I retrieved this from the Google Group function. Please respond if you still have a problem and please respond if you do not so that we can all know the solution.
Points I can make are as follows: Once you have a working link, you do not need to bother explaining what the link technology is. The effect you describe is consistent with having only End Nodes - but you say you are using Network Nodes. (In fact, with your configuration, the first and third nodes could be End Nodes with only the second a Network Node.) The effect you describe is also consistent with having only Low Entry Networking (LEN) connectivity between your nodes so that the APPN capability of the nodes themselves does not come into play. Because you are clearly generally familiar with SNA (since you support SNI - no mean intellectual feat!) it may be traditional - maybe very traditional - SNA with which you are familiar where all relationships were defined and no relationship was allowed to be dynamic - good for security - bad for getting unfamiliar configurations working. Thus you may have defined the locations of your partners somehow (LOCADDR=0 or CDRSC ALSLIST) so that you were successful with sessions over the LEN links. You do not need Border Node support with your configuration since all your Network Nodes have the same NetID. Probably your ADJCLUST definitions are not being used at all. Once you have got this working and you set about testing an SNI conversion where you will have two Network Nodes with different NetIDs, then you should start taking an interest in how Border Nodes work. The whole point about APPN (within what we call a "topology subnetwork") is that VTAM (in this case, other platforms in other cases, of course)works out how to do everything with no definitions other than absolutely necessary. Only if you have security concerns do you need to introduce definitions to limit possibilities. However I had better qualify that last sentence, since we are talking about an SNI conversion, by saying that you might want to introduce some guiding definitions when Border Nodes are introduced in order to avoid pointless searching within the assembled topology subnetworks - just as you may be doing today with your SNI definitions. 08400007 - how familiar "Procedure not valid for resource - Resource not found" Finally I note that you mention CP to CP sessions so that destroys my theory that you are using only LEN connections. Not quite finally - "I read all the books I could find and as far as I understand VTAM should forward the search request to other nodes after coding the next cp with the local cpname (with no BNs)." It may be possible to read too many books <g> and certainly have too many definitions. This sentence implies you are using what I might describe as "LEN logic" in that I can imagine this is a way of encoding the rule that "A LEN node needs to have a definition of a partner LU (for an LU local to the LEN node) which identifies the owning CP of the partner LU as the CP in the adjacent node on the path to the partner LU node." I could imaging that such definitions could cause the 08400007 since this is likely to confuse poor VTAM trying to handle inappropriate static definitions alongside its own appropriate dynamic definitions. Chris Mason > Original message as copied from Google Groups follows < Hi, I have created an APPN network in order to test it and drop my SNI connections. I have 3 APPN network nodes connected in a chain using EE links and MPC. Nodes: NN <-> NN <-> NN Links: MPC & EE EE I want to be able start a session from the first NN to the far NN... Sessions between the ADJ nodes are OK. Its seams like the middle NN will not forward the request to the third NN in the network. I tried to APING and DIRECTRY,SCOPE=NSEARCH but the results are the same... I have an ADJCLUST defined in the following way in every node (with the local corresponding name in each node). DEFAULT NETWORK NN1 NEXTCP CPNAME=NN1 MYNET NETWORK NETID=MYNET NN1 NEXTCP CPNAME=NN1 As far as I understand that should tell VTAM to forward the search request to other nodes... All of the nodes have the same NETID, and I can't figure out why the request is not carried on to the next node. The APING returns with the following sense code 08400007. Does the NN must be border Nodes for it to forward the request??? I coded the BN=YES, BYDYN=LIMITED and added the nodes to the ADJCLUST but the CP-CP session between the CPs are not BN for some reason... I can't figure out what I'm missing... I read all the books I could find and as far as I understand VTAM should forward the search request to other nodes after coding the next cp with the local cpname (with no BNs). Please help... Thanks in advanced, Offer Baruch. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
