Video Screencast Help
Symantec to Separate Into Two Focused, Industry-Leading Technology Companies. Learn more.
Endpoint Management Community Blog
Showing posts tagged with 7.x
Showing posts in English
Tim.Jing | 05 Feb 2013 | 0 comments

This feels like being smacked at every turn. So Monday morning fresh start, we did get ghost image to work last Friday right?

  1. First up, huge issues, GhostUser account keeps sitting there after imaging. Someone has to go hit the OK button to the GP related Eula for the Ghost job to finish(What ever that task is?) So basically no automation.
  2. Second, 2 of the other Child NS's will not keep pxe running.

Cases put in to Symantec, of course it will be at least a day before we get anyone on the phone. So Google, and Connect it is.

By 3PM, we found the cause of both issues.

1. GhostUser does not log in. There is a new version of DeployAnywhere that resolves this, and it's a known issues since 4-17-2012, why is this not in MP1??? So look and get the files from this link below if you have the same issue.

...

Tim.Jing | 04 Feb 2013 | 4 comments

Finally got Image deploy to work where it would take the computer name in the console!

It starts with Using ghost(Which is crap IMO). Then Using Symantec's built in Capture image job, then deploy. And you have to use 64bit PE because there is a known issue where 32bit PE does not have Firm which causes the unattended.xml to not get copied, and the job would hang and fail after imaging.

So today, more image testing with 32 bit, and throwing out all the great logic we had planned with this product.

Original logic:

Image x -> Script in the correct agent installer(hierarchy) -> Copy local all base sw -> Boot to windows and use unattend.xml to install all software and agent.

Now it's back to:

Ghost-> Tasks->

So the big hurdle. Since Ghost is doing it's thing in the back ground, we have no way to manipulate the agent(It configures to the aprents server the image was captured on). So does this mean in...

Tim.Jing | 01 Feb 2013 | 2 comments

To put it mildly, the Altiris SMP 7.1 MP1's DS suite is not looking good. And if you have a working DS 6.x, do not migrate to DS 7.x!

Especially if you came from a previous DS 6.x back ground. The pxe now named SBS is flakier than the combo's at Long John Silvers, and the UI design is as simple as solving the Super Strings theory.

So the facts:

PXE works, but it does strange stuff at times and the only way to resume functions when a machine refused to function within PXE is to Stop all SBS services and clear out the Task Handler folder.

PC join domain, naming. Mother of all nightmares. How this could be such a complicated set of dance moves is shocking. This is an Enterprise product?

PC cant be re-named in WinPE, so the randomized name gets used as the PC name after the jobs at times. 

Here is to a good productive day on resolving the PC join domain after imaging issues. But it's future is bleak based on what we are seeing...

Tim.Jing | 30 Jan 2013 | 6 comments

This week, diving deeper in to the functional side of DS with some Windows 7 testing along with software deploying.

So we all know the new DS is different, but I never knew it was going to be this different. And Different is not good at all from an DS Administrator's perspective. So here is a low down.

  • Software manager is a mess. Maybe from an accountant's side this may be decent, but for an admin, it bestows more work, more clutter, and a interface that serves hardly any benefits. Bottom line, Too many steps to create a job, and not any benefits over the existing platform.
  • New PXE has some issues. The SBS corrupts at times causing machines to give the Device is Not Capable errors. Again, another solution to a problem that didnt exist in DS 6.x or any of the competitor's products. 
  • Software delivery is badly executed as well. There are Tasks and Jobs. While a Job can contain multiple...
Ludovic Ferre | 29 Jan 2013 | 0 comments

Update! Added the Dataclass and Filters top level items as they also needed fixing. Also I added a reference to my original SP1 to SP2 upgrade post, with steps for fixing the issue [1].

####

I attended an upgrade for a customer installing MP1.1 on 4 servers in hierarchy (1 parent, 3 children). We had some minor problems when we got started but nothing major:

  • 2 servers (including the parent) did not have the MP1.1 upgrade option because SMP 7.1 SP2 was not appearing as installed in SIM. We just ran the upgrade as a new installation.
  • The parent configuration went much to fast to be healthy. In effect it did not install most of the products, so we ran the "aexconfig /configureall" manually and this fixed the environment
  • One role had been granted undesired rights (inherited from the hidden parent folder)  on main console locations: "Reports", "Console menu", "Settings"...
Tim.Jing | 25 Jan 2013 | 0 comments

The white board sessions seems to get longer, and the endless syntex errors are just a blur dulled by time.

