0

I have a Service Fabric standalone cluster where I can connect to it with Connect-ServiceFabricCluster just fine, but whenever I run other commands such as Start-ServiceFabricClusterConfigurationUpgrade, it keeps timing out. I've tried running commands with longer timeouts (as long as 10 min) with the same results. Any ideas what I might be doing wrong?

This works..

PS C:\WINDOWS\system32> Connect-ServiceFabricCluster -ConnectionEndpoint "localhost:19000"
True

ConnectionEndpoint   : {localhost:19000}
...

But these commands time out...

PS C:\WINDOWS\system32> Get-ServiceFabricClusterConfiguration -TimeoutSec 300
Get-ServiceFabricClusterConfiguration : Operation timed out.
At line:1 char:1
+ Get-ServiceFabricClusterConfiguration -TimeoutSec 300
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : OperationTimeout: (Microsoft.Servi...usterConnection:ClusterConnection) [Get-ServiceFabr
   icClusterConfiguration], TimeoutException
    + FullyQualifiedErrorId : GetClusterConfigurationErrorId,Microsoft.ServiceFabric.Powershell.GetClusterConfiguratio
   n

Start-ServiceFabricClusterConfigurationUpgrade -ClusterConfigPath ClusterConfig.Unsecure.DevCluster.v2.json
Start-ServiceFabricClusterConfigurationUpgrade : Operation timed out.
At line:1 char:1
+ Start-ServiceFabricClusterConfigurationUpgrade -ClusterConfigPath Clu ...
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : OperationTimeout: (Microsoft.Servi...usterConnection:ClusterConnection) [Start-ServiceFa
   ...gurationUpgrade], TimeoutException
    + FullyQualifiedErrorId : StartClusterConfigurationUpgradeErrorId,Microsoft.ServiceFabric.Powershell.StartClusterC
   onfigurationUpgrade
Jessica
  • 167
  • 2
user2644378
  • 11
  • 1
  • 4
  • Some commands can very long to execute on the cluster, as the operations takes place one server at a time, and services needs to be migrated around as they're applied. Changing the cluster certificate is something I did recently which took almost an hour on a 5 server cluster. – Svend Jan 13 '20 at 08:27

1 Answers1

0

I believe this a single node dev cluster? If so, it can't be upgraded and needs to be recreated instead.

You can upgrade your cluster to the new version only if you're using a production-style node configuration, where each Service Fabric node is allocated on a separate physical or virtual machine. If you have a development cluster, where more than one Service Fabric node is on a single physical or virtual machine, you must re-create the cluster with the new version.

To fix:

  • Create backups for all desired service partitions.
  • Update the client tooling using the web platform installer.
  • Remove the old cluster and recreate it using the Local Cluster Manager tool (in the tray, if on Windows).
  • Redeploy applications.
  • Restore service backups.
LoekD
  • 11,402
  • 17
  • 27
  • Hi! Yes, are there instructions on how we can recreate the cluster and transfer over all the apps? – user2644378 Jan 13 '20 at 22:08
  • I've put some steps into the answer, hope that helps – LoekD Jan 14 '20 at 06:54
  • Thanks! Unfortunately, it seems we'll need to enable backup for the service fabric, and to do that, we'll need to be able to upgrade the configurations. Is this true? If yes, it seems like we're in a deadlock situation? – user2644378 Jan 14 '20 at 17:56
  • I think you're talking about the BackupRestoreService, but this is not required. You can do it old-skool like [this](https://xpirit.com/backup-and-restore-in-azure-service-fabric/). Do deploy the new cluster with the backup service though. – LoekD Jan 15 '20 at 07:12