t start.

Helm values for Zeebe:

image:
      repository: camunda/zeebe
      tag: 1.1.0

    clusterSize: "1"
    partitionCount: "1"
    replicationFactor: "1"
    logLevel: DEBUG
    cpuThreadCount: "2"
    ioThreadCount: "2"
    pvcSize: "10Gi"
    pvcAccessModes: [ "ReadWriteOnce" ]
    resources:
      requests:
        cpu: 3000m
        memory: 2Gi
      limits:
        cpu: 6000m
        memory: 3Gi
    probePath: /ready
    readinessProbe:
      failureThreshold: 3
      periodSeconds: 30
      successThreshold: 1
      timeoutSeconds: 12

Any help is appreciated!

Complete error log in Zeebe cluster: StorageException: Failed to acquire storage lock; Zeebe - Pastebin.com

2

Upgrading is not supported from 0.26 to 1.x.

3

I see. I had to delete our PV/PVC for fresh deployment, this resolved the issue.

I’m also getting this error in operate io.camunda.operate.exceptions.OperateRuntimeException: Import is not possible for Zeebe version: 0.25. I assume this is also because of unsupported upgrade as our Elasticsearch still contains data from earlier version?

4

Correct. You need to do a completely new install to go between 0.x and 1.x.

2 Likes
5

Thank you!

6

btw is it the same also for any upgrades henceforth from 1.x ? Meaning we need to always start clean slate?

7

There is no upgrade path guarantee between major versions, no.

8

thanks. is this somewhere documented?

9

There is this: Public API | Camunda Cloud Docs