• Navigate to C:\Program Files\Microsoft SDKs\Service Fabric\ClusterSetup
  • (Assuming that %systemdrive% is the C: drive)
• Within this ClusterSetup directory, update the 4 manifest files
  • Update each of the ClusterManifestTemplate.json files in these folders
    • NonSecure/FiveNode
    • NonSecure/OneNode
    • Secure/FiveNode
    • Secure/OneNode

note: will probably need admin privileges to update

• Stop the cluster
• Remove the cluster
• Delete SfDevCluster folder from C:
  • (Assuming that %systemdrive% is the C: drive)

• Reset the cluster (if applicable, or create the cluster)

source: SO post

What to update in these manifest files

• Update the fileshare of the following path values
  • diagnosticsStore.connectionstring
  • FabricDataRoot
  • FabricLogRoot

note: the fileshares in these path values may be "%systemdrive%" or the actual drive letter

sources:

• SF issues github
• SF replicator log size

This file is for use by SF reliable stateful services. It is defaulted to 8GB!!

• according to the referenced docs, it is static; so it remains at whatever size it is set to
• according to the referenced docs, the minimum value is 1024 (1GB)

Within the fabricSettings array node, add the KtlLogger object

"fabricSettings": [
    {
        ...
    },
    {
        "name": "KtlLogger",
        "parameters": [
            {
                "name": "SharedLogSizeInMB",
                "value": "1024"
            }
        ]
    },
    {
        ...
    },
]