I'm having a problem when trying to add a new agent in the site configuration for project deployment in BCC.
When I create a new agent that belongs to a host different from the one of the BCC (different real servers), the new agent is not being found, showing the inaccessible error, and this is the status on DeploymentServer: ERROR_TRANSPORT_UNINSTANTIATED
I try to create an agent with this configuration:
and the BCC host is 126.96.36.199.
So, the agent is not being found when the host is different.
P.S: I've tested telnet and the server are communicating with each other in the port used.
Does anyone know what's happening or what's missing?
Any help would be appreciated,
ERROR_TRANSPORT_UNINSTANTIATED only means that the server you are trying to connect with the specific port does not exist or is shutdown.
Please check whether the commerce server at 188.8.131.52 is running on rmi port 8870
You can easily check that in dyn/admin of commerce server at /atg/dynamo/Configuration and check the rmi port.
Also make sure the port and host are accessible from bcc server instance if this is under any linux environment, since it shows the 2 servers are on different ips.
The other instance for which you are setting up the agent (184.108.40.206) is not a CA server(BCC) right? If BCC, i don't think you can set up an agent for another BCC instance.
Also, can you make sure that the nucleus is running in the target instance (220.127.116.11)? Normally you get this error when the instance is down or when AgentTransport component is not running in the target instance? If the instance is running, can you try to access /atg/epub/AgentTransport in that target instance's dyn/admin?
The server mentioned is running, as for the linux, I've tested if there was any connection between both servers, bcc and agent, which have different ips. The connection was tested through the telnet command, connecting to ip 18.104.22.168 and port 8870 and it was successful. Though, when configuring the agent through BCC, the error I've mentioned before happens.
I don't know if it could be anything wrong with the agent configuration or if there's anything in ATG that is forbidding the connection between servers with different ips. Because, if it is an agent that is present in the same server, with the same ip as the publishing, it works.
In the same instance (server) of the agent (22.214.171.124) there are two instances of production (impossible to connect to the agent through the bcc), one instance of lock production and one of agent (CSC), but there is no publishing running on that machine.
The agent instance is running normally, including the AgentTransport, which is successfully running and I can access it through dyn/admin.
Thanks for confirming that.
Can you do the following please? As Guruvinder mentioned, please go to /atg/dynamo/Configuration in the target instance and make sure that the rmiPort is 8870 as you have configured. Also check whether you have a siteHTTPServerName which is not 126.96.36.199. If yes, try changing the host name while setting up the agent.
If everything is correct, click delete default cluster name inside /atg/epub/DeploymentAgent in target instance. And also make sure that all the datasources are pointing to the same in case of SwitchingDataSource
Let me know how it behaves after that.
I had this problem once, access /atg/epub/DeploymentServer on CA server and click on instanciate agent (or something like this) on the agent that is not being instantiated. Then, check the log if there is something like this.
008-11-10 17:57:58,908 ERROR
[nucleusNamespace.atg.epub.DeploymentServer] transport error from
java.rmi.NoSuchObjectException: no such object in table
If yes, then you have a firewall problem on your network, like others sad here. To resolve it, check this workaround or check your network configurations to work with RMI.