Partition is below target replica, node is stuck

Question

Partition is below target replica, node is stuck

mtalhaoui opened this issue 6 years ago · 10 comments

I'm trying to run a service fabric mesh applicaton locally in vs 2017, but as soon as the application is build I get some errors and warnings in the service fabric explorer. I am not able to run a hello world application.

Has anybody run into the same issue?

'System.FM' reported Error for property 'State'.
Partition is below target replica or instance count.
fabric:/Application2/Console1 1 1 d4f26725-aac7-4d82-a803-d442f78101f6
N/P InBuild _Node_0 132042979063713078

'System.RAP' reported Warning for property 'IStatefulServiceReplica.ChangeRole(P)Duration'.
The api IStatefulServiceReplica.ChangeRole(P) on node _Node_0 is stuck. Start Time (UTC): 2019-06-06 12:31:51.604.

Answer 1 · 2019-06-06T12:38:49.000Z

Does the issue persist when you remove the application?

Answer 2 · 2019-06-06T12:41:27.000Z

As soon as the application is closed, it goes away.

Answer 3 · 2019-06-06T12:43:15.000Z

Can you see it running in local Docker without issues?

Answer 4 · 2019-06-06T12:50:33.000Z

I have tried: docker ps/docker container ls and and they don't display anything.

Answer 5 · 2019-06-06T13:07:12.000Z

I had this sometimes, which is understandable in preview. Usually it goes away if I right-click and "remove mesh resources"
. Also try to drop all SF related docker containers.

Answer 6 · 2019-06-06T13:18:01.000Z

I have removed the Mesh App Resources. I have removed all the docker containers with the following command: docker system prune.

Now it seems I have an extra error message:

'System.Hosting' reported Error for property 'CodePackageActivation:Console1:EntryPoint:132042976053925389'. There was an error during CodePackage activation.System.Fabric.FabricException (-2147017731) Failed to start Container. ContainerName=sf-5-48102e82-f9dd-419b-81f9-09f9cba20f15_8b3651e9-6139-4287-a5b3-1bba66d9a1d1, ApplicationId=SingleInstance_3_App5, ApplicationName=fabric:/Application2. DockerRequest returned StatusCode=NotFound with ResponseBody={"message":"network servicefabric_nat not found"}

Answer 7 · 2019-06-06T13:55:01.000Z

I think you might need to re-create Mesh cluster as it adds docker networking component which is gone after prune.

Answer 8 · 2019-06-06T14:05:30.000Z

I have removed the Local Cluster and created a new Mesh 1 node.

I still have the same issue.

Answer 9 · 2019-06-06T15:47:02.000Z

I have tried to use 'reset local cluster' and this seems to have solved the issue caused by the 'docker system prune' command. I now have the original issue again: partition is below target replica.

It appears the issue was not solved by this.

Answer 10 · 2019-06-18T12:49:30.000Z

I've managed to solve this issue.

The problem was in my installation of vs 2017. The installation was seperated over 2 hdd's.