Workflow intermittently hangs on Timer activityIssue Workflow intermittently hangs on Timer activity and the error seen in logs is "Attempt to get mutex wf_context_b62d7916dbc9c8187415ff971d9619a3 with spin failed because max spins was hit"Cause>The timer event is unable to get the context mutex for the workflow, so the event has to be deferred into the wf_command table.-> What should be happening but apparently isn't is:a) The mutex is unavailable because another thread is currently running the flow, that active thread should be picking up these events out of the 'wf_command' table and processing themb) There is a race condition in the engine that can cause wf_command events to be missed by the active thread. The backup for that is the Workflow Nudger scheduled job. This runs every 5 minutes and looks for flows that have commands stuck in the wf_command table and sends them a nudge.1. First check if this OOB 'Workflow nudger' scheduled job is active on the instance 2. If Workflow Nudger is active and is running every 5 minutes, there's an error that could be causing workflows to not get nudged:[code]<pre><code>2019-11-13 13:04:58 (208) worker.2 worker.2 txid=8c7d2716dbc1 Completed: Workflow Nudger in 0:00:00.098, next occurrence is 11-13-2019 16:09:58<br/>019-11-13 13:04:58 (203) worker.2 worker.2 txid=8c7d2716dbc1 WARNING * WARNING * Evaluator: com.glideapp.workflow.model.ModelRunTimeException: Cannot find and load workflow version:ec62e4f9db9a730480f296cadb9619cb<br/>Caused by error in Schedule Item: 'Workflow Nudger' at line 1<br/>==> 1: new SNC.WorkflowScriptAPI().nudge();<br/>com.glideapp.workflow.model.cache.WorkflowModelFactory.getModel(WorkflowModelFactory.java:37)<br/>com.glideapp.workflow.stages.WorkflowStageList.<init>(WorkflowStageList.java:109)<br/></code></pre>[/code] That is 'wf_workflow_version' (ec62e4f9db9a730480f296cadb9619cb) has been deleted, but there are many contexts still running for that workflow version:ResolutionTo resolve this issue either:1) Restore this workflow version2) or Cancel all running workflows for that versionThen monitor the workflow nudger job and if there are no further errors, we should see the number of records in the 'wf_command' drop significantly.