Conference azur::mcc

5114.0. "%MCC-E-DSPMMLOCFILERR cannot be ignored !" by KETJE::PACCO (Gallia divisa est in partes tres) Tue May 25 1993 13:07

    Since I rebooted my ULTRIX SYSTEM V4.3, whenever I want to access
    DECmcc I get the following:
    
    The ICONIC MAP hangs with a message window, and the POLYCENTER Map
    window, without any data or text.
    
    In the FCL_PM:
    No such host mgkk01.bc for MM mcc_registration_fm
    %MCC-E-DSPMMLOCFILERR, dispatch local management module file access
    error during probe
    
    
    
    "mgkk01.bc" is the hostname of my system.
    "MM mcc_registration_fm" was "MM mcc_dna5_am" when I tried to start
    	the event logging instead of the FCL_PM.
    
    
    According the DECmcc message manual %MCC-E-DSPMMLOCFILERR should be ignored,
    but should be of type "Information" instead of "Error" ???
    
    The message "No such host mgkk01.bc for MM ..." cannot be found
    anywhere.  I am lost !
    
    Another test gives the following:
    
    hostname
    mgkk01.bc
    manage
    MCC> enroll Q_ENR_DUMP_TABLE
    No such host mgkk01.bc for MM Q_ENR_DUMP_TABLE
    %MCC-E-TRANSMITERROR, error trying to transmit a packet
    
    then after I give the command
    hostname mgkk01
    manage
    MCC> enroll Q_ENR_DUMP_TABLE
    Cannot find MM file
    %MCC-E-ENROLL_FAILED,  enrollment failed: table is not fully inserted.
    
    I suspect the ULTRIX hostname has something to do with this.
        
    
    Before I rebooted my system I added some SNMP MIB's, but I do not
    see any relation with it.  Much longer ago the name of my system was
    changed because of BIND, from mgkk01 to mgkk01.bc.
    
    What can I do to get out of this black hole?
    
    Regards,
    	Dominique.
    

Please post the results of:

# hostname
# grep `hostname` /etc/hosts
# nslookup `hostname`

Have you previously rebooted after changing your hostname for bind?
Are you sure that MCC has worked since that time?

I suspect that you may need to remove

	/usr/mcc/mcc_system/mcc_dispatch_table.dat

and then re-enroll all modules.

Please file a QAR when you have isolated this problem.  There is at
least a documentation problem here, if nothing else.

-- Erik

    
    > mgkk01> hostname
    > mgkk01.bc
    > mgkk01> grep mgkk01 /etc/hosts
    > 45.0.16.26 mgkk01
    > mgkk01> nslookup mgkk01
    > *** Can't find server name for address 45.0.16.25: Timed out
    > *** Default servers are not available
    > mgkk01> nslookup
    > *** Can't find server name for address 45.0.16.25: Timed out
    > *** Default servers are not available
    
    As you see there was a mismatch between 'hostname' and the contents
    of /etc/hosts.
    This mismatch was also the cause that
    "/etc/ifconfig ln0 `/bin/hostname` broadcast 45.0.31.255 netmask
    255.255.240.0"
    could not be executed in the /etc/rc.local file
    
    > mgkk01> /etc/ifconfig ln0 `/bin/hostname` broadcast 45.0.31.255
    netmask
    > 255.255.
    > 240.0
    > mgkk01.bc: bad value
    
    Once this was rectified, DECmcc could be executed without any problem.
    
    There was no need to delete
    /usr/mcc/mcc_system/mcc_dispatch_table.dat
    
            Thank you Erik,
            Dominique.

    I thought this was documented or release-noted a long time ago.
    
    tcp/ip must be up and running before DECmcc/Ultrix will run.
    
    /etc/ping `hostname` must return success.