Surgient Success

We've encountered problem that has resulted in the potential loss of a days work. A snapshot was taken of a deployed application which took 24 hours to then error out (conveniently the original scheduled end time).

We are trying to replicate the problem the process that created the error was:

A Developer deployed an App made some changes and did a save & quit including the ram state. (save took 15 mins).

Developer then re-deployed the snapshot and made further changes to it and hit save & quit (no ram state). The estimated time to do this was around 24 hours (reservation end time less 15 minutes or so).

Upon waiting the 24 hours it then errored:

State: Saving snapshot...
Message: Surgient.Platform.Exceptions.AutomationException: 00370008 An error occurred creating a snapshot for on virtual machine vqms-PoolVm-6: ScheduledDeployment reservation timed out while waiting for an automation script to complete. .

 
Can we tell what caused the problem and why it took so long? It could have potentially lost the Developer a significant amount of work.
 

Surgient makes local backups of snapshot files so your work is not lost.  Please contact Surgient support for how to recover your snapshot data from backup.

Please include the log files that are captured by running C:\Program Files\Surgient\surgient-support.exe.  That can help determining what the issue with taking the second snapshot was.

From Ariel: 

Logs attached. Bare in mind that the problem occurred with the Capital Cell 1 application that terminated on the morning of the 4/7/08.

 Looking through the scripting.log, I see the following:

 
/4/2008 11:16:30 AM|Warning|snapshot-server.395 |[snapshot-server-395] Failed to log off Terminal Services sessions prior to guest OS shutdown.  If the guest OS is Linux, this is expected behavior. System.Threading.ThreadAbortException: Thread was being aborted.
   at System.Runtime.Remoting.Messaging.StackBuilderSink.SyncProcessMessage(IMessage msg, Int32 methodPtr, Boolean fExecuteInContext)
   at System.Runtime.Remoting.RemotingServices.ExecuteMessage(MarshalByRefObject target, IMethodCallMessage reqMsg)
   at Surgient.Platform.Services.Internal.ServiceProxy.Invoke(IMessage msg)
   at System.Runtime.Remoting.Proxies.RealProxy.PrivateInvoke(MessageData& msgData, Int32 type)
   at Surgient.Platform.Services.CommandService.Execute(RequestContext ctx, Server target, String command, NameValueCollection parameters)
   at Surgient.Automation.AutomationUtil.LogoffRdpSessions(Server server)
   at Surgient.Automation.AutomationUtil.KillRdpSessionsAndShutdownServer(ScriptContext context, Server svr)
7/4/2008 11:16:30 AM|Info|snapshot-server.395 |[ScriptEngine] The Script snapshot-server was aborted by request.
7/4/2008 11:16:30 AM|Info|snapshot-server.395 |[Surgient.Platform.Persistence] ScriptInstance (#761) deleted.  Fundamental properties of the deleted object:  ScriptInstance #761 [CommandId=11; LastSavedCheckpoint=0; SerializedRequest=System.Byte[]; Request=; SerializedData=; Data=; Id=761; IsVolatile=False; LastUpdateOn=7/3/2008 1:39:07 PM; IsDcgNode=False]
7/4/2008 11:16:30 AM|Info|snapshot-server.395 |[ScriptEngine] Script instance 395 "snapshot-server", vqms-PoolVm-6, ScheduledDeployment #95, DeploymentGroup #59, Owner: Nigel2 (serverconfigid="15", ownerid="4", systemlibraryid="1", scheduleddeploymentid="95")  has completed with status Aborted.

What OS is the image based on?  Is it Linux?  If not, were there any changes made to Terminal Services before saving the snapshot?

The VQMS.log had rolled along with the servicehost.log.

Can those be provided?   I'll also need the agent.log for the host.
 

I couldn't see how to add attachments to the forum so I'm e-mailing them through (probably not a good idea to have them on here anyway).

To answer the other questions the OS is Windows 2003 and no TS changes were made prior to the save. Although an IP address was moved from the loopback adapter to one of the nics.