Video Screencast Help

Jobs doesnt work but individual tasks do (v.7.1.8400)

Created: 29 Jan 2013 | 15 comments

Hello everyone,

 

I'm having an issue where version 7.1.8400 doesnt want to run any jobs we created. For example we've made job where we deploy specific image then reboot to prodaction and after that we want to run two custom tasks (windows activation and agent removal). However after image deployment nothing happens. Deployment window disappears and machine doesnt reboot if you dont turn it of by pressing power off button. If you run those tasks induvidually they all work just fine so I dont think it's up to them.

 

What could be a reason for this issue?

 

 

Thanks in advance,

Leo

Comments 15 CommentsJump to latest comment

mclemson's picture

Could you post a screenshot of your job configuration, including a screenshot of each task in the job?  Also, if you leave the job alone, rather than reboot manually, what does the job/task status look like on the server for that test computer?  (If the image fails, then that's why the reboot isn't occurring; if the image succeeds but the reboot fails, then there might be other causes.)

Mike Clemson, Senior Systems Engineer, ASC
Intuitive Technology Group -- Symantec Platinum Partner
intuitivetech.com

leomilano's picture

Image is 'running' status forever after the deployment window disappears on client computer. I've tested by leaving that computer over night to run and next day it still hasn't rebooted.

 

 

It's very strange that software of this scale doesn't include original task for it's agent removal.

mclemson's picture

You shouldn't capture images without running sysprep or the Prepare for image capture job.  This will cause duplicate SID and other Windows issues, and will cause duplicate GUID issues in Altiris which are a pain to resolve.  You should never have to run a cleanup script like the one you're using to remove the agent and reset the GUID.

You may want to review this article:
http://www.symantec.com/docs/HOWTO30267
 
Specifically, under 2) Getting Going, read #5, which explains the types of jobs you should use to gather and deploy images.  The 'Prepare for image capture' task mentioned is the one that strips out the GUID from the agent.  There is no task for removal since the agent doesn't need to be removed.
 
Does this help?

Mike Clemson, Senior Systems Engineer, ASC
Intuitive Technology Group -- Symantec Platinum Partner
intuitivetech.com

leomilano's picture

Screenshots were taken from some testing job. Sure we prepare images with sysprep and that's not the case.

Second shouldn't decision for removing agents be up to us not up to Symantec? I cannot see any benefit of leaving it running in background if you dont plan to image that machinge for very long time. Older Altiris had this feature but for some strange reason it has been removed.

mclemson's picture

If you don't need the agent, you can remove it after imaging.  I had thought you were trying to resolve not preparing the image/sysprepping by running commands post-image.

Mike Clemson, Senior Systems Engineer, ASC
Intuitive Technology Group -- Symantec Platinum Partner
intuitivetech.com

Thomas Baird's picture

and using /clean is not supported in the current release.  They're re-adding it in v7.5.  Today, you should use aexnsagent /uninstall instead.  In 7.5, /clean will actually call the /uninstall to preserve product integrity.

You'll probably have to dig a bit to find out why the reboot to production doesn't work - sounds like the previous job ran, but did not run successfully (like a failure with DeployAnywhere, for instance, which shows a success with the image, but fails AFTER the image - though you don't have that selected).

So change the job, just for a test.  Run a script with a message box or something and a pause.  You finalize the box or pause in command prompt, the script moves on.  The next task is the reboot to prod task.  The script mentioned is to REPLACE the imaging task you have above.

Then, because you did NOT image it, you replace the other tasks with otherwise bogus tasks, just to see if they run.

 

If that process completes, then we know that the failure is right around the imaging task.  If it's the image, try a new one, as suggested above.

 

In short, let's narrow this down.  :D

Thomas Baird
Private Consultant

 

leomilano's picture

Thanks for letting me know. I will remvoe /clean line. I need to test as soon as I have some free time for this.

Atm. I'm strugguling with another issue where image capture doesn't start (error code -1). Changing DNS, rebooting client machine & Altiris server doesn't help at all.

 

I'm sorry to say but Altiris 7.1 is very unreliable software.

Thomas Baird's picture

If you have issues with capturing an image, we'd need to know where the error occurs.  For instance, it may fail in the console and we'd find details in the server logs.  Or it may fail in automation and we may see the job never arrive, or something in the agent / pectagent logs.  Or it may fail during Ghost, and we'd see the errors in the Ghost logs.

So if you do need help with this, 1) a new thread is probably good and 2) if you could give us more details, it'll help a lot.

 

Good Luck!

Thomas Baird
Private Consultant

 

leomilano's picture

Hi Thomas,

 

Please tell me which logs (including their location) do you need and I will try to pride them to you.

Sally5432's picture

If you run a reboot to pxe command and then a a reboot to production command on the computer, does that work?

I had an issue where image jobs were failing to run post image tasks, but it was because when I sent a reboot to production command - it was just going back to PXE.

If I deleted the machine from the console, the issue went away.  It's a bug with MP1 I think.

---
Don't forget to mark posts as helpful if they are, and mark answers as solutions.

leomilano's picture

Doesn't seem to work. :(

 

 

I've created job which includes 2 original tasks. Machine reboots in PXE and after a few seconds of 'loading Altiris'-phase, machine continues to reboot to the OS. It looks that it just skips the task and go to the next bootable source.

Nelo's picture

Try separating automation jobs/tasks with production jobs/tasks.  Make sure the check box to move to the next task if the current one fails is not enabled.

Tim.Jing's picture

We have teh same issues with some Jobs, but not all. Some image capture with sysprep/Ghost will not even get in to PE even witha  reboot to PE task as the first thing.

But a Image capture will work if we use f8 to get in to PE.

Thomas Baird's picture

Wow guys - there are a lot of different issues here.  I strongly suggest different threads if you're not on the original question.  Here's why:

1) Reboot to PE/Production "failures" can happen for a handful of reasons, including not sending the SBS files to the right PXE server, systems not getting the task, systems failing to run the task because they failed something else, systems not being recognized by the PXE server, SBS files ariving but not being handed off to the PXE Server Service from Interface.  I've seen all of these at one time or another.

2) Using F8 to get into PE via PXE has little or nothing to do with a successful reboot to task.  It just means the system was waiting to get somewhere, and did so.  Consider it like this.  Let's say you send a reboot task to a computer.  You go to the computer 30 minutes later, and while sitting there, you shut down and restart it.  When you return to the console, the task shows as successful.  Did it succeed, or did you rmanual intervention make it look like it succeeded?  When you use F8 to get into PE, you've learned nothing at all about any reboot-to task.

3) If a system "fails to reboot to production" we need to know EXACTLY what tha means.  For instance, after imaging, maybe the next job never fires because the imaging task shows as failed?  Maybe the system DID reboot but went back into automation?  Maybe the system tried to reboot but hung?  Maybe it went to production, but the image was damaged and it went back to automation and never reported up success?

 

Please, all of you, first, if your issue is not the same as the person who posted, start a new thread.  Second, when you say something that is designed to happen in automation "fails" take note of exactly what you see happen and document that here.  We can't help troubleshoot if we're just guessing, and I think we've done enough of that so far or we'd have more fixed issues.  :D  Even an "Well, I see ghost run, and then nothing happens - it just sits there" is a really good place to start.

Thanks!

Thomas Baird
Private Consultant