I have upgraded our OEM from 11g to 12c Cloud Control using the two system method. I am in process of deploying/configuring agents using the Enterprise Manager 12c Upgrade Console. I have encountered one agent in which the deployment has reportedly completed successfully but the Health Check shows the ping test failed and I don't understand why. I have found some information suggesting looking at the emd.properties and the gcagent.log file but I see no errors listed or indication there that anything is amiss. The os on the target is
Linux 2.6.9-42.EL #1 Wed Jul 12 23:15:20 EDT 2006 x86_64 x86_64 x86_64 GNU/Linux
the RDBMS version is 188.8.131.52.0 and the existing agent is 10.2.0.4.0., I don't know if that matters.
I don't believe it is a firewall issue as I have the proper ports open in both directions. I am not clear where to look or what to change to resolve this issue and be able to switch over. As far as I know the agent has been reconfigured from unsecure to secure or vice versa since the backup of the initial repository was made and I'm not sure how to proceed with that quesiton. One note I found appears to suggest unsecuring and resecuring the console but it isn't clear if they are suggesting the old or new console. There is some confusion as well as to whther they are referring to the agent or the console as well. Please help me resolve this problem.
As Nazim mentioned, check the port . Few things that u can look for:
Look at the log file for more error message:
For 2-system upgrade, if your old Management Agent was running in secure (or unsecure) mode before backing up the Management Repository, then ensure that it continues to run in the same mode while you deploy and configure the new Management Agent for it. Do not resecure the Management Agents after backing up the Management Repository. If you do so, the ping test might fail while performing the healthcheck because of a mismatch between the configuration stored in the repository and the actual configuration of the Management Agent. You will see a KEY_MISMATCH error in gcagent.log.
If any Management Agent is in unsecure mode, and if you have already upgraded the OMS, then run these commands on the upgraded OMS (12c OMS), generate the health check report again, and then switch over:
$<OMS_HOME>/bin/emctl secure unlock -console
$<OMS_HOME>/bin/emctl secure unlock -upload
If you still has confusion, i suggest open SR with support and they will guide you
Thanks for the information. I have a couple of followup questions:
1) What ports shoulw I be looking at? I have deployed other agents with no problems and am using the same ports. Is there perhaps another port I should look at?
2) Here is where I display my lack of understanding but is there a way to tell if a management agent is in secure or unsecure mode?
3) If I do unlock the agent and upload should I lock it after I switch over the agent? THis is something I don't understand clearly.
4) I looked at the gcagent.log file and don't see evidence of a KEY_MISMATCH. Is there some other error I should look for?
run emctl status agent from both the 11g and 12c agent homes. this will display the agent url for each of those agents.
the agent url will include the port, and it will be http if unsecure, and https if secured.
best practice is to run with the OMS secured and locked, so yes after you get all agents switched over the recommendation would be to secure and lock (requires agent communication to use https only), at this time the agents would need to be secured as well (emctl secure agent) so they can communicate properly.
Oracle Enterprise Manager Cloud Control 12c Release 2
Copyright (c) 1996, 2012 Oracle Corporation. All rights reserved.
Agent Version : 184.108.40.206.0
OMS Version : 220.127.116.11.0
Protocol Version : 18.104.22.168.0
Agent Home : /oracle/agent/agent_inst
Agent Binaries : /oracle/agent/core/22.214.171.124.0
Agent Process ID : 9909
Parent Process ID : 9856
Agent URL : https://host.com:3872/emd/main/
Repository URL : https://host.com:4904/empbs/upload
Started at : 2013-01-10 12:41:11
That makes sense but one other question belying my lack of understanding. The 12c agent is not running so the emctl status agent command simply says that so I can't see if it is secured. The 11g agent is. What are the ramifications of stoppong the 11g agent, starting the 12c agent and checking the status? If I do that do I then need to stop the 12c agent restart the 11g agent and then do the switchover?
As you can see I am still struggling wtih this and I appreciate the help.
I noticed one other anomaly and I am not sure if this has bearing on anything. The OEM 12c Upgrade Console lists the agent as being Linuxx86 but the operating system is Linux x86_64 and the RDBMS is 64bit. Could that have some bearing on the situation?