Tenticle NullReferenceException

When we try and deploy to our production environment, we get the following error (since upgrading to 3.0). This only happens to the production environment, and the actual deployment succeeds (both services and websites have been tested, and are on separate hosts).

Object reference not set to an instance of an object.
System.NullReferenceException: Object reference not set to an instance of an object.
   at Octopus.Shared.Diagnostics.AbstractLog.PlanFutureBlock(String messageText) in y:\work\refs\heads\master\source\Octopus.Shared\Diagnostics\AbstractLog.cs:line 72
   at Octopus.Server.Orchestration.Deploy.DeploymentPlan.RetentionRequired(String machineId, List`1 retentionTokens) in y:\work\refs\heads\master\source\Octopus.Server\Orchestration\Deploy\DeploymentPlan.cs:line 363
   at Octopus.Server.Orchestration.Deploy.ActionCommand.ApplyRetentionPolicy(Machine machine) in y:\work\refs\heads\master\source\Octopus.Server\Orchestration\Deploy\ActionCommand.cs:line 79
   at Octopus.Server.Orchestration.Deploy.Package.PackageActionHandler.Execute(ActionCommand command, Machine machine) in y:\work\refs\heads\master\source\Octopus.Server\Orchestration\Deploy\Package\PackageActionHandler.cs:line 27
   at Octopus.Server.Orchestration.Deploy.Guidance.Execute(Action callback) in y:\work\refs\heads\master\source\Octopus.Server\Orchestration\Deploy\Guidance.cs:line 50
   at Octopus.Server.Orchestration.Deploy.DeploymentTaskController.ExecuteActions(PlannedStep step, Machine machine) in y:\work\refs\heads\master\source\Octopus.Server\Orchestration\Deploy\DeploymentTaskController.cs:line 196
   at Octopus.Shared.Tasks.OctoThreadClosure`1.Execute() in y:\work\refs\heads\master\source\Octopus.Shared\Tasks\OctoThreadClosure.cs:line 28
Octopus.Server version

Hi Andy,

I’m not from octopus, but I’m also encountering the same error. Does that build happen to have a manual intervention step? It seems (from my testing anyway) that if you remove the manual intervention step, the deploy works fine. Have you noticed this as well?



You are correct, removing the Manual Intervention step does resolve the issue (and as the manual intervention only runs on a production deploy for our projects, it explains why only production deploys are effected).

Thanks for the heads up (and lets hope a fix is done soon!)


We have a github issue logged for this https://github.com/OctopusDeploy/Issues/issues/1701 . I’ve added a note with both your tickets to the issue to make sure we let you know as soon as this is fixed.

Sorry for the inconveniences caused by this issue