After setting node type i,e Reliability tier to sliver to bronze on azure service fabric, error on cluster health is waring

0
votes

After setting node type i,e Reliability tier to sliver to bronze on azure service fabric, error on cluster health is waring here below is the error evaluation from service fabric.(Even in vmss of service fabric

Services Warning Unhealthy services: 100% (1/1), ServiceType='ClusterManagerServiceType', MaxPercentUnhealthyServices=0%. Service Warning Unhealthy service: ServiceName='fabric:/System/ClusterManagerService', AggregatedHealthState='Warning'. Event Warning Unhealthy event: SourceId='System.PLB', Property='ServiceReplicaUnplacedHealth_Secondary_00000000-0000-0000-0000-000000002000', HealthState='Warning', ConsiderWarningAsError=false. The Load Balancer was unable to find a placement for one or more of the Service's Replicas: ClusterManagerServiceName Secondary Partition 00000000-0000-0000-0000-000000002000 could not be placed, possibly, due to the following constraints and properties:
TargetReplicaSetSize: 5 Placement Constraint: NodeTypeName==NOde Depended Service: N/A

Constraint Elimination Sequence: ReplicaExclusionStatic eliminated 2 possible node(s) for placement -- 1/3 node(s) remain. ReplicaExclusionDynamic eliminated 1 possible node(s) for placement -- 0/3 node(s) remain.

Nodes Eliminated By Constraints:

ReplicaExclusionStatic -- No Colocations with Partition's Existing Secondaries/Instances:

FaultDomain:fd:/0 NodeName:_NOde_0 NodeType:NOde NodeTypeName:NOde UpgradeDomain:0 UpgradeDomain: ud:/0 Deactivation Intent/Status: None/None FaultDomain:fd:/2 NodeName:_NOde_2 NodeType:NOde NodeTypeName:NOde UpgradeDomain:2 UpgradeDomain: ud:/2 Deactivation Intent/Status: None/None

ReplicaExclusionDynamic -- No Colocations with Partition's Existing Primary or Potential Secondaries:

FaultDomain:fd:/1 NodeName:_NOde_1 NodeType:NOde NodeTypeName:NOde UpgradeDomain:1 UpgradeDomain: ud:/1 Deactivation Intent/Status: None/None

Help me to slove this problem

1
azurenodesazure-service-fabricazure-vm-scale-set

1 Answers

2
votes

When you create your cluster with Reliability tier Silver it will provision 5 replicas of the system services, i.e. the services that essentially are Service Fabric.

Downgrading from Silver to Bronze means that you change the target replica count of these services from 5 to 3.

In order for SF to place replicas on nodes it evaluates a set of constraints, on of these being that it does not want two replicas of the same service partition to end up on the same node.

As it looks from your error you have one Node Type with 3 nodes in it but you still have Silver reliabilty tier, that means that SF is unable to find a node for the last two of your replicas for the system services (in your log it is System/ClusterManagerService, but same applies for all system services).

Make sure that your cluster has at least as many nodes as your reliability tier needs, i.e. 3 nodes for a Bronze tier, 5 for a Silver and so on.

Also, what you are seeing is a warning that the cluster is not able to uphold it's characteristics, but it should still be running, right?