But we are making steady progress. Altiris 7 doesnt suck like everyone said it would, and other than a very crappy(Personal opinion) console design, everything is pretty smooth.

Things accomplished:

Hardware independent image done, took 3 days, completely modular.

Base Software installed through Unattended.xml. Again modular based on what we copy in to the partitioned image duing WinPE.

Agent installer logic build into WinPE with variables to install agent to the correct Parent/Child Server.

PCAnywhere tested and working perfectly.

SQL sort issues kind of resolved- gotta specify ASC and DESC on a few servers. So we just specified for all at the top and let hierchy replicate them down.

The Grocery List for next week:

Populate Software manager for SVS, Desktop Deploymen.

Win...

Ludovic Ferre | 23 Jan 2013 | 0 comments

The Symantec product listing xml was updated Monday.

Here are the information we have from it:

File name: symantec_v2.pl.xml
File hash: 3e447dd07844f9fad531a3240215a11b
File date: 2013-01-18 15:37
File size: 13,990,445 bytes (~14MB)
Release date: 2013-01-21

From the Git commit [1] we can see the following changes:

  Add /solutions/7_1/mobilemgmt/7_2_sp2_1_rtm/symantec_mobileframework_7_2_sp2_1_x64.msi_info ...
Ludovic Ferre | 22 Jan 2013 | 1 comment

I just finished a remote session with a customer that found a computer from the database that is not sending data back.

I had received the log files yesterday and the log viewer was all red and blue. Upon inspection it was clear that the agent COM components are not working (not registered or unregistered), so no basic inventory, client session management or sub-agents could perform their normal tasks.

Still the agent was getting it's policy file from the server, but not being able to do anything with the given policy (given its state).

We checked the computer Windows logs and found nothing interesting in there, so we crafted a SQL query to detect other computers with similar issues:

select 
       distinct(s.ResourceGuid),
       i.Name, MAX(_eventtime) 'Last config request',
       MAX(s.ModifiedDate) 'Last Basic Inventory',
       MAX(s.createdDate) 'First inventory',
       DATEDIFF(d, max(s.modifiedDate),
       MAX(c....
Tim.Jing | 14 Jan 2013 | 5 comments

Items completed:

3 SMP servers, 1 parent, 2 child with hierchy completed.

SQL maintenence plan created per Symantec recommendations.

AD import completed.

Subnets manually created(We are using the 8 scheme which would have been tough to import)

Tested PCAnywhere functions, decent product, not very good at screen aspect ratios for such a mature product. Very bare bones?

Only issues are: 1 server had to be respun in VM and rebuilt becuase of various strange errors with SMP agent. Also some Java oddities when importing OS files, which thankfully can be bypassed by simply pressing cancell to the credential challenge(This was on XP and 7 desktops with the Symantec's proposed GP configuration in palce. But that error still persisted, attached is the photo of the error, on an XP machine you can press cancell and continue with the file upload).

The design of the product still irks me a bit for the options on Site Servers. Even with fiber...

Tim.Jing | 09 Jan 2013 | 2 comments

A few days in to the migration, have some questions came up. So lets go through it.

Our infrustructure is more like 1 huge building than 80 separate sites. The availability of having fiber connected to all the sattalite campuses means we can centrally manage as if they were in the same building. The problem is SMP isnt really designed for that kind of infrustructure. And the few issues we encountered are:

1. The technicians cannot manage the infrustructure in real time from 1 console. The amount of our clients meant we had to use hierchy, which means teh Parent server sees things later(Synch schedule). While all synched machines will be on the Parent server, any new clients will take 24 hours for the Parent console to see. Which means no OOB deployment from the parent Console.

2. Replication of packages maybe required as Task Servers, which means we would have to replicate our packages atleast 5 times(5 Task Servers). We dont need that as our San can handle the...