Hi there,

I'm trialing VisualCron to see if it will work for a project we're planning. During my testing over the last two days the VisualCron service has crashed a few times, with the errors at the end of this message.

I see that Windows Server 2012 is not listed in the system requirements for the software, is Windows Server 2012 not supported by VisualCron? Or is this something else?

I'd appreciate any help you could give.

Thanks,

Tom

Application: VisualCronService.exe
Framework Version: v4.0.30319
Description: The process was terminated due to an internal error in the .NET Runtime at IP 000007FD0BA23CB9 (000007FD0BA00000) with exit code 80131506.

Faulting application name: VisualCronService.exe, version: 6.2.2.14786, time stamp: 0x5165168e
Faulting module name: clr.dll, version: 4.0.30319.18046, time stamp: 0x51552911
Exception code: 0xc0000005
Fault offset: 0x0000000000023cb9
Faulting process id: 0x934
Faulting application start time: 0x01ce731120a17b11
Faulting application path: C:\Program Files (x86)\VisualCron\VisualCronService.exe
Faulting module path: C:\Windows\Microsoft.NET\Framework64\v4.0.30319\clr.dll
Report Id: 72dae0ab-df07-11e2-93f4-d4ae52c9f064
Faulting package full name:
Faulting package-relative application ID:

Yes, it is supported. What do you do when it crashes? Please also send the log_serverDATE.txt from the log folder to support@visualcron.com

---

Most of the time it's when I'm editing powershell entries, but it seems to happen when the application is idling waiting for me to do things too.

I've attached the log file requested.

The server we're running it on is a Hyper-V host, if that matters.

Just editing it sounds strange. It would be great if you found a pattern exactly what you do and compare it to the crash time.

---

Well, it's just happened now while I was editing some PowerShell in one of my test job's tasks, I was in the middle of typing the word 'return'.

Prior to that I had tried to run my job, and it had got stuck at the same task I am editing. I say stuck: it was just taking longer than I'd expect it to take given the PowerShell script I'm using.

I don't think I terminated the job before editing the task.

Ok, then it is probably not related to writing/editing in Client but rather that it crashed when you tried to run it earlier. Please provide more feedback based on this.

Try for example creating a very simple row and run it compared to a more complex powershell that you normally run.

---

I think I've figured out what's crashing the service: I'd referenced a non-standard PowerShell cmdlet in my script without adding the required snapin first.

Here's my script as it stands now:

Add-PSSnapin VeeamPSSnapin
$cdsJobStartString = "{JOB(Active|LastRun|dd/MM/yyyy-HH:mm)}"
$cdsStart=[datetime]::ParseExact($cdsJobStartString, "dd/MM/yyyy-HH:mm", $null)
$cdsVeeamJob = Get-VBRJob -name '{JOB(Active,Variable,VeeamJob)}'
$cdsVeeamJobName = $cdsVeeamJob.Name

This runs fine, but if I comment out the top line the VisualCron service crashes.

If I run the same script in a PowerShell session with the top line commented out I get the following error, which VisualCron doesn't seem to be handling:

Get-VBRJob : The term 'Get-VBRJob' is not recognized as the name of a cmdlet, function, script file, or operable program. Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
At line:4 char:16
+ $cdsVeeamJob = Get-VBRJob -name '{JOB(Active,Variable,VeeamJob)}'
+ ~~~~~~~~~~
+ CategoryInfo : ObjectNotFound: (Get-VBRJob:String) [], CommandNotFoundException
+ FullyQualifiedErrorId : CommandNotFoundException

Some of the tasks we need to run require PowerShell. My hope was that VisualCron would handle unexpected errors meaning I wouldn't need to account for them all in the script myself ('re-inventing the wheel' as I go).

Our intetion is for this to run unattended on a regular basis, meaning we need to be alerted when things go wrong.

As such we'd need an error such as this to be handled. If for whatever reason the PowerShell snapin we require failed to load we would want VisualCron to log the error and notify us, not to crash.

Briefly I've also just found that PowerShell loops don't behave as I expected inside VisualCron tasks: If I make a job with a single task that runs the following script:

while ($true -ne $false){
write-output "Looping!"
}

And then set the timeout for the task to 30 seconds, this task runs beyond 30 seconds, and I am unable to terminate it from the GUI.

Thanks for the information - we will try to reproduce this and also see what we can do about generally loading snapins.

---

Hi,

could you please let us know if this beta version works better for you:

http://neteject.com/down...Cron/VisualCron7.0.0.exe 

---

Thanks, I'll give that a try and let you know.

Tom

The beta handles the Powershell cmdlet not-found error properly, thanks for that.

However it still fails to stop a looping Powershell script after the timeout. The same test I performed earlier still runs past the 30 second timeout I set.

I've actually got round this now by putting in VisualCron loops. So rather than having a looping Powershell script waiting for some condition to arise, I have a single 'if' statement run that returns 'true' if the condition is true. VisualCron is then set to loop that single task as long as the return code isn't 'true'.

Are we okay to use this beta in production?

Thanks,

Tom

Could you attach a sample looping script (preferably with a pause so it does not use too much cpu).

I would say, at this stage, that the latest beta is ok for use in production but please update it when the public version is released.

---

No problem:

I have a job called 'Test' with a single Powershell task called 'Loop'. The Powershell task is set to timeout after 30 seconds and the job itself is set to timeout after a minute.

The Powershell task runs the following script (input as text, not a script file):

While ($true -ne $false){
Write-Output "$(Get-Date -f dd/MM/yyyy-HH:mm:ss)" >> C:\Test.txt
Start-Sleep -Seconds 10
}

If run normally this script should obviously loop forever, writing the date and time to a log file I've already made, and then pausing for 10 seconds.

Given the timeouts on the task and the job however I'd expect VisualCron to terminate it after 30 seconds.

In actual fact the job runs past both timeouts (right now its still running and has been for four and a half minutes).

On top of that if I try to stop it 'via the right-click menu on the task, the job, or from the 'Running Jobs' window nothing happens, it just keeps running.

Interestingly I've noted that the Windows process ID listed in the 'Running Jobs' window for this task is '0'.

If I restart the VisualCron service the loop stops.

Thanks for the test script.

We have now implemented both Stop of Powershell script and update of Windows PId in "Running Jobs" window for version 7.0.0. We will make a new beta build tomorrow so you can test it.

---

That's great! Thanks for your quick responses on this.