Conference netcad::hub_mgnt

Hi,

I thing you have two separate problem. Problem with DECbridge 900MX is probably
caused by old firmware. I was using DECbridge 900MX with firmware version 1.4.0
and it worked well.

I had also the same GIGAswitch configuration as you and behaviour was exactly the
same. Pressing <BRIDGING> => Core dump in 100% cases.

Regards

Jarry

    Hi,
    
    The DECbridge 900MX firmware is 1.4 ,not 1.1. Sorry for my miss typing.
    
    The problem occour when you are using IP address only. MAC or DECnet is
    OK. 
    
    Looking For Engineer Help!
    
    Thank!
    
    David Lee.
    HKMCS

	We can do a couple of things to track this one down:

	1. We will try and reproduce the problem here and get back to
           you with questions and/or status.

	2. Can you make the core dump available over the net?

	3. If it will take some time to make the core dump available, you
	   can help us track this quicker.   If you have dbx on the
           workstation, go into the directory with the core dump and do a
	   

		dbx hubwatch core


	   Once in dbx do a where command.  This will show where the crash
	   occurred along with the subroutine call list up to that point.
	   If you could get us that information it would be a big help

							John
           

    Hi,
    
    The core will available on next Monday. I found 2 more problem on
    GIGAswitch.
    
    1) I cann't modify the FGL2 or FGL4  line card configuration using 
       HUBwatch. For Example, I want to change the timer or connection 
       policy. After click [APPLY] button, the following error message
       comes out:
    
       Timed out while waiting for SET reponse from FGL2 line Card.
       The SET may have failed.
    
       What Wrong?
    
    2) In FDDI MAC Summary screen, the field UN DA Flag display grable.
    
       Status
    
        :
       UN DA Flag: ZZuyZZuyZZuy..........
    
    3) The colour problem about FGL4 port status is not correct.
    
    
    Thank for your Support.
    
    David Lee.
    HKMCS
     
    
     

    
	We were not able to reproduce this problem with our GIGAswitch.
	Just to check our boards were at the following revisions:

	FGL-2   2.08
	FGL-4   1.05
	Clock	2.10
	SCP     2.16

	One other thing that can be tried to give us more information is
	the following.  Set the environment variable SNMP_TRACE and
	run HUBwatch.  You'll get all sorts of SNMP messages before
	the front panel comes up.  Go to the view with the bridge
	button.  Once you hit the bridge button see if any SNMP
	messages show up on the trace before you get the segmentation
	fault.  If so, can you let us know what they are.

						John

    How about the ports status colour of FGL4, is it correct ?
    
    Thank!
    
    David Lee.

                                                                          
                  
Hello ,

There are three files named core_900mx, core_giga and core_add_ip which is in 
the location "hgcs05::/usr/OV/RHKJC" directory. IP address of hgcs05 is 
16.157.0.220 and Decnet address is 22.921. 

File		Sympton
----		----------------------------------------
core_giga	Click the [briding] button, Segmentation Fault (core dump)
core_add_ip	Setting IP address filtering in DECbridge 900MC module
		Segmentation Fault (core dump).
core_900mx	Display forwarding database using IP address.
   		Segmentation Fault (core dump).

Also,  could you tell me the pointer of lastest version for GIGAswitch ?
    
    
Thank!

David Lee.

        The BL2.1 release is the latest full customer release for
    GIGAswitch/FDDI.
    
        My guess on why the sets don't work is that possibly you neglected
    to specify the read-write community, or got the case wrong.  The
    default read-write community (required for GIGAswitch/FDDI SNMP set
    commands) is the box address of the GIGAswitch/FDDI system.  The box
    address is the lowest address in the multi-address rom, lower case, no
    spaces, no "-", no ":" (I.E. 08002ba00280).  You get the box address by
    either looking on the front of the GIGAswitch/FDDI system (it's on the
    label on the front of the clock card (slot 7)), or by running OBM and
    doing a show box configuration command.
    
        If Hubwatch uses ifOperStatus to determine the LED colour, then
    it's probably the known bug which should be fixed in BL2.2 (entering FT any
    day now).

    Hello,
    
    The GIGAswitch System Release Ver(2.1) contain the following firmware
    version.
    
    SCP OP2.10, SCP BB 2.00 SCP DL 1.00
    FGL-2 v2.00
    FGL-4 v1.00
    AGL-2 v1.00
    CLK   v2.10
    PSC   v2.00
    
    It seem to have minor version different in -5.
    
    I want to have the lastest ( may be FT version ) firmware pointer ?
    
    Thank!
    
    David Lee.
    

    David, you might find a pointer to the latest Gigaswitch firmware
    in the Gigaswitch notes conference, on SCHOOL::GIGASWITCH
    
    Bob

       I reiterate;  BL2.1 is the latest customer release.  BL2.2 has not
    been released yet, not even the FT version.
    
        There have been interim bug fix releases given to specific
    customers to fix specific problems.  In general, bug fix releases are
    not given the same level of testing as the official releases.
    
        If you feel you require something more recent than BL2.1, contact
    Sandra Renda (NPSS::RENDA).
    
        As Bob suggests, the GIGASWITCH notes conference is a good place to
    find current information about the GIGAswitch/FDDI system.

    Thank Bob and Micheal.
    
    David.

    
    
    Was there a solution to this problem? I have a customer with the same
    problem and he also appears to be at the recommended revs for all of
    his options (gigaswitches, bridges etc.).
    
    He has a core dump available if one is needed.
    Should I IPMT this call?
     

    
    
    HHHHHHHHHHEEEEEEEEEEEEELLLLLLLLLLLLLOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
    
    

        Did you submit the IPMT, or are you expecting to get the problem
    resolved here?

    IPMT submited.
    
    Thank!
    
    David Lee
    

    David,
    
    I have a customer having the same problem. Can you tell me the IPMT
    number and status of this problem.
    
     Reagrds
    
    Richard

    
    Suppose I should have mentioned:
    
    Hubwatch version 4.0.1
    OSF 3.2
    
    When he brings up the FGL2 card and clicks on bridging it core dumps.
    After reading a few more notes on this problem I was under the
    impression it was fixed in v4 of hubwatch??
    
    Thanks
    
    Rich

    
    Can anybody give me a status on this problem please.
    
    Rich

    
    Rich,
    
    	Can you specify the customer's GIGAswitch/FDDI configuration? Also,
        if possible, could a 'dbx where' (pointed out in a prior reply) be
        done on the core file? It would also be useful if we could get an
        SNMP trace just prior to the crash. The last half page or so of
        output should be enough to get an idea of where the problem may be.
    
        I have tried this on several GIGAswitches with no luck at
        reproducing it, although we do not have a fully populated
        (max ports) switch available (anyone have one on the net that we 
        could look at?).
    
    Regards,
    Bob

      ...I will contact Bob offline about access to a fully configured
    GIGAswitch/FDDI system...
    
    MDL

    Thanks Bob,
    
    I'll get the info you requested. watch this space!
    
    Again thanks for your interest
    
    Rich.