[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference azur::mcc

Title:DECmcc user notes file. Does not replace IPMT.
Notice:Use IPMT for problems. Newsletter location in note 6187
Moderator:TAEC::BEROUD
Created:Mon Aug 21 1989
Last Modified:Wed Jun 04 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:6497
Total number of notes:27359

4993.0. "CMA exceptoin crashes DECmcc starting hubwatch" by GIDDAY::DRANSFIELD (Mike Dransfield, Sydney RSSG) Tue May 04 1993 04:14

    I have a cma exception crashing DECmcc V1.3 when I try to start
    HUBwatch T1.1-18.
    This only happens on one particular machine and not on others running
    the same software. I urgently need this fixed as it is for an
    evaluation of hubwatch...
    
    the error is:
    %CMA-F-EXCCOP, exception raseid; VMS condition code follow
    -SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual
    address=00000018, PC=002407B8, PSL=03C00004
    %TRACE-F-TRACEBACK, ...
    					rel pc		abs PC
    					0004B4C4	0004B4C4
    					0007C382	0007C382
    					000771DC	000771DC
    					0007A878	0007A878
    
    
    thanks,
    Mike
T.RTitleUserPersonal
Name		DateLines
4993.1HUBwatch doesn't use CMAQUIVER::HAROKOPUSTue May 04 1993 18:0015
        Hi Mike,
    
        We haven't been able to reproduce this here but
        HUBwatch doesn't use CMA so I don't think this is a HUBwatch
        problem.   Does anyone know of any bugs in the MCC applications
        launching?   
        
        Try running hubwatch  standalone from the system that is
        giving you the problem and let me know what happens:
    
        $watch/agent <name> /ip=<ip addr> /comm="comm name" /v=l
    
        Thanks,
    
        Bob
4993.2more info..GIDDAY::DRANSFIELDMike Dransfield, Sydney RSSGTue May 04 1993 20:1615
    Hi Bob,
    thanks for the reply.
    HUBwatch works just fine. I don't think the problem is anything to do
    with HUBwatch as such, it just so happens that when I try to launch
    HUBwatch that MCC dies.
    I should also add that this is VMS 5.5-2, MCC V1.3 CMA 1.10-030 (the
    same version as the release notes say it has to be)
    
    I can't reproduce it either, except on this one particular site.
    The machine is a VAX 4000/60 with plenty of memory and resources set up
    as per mcc_audit or better.
    
    I'll open a CLD today...
    thanks,
    Mike
4993.3Another site another CMA exception...GIDDAY::DRANSFIELDMike Dransfield, Sydney RSSGTue May 11 1993 08:0511
    Ok,
    another site, another 4000/60, another CMA exception.
    This time nothing to do with HUBwatch (it isn't even here)
    This is a clean install of V1.3 BMS and V1.0.2 TSAM on VMS 5.5-2.
    I think I was messing around with registering stuff.
    I haven't got the traceback, but am going back on site tomorrow and
    will get a copy to put here.
    
    Another CLD?
    thanks,
    Mike
4993.4append to existing cld if possibleGOSTE::CALLANDERTue May 11 1993 16:515
    if you could append this new data to the existing CLD that would
    be appreciated.
    
    thanks
4993.5same traceback as .0...GIDDAY::DRANSFIELDMike Dransfield, Sydney RSSGWed May 12 1993 23:478
    re: .4
    thanks for the reply..
    I just checked and the traceback is exactly the same as the one
    which is caused by invoking HUBwatch. So the problem is clearly nothing
    to do with HUBwatch and only seems to happen on fast, well configured
    machines!
    thanks,
    Mike
4993.6Reproduceable any time...TOOK::LYONSAhh, but fortunately, I have the key to escape reality.Mon May 24 1993 11:508
    I have also seen this error, on a 4000/90 with 80 meg.
    At the time, we were trying to set filters in a "bridge"
    type of device. (e.g. gigaswitch).
    
    It can be reproduced any time you want, here in TAY2.
    
    drl
4993.7VCSESU::WADEBill Wade, VAXc Systems &amp; Support EngWed May 26 1993 11:515
    Is there a QAR number for this?
    
    I checked the External QAR Database DECmcc Version 1.3 with no luck.
4993.8cld number..GIDDAY::DRANSFIELDMike Dransfield, Sydney RSSGTue Jun 01 1993 20:336
    re: .7
    I have a CLD open on this
    Local call is STLQ50264, I don't know what the IPMT number is but the
    CLD is being handled by Alex TOOK::SOKOL.
    thanks,
    Miker
4993.9**NOT** CMARACER::daveAhh, but fortunately, I have the key to escape reality.Thu Jun 03 1993 09:015
It's not really a CMA execption, it's simply the fact that the
CMA code "signs up" to deal with all exceptions that someone else has
not dealt with.  Kind of like a "last chance".  The result is a bunch
of exceptions that all look the same but have very different causes.
4993.10How do we fix this?GIDDAY::DRANSFIELDMike Dransfield, Sydney RSSGTue Jul 27 1993 21:457
    re: .9
    So how do we find out what the problem is?
    
    This is still a serious problem. How can we get enough information to
    get it fixed?
    thanks,
    Mike
4993.11problem solved at lastGIDDAY::DRANSFIELDMike Dransfield, Sydney RSSGWed Mar 02 1994 16:4410
    This problem is now fixed.
    It seems if the mcc_common:mcc_appl_hubwatch.def has a
    /community in it, then it fails (or maybe it is having three arguments)
    
    I removed that and it now works ok..
    
    (I now have an exception when doing a "look up" in the impm, but I will
    put that in another note)
    
    Mike