[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

2817.0. "%MCC-E-NORECORD during a creation of rule" by BEAGLE::LOUIS () Wed Apr 22 1992 09:12

mcc1.2.4 ultrix
    
    Hi,
    
    After a hang & crash of the station during an enable of multiples rules
    selected, we lost all rules and we can't create new rules using iconic
    map or fcl.
    
    The error message is the same with create domain rule or 
    create mcc 0 alarm rule (& the same using comparaison , occur...)
    
    
csh> manage
DECmcc (T1.2.4)
 
MCC> create mcc 0 alarm rule alarm_ricard 
    EXPRESSION = (change_of (node4 ricard line bna-0 data blocks sent,*,*))

MCC 0 ALARMS RULE alarm_ricard 
AT 1992-04-22-10:28:43.379+02:00Iinf

Software logic error detected
                      MCC Routine Error = %MCC-E-NORECORD,  specified 
    			entity definition record does not exist
MCC>exit

    Here follows the printenv to show environment variable  :
    
    Is there a need to define MCC_COMMON and MCC_SYSTEM ?
    
    
Script started on Wed Apr 22 10:45:37 1992
csh> printenv
HOME=/usr/users/mcc
SHELL=/bin/csh
TERM=vt300
00
USER=mcc
    PATH=/usr/users/mcc/bin:/usr/ucb:/bin:/usr/bin:/usr/local:/usr/new:/usr/hosts:.
DISPLAY=:0.0 
LANG=
TERMCAP=db|vt300|vt300-am|dec vt300:	:hs:	:es:	:ts=\E[1$}\E[;H\E[K:
  
MAIL=/usr/spool/mail/mcc
EXINIT=set redraw wm=8
MSG_LOCATION=/usr/mcc/mcc_system
MCC_MIR_LOCATION=/var/mcc
MCC_NS_LOCATION=/var/mcc
MCC_ALARMS_LOCATION=/var/mcc
NLSPATH=/usr/mcc/mcc_system/%N
MCC_DNS_SELECTION=MIR 
MCC_NODE_IDP=49::
MCC_RPTS_LOCATION=/usr/mcc/reports
MCC_TIME_IN_MODE=2
MCC_TIME_OUT_MODE=2
MCC_TIME_ABS_TRUNC=79
MCC_MAPS=/usr/mcc/maps
MCC_ICONS=/usr/mcc/icons
II_SYSTEM=1
csh> pwd
    
    
    
/usr/users/mcc
csh> cd /var/mcc
    
	We tried to rename the alarms MIR files (.pag and .dir
    	but not the target MIR files), and reenrolled the mcc_alarms_fm
    	under root, but the symptom described persists.
    
    	Here is the MIR files contents after having reenrolled.
    
csh> ls -l *MIR*.*
-rw-r--r--  1 root            0 Apr 22 10:12 MCC_ALARMS_ATTRIBUTE_MIR.dir
-rw-rw-rw-  1 root         4096 Apr 16 08:50 MCC_ALARMS_ATTRIBUTE_MIR.odir
-rw-rw-rw-  1 root        29696 Apr 17 15:11 MCC_ALARMS_ATTRIBUTE_MIR.opag
-rw-r--r--  1 root         1024 Apr 22 10:12 MCC_ALARMS_ATTRIBUTE_MIR.pag
-rw-r--r--  1 root            0 Apr 22 10:12 MCC_ALARMS_INSTANCE_MIR.dir
-rw-rw-rw-  1 root         4096 Apr 14 18:27 MCC_ALARMS_INSTANCE_MIR.odir
-rw-rw-rw-  1 root         5120 Apr 17 15:11 MCC_ALARMS_INSTANCE_MIR.opag
-rw-r--r--  1 root         1024 Apr 22 10:12 MCC_ALARMS_INSTANCE_MIR.pag
-rw-rw-rw-  1 root            0 Feb 12 11:29 MCC_TARGET_DATA_MIR.dir
-rw-rw-rw-  1 root         1024 Feb 12 11:29 MCC_TARGET_DATA_MIR.pag
-rw-rw-rw-  1 root            0 Feb 12 11:29 MCC_TARGET_INST_MIR.dir
-rw-rw-rw-  1 root         1024 Feb 12 11:29 MCC_TARGET_INST_MIR.pag


    
	
    We have a demonstration for a customer tomorrow, and we have
    choose a station Ultrix because we don't have any other solution.
    I don't know a lot Ultrix, so perhapps the solution of this problem
    is easy , and i forget to check something. 
    So It is very important for us to have a solution quickly.
    
    Thank's a lot for any suggestions
    
    
    yves 
    
T.RTitleUserPersonal
Name
DateLines
2817.1Try new master repositoriesTOOK::MINTZErik Mintz, DECmcc Development, dtn 226-5033Wed Apr 22 1992 11:3918
Try removing the files /var/mcc/mcc_mir_repo_table.{dir,pag}
These are the master repository table, which point to other
repositories.  It is possible that these are the files which
became corrupt when the system crashed.


By the way, 

> II_SYSTEM=1

I don't know why this is defined this way.  For Ingres to work properly,
II_SYSTEM must point to a valid Ingres directory.  Since the DECmcc
installation procedure uses this variable to determine if Ingres is
installed, it may cause some confusion.  If Ingres is not installed,
the variable should not be defined.

-- Erik

2817.2is it dangerous ?BEAGLE::LOUISWed Apr 22 1992 12:4615
    Thank's Erik for your quick answer, but i am affraid about removing
    this files.
    I 've got near 50 domains and 200 entities with the local repository
    table. (i don't use DNS)
    So, what will append for these domains & entities  ( I can't lost
    them) if i remove this files ?
    
    For more information, a lot of others features, like create a domain, or
    add a node4 or an snmp entites or more work fine . I didn't test
    yet the report and export because it will take to much time.  
    
    
    	Thank's
    
    yves
2817.3Should be OKTOOK::MINTZErik Mintz, DECmcc Development, dtn 226-5033Wed Apr 22 1992 13:196
The master repository table will be re-created automatically.
You should not lose any of your other repository data.
But if you are concerned, rename it, so you can get it back if needed.

-- Erik

2817.4Now it's OK !...BEAGLE::LOUISWed Apr 22 1992 13:4511
    
    	Ok Erik,
    
    	I enroll the MCC_ALARMS_FM again after rename the files
    	/var/mcc/mcc_mir_repo_table.(dir/pag) and now the
    	create of rule works fine.
    	All other entities & domain look like ok
    
    	Thank's again for your quick solution
    
    	yves