Conference azur::mcc

| Is the DECbridge 600 supported?

    Yes.

| If so, is there something special that needs to be done
| to register it?  All I did was give it the ethernet address, and it
| didn't get registered...  I can see it with NCP and it says it's
| reachable...

    Shouldn't be anything special to register it.  You can see if it's
    reachable with the following two commands:

    MCC> test bridge <address>
    MCC> show station <address> all status

    If those commands both work, there shouldn't be any reason why you
    can't register it.

    MCC> register bridge <name> address <address>

    If you are having trouble, please post the errors you are getting.

| Is it an oversight that the BRIDGE AM was not included in the
| MCC_STARTUP_BMS.COM file?

    Um, it was a product marketing decision. :-)  The Bridge AM is no
    longer part of the DECmcc BMS package, you need to obtain the DECmcc
    ELM AM package to get the Bridge AM now.

    Ed,
    	I'm confused by your reply to the second part of my basenote...  I
    guess I should've stated that I'm working with EMS V2.2.  I have the
    ELM AM, however, I didn't *think* I was doing anything with it... Are
    you telling me that the icon I get in the toolbox comes from ELM?  
    	Also, that still doesn't explain why the BRIDGE_AM.EXE file is in
    MCC_STARTUP_BMS.COM....NOT! or does it?  
    
    confused...
    
    /Mic
    
    BTW, the error message I'm getting from the DECbridge 600 is that it's
    partially registered....thanks for your help so far!
    

Well, yes actually.

The bridge icon (and corresponding AM) is provided by the ELM kit.

The ELM kit is part of the EMS package, but not part of the BMS
package, and thus is not included in the BMS startup procedure.

-- Erik

    
    Thanks Erik..
    
    /Mic
    

    Regarding the basenote, 4096, I have done some more research on the
    DECbridge 600.  Apparently, the customer has set it up as a Phase V
    Router.
    
    From NCP, I can see it's DECnet address, 1.501, however, I can't get
    any information on it, ie:  NCP TELL 'node' SHOW EXEC CHAR;  I suppose
    I need to run NCL?  And be on a DECnet Phase V node?
    
    Anyway, here's my current dilemma...  I tried to register this guy,
    using the BRIDGE AM, no luck...  I come up with partial registration...
    can't access node, (something along those lines).
    
    Next, I tried to register it as a DNA5 global entity, however, when you
    use the IMPM, the line for the ethernet address is inaccessible!  A
    bug?
    
    Next, I tried to register using FCL,
    
    FCL> REGISTER NODE 'nodename' ADDRESS=08-00-2B...
    
    I get a similar error message, regarding unable to access entity...
    
    Just for chance, I tried using the STATION AM, using the hw as well as
    the ethernet address...same results, and I event tried the DNA4
    AM...obviously same results...  Any ideas?
    
    thanks,
    
    /Mic
    
    Mic JUNG::GALVIN
    

Are you sure the beast is a DECbridge 600?

Could it be a DECnis ???

    � Could it be a DECnis???
    
    BINGO!
    
    That's what it is.  Have you had any experience with it Chris?  If so,
    how do I register/manage/alarm....pray tell!  I'm on the edge of my
    seat... 8^).
    
    thanks,
    
    /Mic
    

    The DECNIS should be registered as a Node 5.  Then when you open it you
    will see the normal router-type stuff and some bridge stuff too.

    Hi.  What about if the DECnis has been setup as a Phase IV Router? 
    When I tried using the DNA5 AM, I believe I get a partial
    registration...  btw, if the icon comes up with a '.', does that simply
    mean we're talking Phase V jargon???  smile.  Thanks in advance.
    
    /Mic
    

The routing  algorithm  in  use has nothing at all to do with the management
protocol.

>     Hi.  What about if the DECnis has been setup as a Phase IV Router? 

It is  still  a  Phase  Vnode,  even  if  it is running the Phase IV routing
algorithm.

>     When I tried using the DNA5 AM, I believe I get a partial
>     registration...  

Make sure  you  can talk to it (using the synonym which is what DNA5 AM will
be trying to use).  Try using NCL to it.

The times  I  have  seen  this  are  because  the synonym information is not
usable (DECdns problems, etc.).

Graham

    Or, if you are not using DNS, maybe the node isn't defined in the local
    NCP database.  If NCL doesn't work either, I usually go with the
    following (even though it doesn't make sense) because sometimes NCL's
    error messages are not as meaningful as they could be:
    
    	ncp> tel <the-decnis> show exe
    
    Now, this error message is definitely going to fail because ncp cannot
    be used to talk to a DECNIS.  But it is the actual error that is
    important.  If you get:
    
    	%NCP-F-CONNEC, unable to connect to listener
    	-SYSTEM-F-UNREACHABLE, remote node is not currently reachable
    
    then you've got a network problem or maybe the DECNIS is defined with
    the wrong DECnet Phase IV address (or running with a different address
    than it is defined with).
    
    If you get:
    
    	%NCP-F-CONNEC, unable to connect to listener
    	-SYSTEM-F-NOSUCHNODE, remote node is unknown
    
    then the name isn't defined
    
    But if you get:
    
    	%NCP-F-CONNEC, unable to connect to listener
    	-SYSTEM-F-REJECT, connect to network object rejected
    
    Then you probably have the DECNIS defined in NCP properly and therefore
    the problem is in MCC somewhere.

    Kelly,
    	Hi.  I got the  Network Object Rejected message, earlier, now I'm
    ok...FM, I guess?  I do have a question with accessing the DECnis using
    the DNA5 AM, how come it's so sloooooooooooooow!
    	I click on the DECnis icon, from the IMPM, go get coffee, walk
    about 10 minutes, and viol�, the child entities finally come on the
    screen...!
    	My configuration is EMS V2.2, VMS5.5-1, 36MG of Memory on a 3400,
    yeah I know...the CPU's too slow...but THAT slow?  I don't have any
    ALARMS polling, I don't have any EXPORTER jobs running, no HISTORICAL
    stuff....just barebones IMPM...any ideas?
    	A MONITOR SYSTEM *does* show CPU 100% flat out, but come on the
    DECnis is in the next roommmmmmmm!
    
    /Mic
    

    I'm only another poor, dumb user, but my guess is this...
    
    From a terminal window, before you double click the DECNIS the first time, 
    do an NCP> ZERO NODE <the decnis> COUNTERS.  Then double click the DECNIS
    icon.  When it finally comes up, go back to the terminal window and do
    an NCP> SHOW NODE <the decnis> COUNTERS.  Check out how many connections 
    have been sent to the DECNIS during the elapse time.  
    
    My guess is that MCC is creating, using, and tearing down a seperate
    logical link for each of some long list of operations it does when you
    first open something.  For a very fast system and a lightly loaded
    router both being on the same lightly loaded LAN, this might not be
    completely terrible.  But it still amounts to sending aproaching three
    times the number of messages.  
    
    As to why MCC consume all the CPU time when you do this, who knows.
    But, don't feel too bad, it does the same thing to me and I'm using a
    VXT 2000 and a VAX 7000-610 with 128Mb and no other users.

Of course, all this changes with version 1.2-a of MCC.