Infrastructure Deployment Walkthru

Document ID : KB000072390
Last Modified Date : 06/03/2018
Show Technical Document Details
Introduction:
CA Client Automation uses Infrastructure to deploy agents to machines with no agents.
There are many steps in the process and the only Error you get if it fails is "No Primer Transport"
This document helps you to figure out where and hopefully why it is failing.
Question:
I am getting  "No Primer Transport" , which is a generic Error, How do I figure out where in the Infrastructure Deployment process is it failing ?
Environment:
CA Client Automation - All Versions
Answer:
BASIC INFRASTRUCTURE DEPLOYMENT TROUBLESHOOTING CHECKLIST

If you are having issue using Infrastructure Deployment, here is a good checklist to go thru before calling support,
then you can find out exactly where your issue is and if you can tell the tech which step you are stuck at,
it should make it quicker and easier to resolve your issue.

1.     Can the remote admin$ (Or configured replacement) share be accessed?

1a.   Can this share be accessed as 'localsystem'?

2.     Do a 'ping' of the target and then a 'ping -a' of the returned IP to verify it is online and resolves to the expected DNS name.

3.     If correct, do the same from the agent to the DM (Note the SS is not yet involved at this point).

4.     Check the target admin$ share for 'dmsetup.exe' and 'dmkeydat.pmr' (.cer)

5.     If there, check for the existence of the path '..\program files\ca\dmprimer'

6.     If not there then trigger port 135 TCP is likely being blocked from DM to agent.
         NOTE: Note, even though there is an option to 'Deploy via SS', this part always comes from the DM.

7.     If the folder exists, try to CAMPING the agent from the DM, this SHOULD succeed.

8.     Even if this fails, which is potentially valid if you are deploying via SS and have restrictions?
         It should NOT fail when you camping the agent from the SS.

9.     Ping the SS from the agent, perform a reverse lookup and verify correct info. as in #2 above.

10.   CAMPING the SS from the agent and ensure proper result.

11.    Run 'CAMPING <SS name> -s 40000' to ensure camping works with larger packets of the size used in a deployment.

12.    Perform the same camping test from the SS to the agent.

13.    Now try 'camping -a <SS name>' from the agent and make sure no error is shown in the output.

14.    Try 'camping -a <agent name>' from the SS and make sure no error is shown in the output.

15.    If that all works, check the '..\program files\ca\dmprimer' folder for <SS/DM name>.pmr (.cer in 12.5)

16.    If not there, it's likely an authentication issue. Contact support.

17.    If there, check the '..\program files\ca\dsm\dmprimer' folder for the existence of a folder called \CAUnicenterDSM\.

18.     If not there, check the same folder for a file called 'primer1.pack'.

19.    If primer1.pack is there but no folder from #17, then the transfer confirmation was never received or responded to by the SS.

20.    If the '..\ca\dsm\dmprimer\CAUnicenterDSM' folder exists then the agent DID receive the agent package and acknowledge that, but the agent install failed.

21.    From here the agent should be installed.

22.    CAF should start automatically CAF STATUS

23.    Check if all the CA agent processes are installed DSMVER


 
Additional Information:
Infrastructure should only be used to deploy and Agent , if you are upgrading an Agent (that has an SDAgent),
you should use a SD Package thru Software Delivery.