| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 231.1 | Is it reproducable? | BARREL::LEMMON | | Fri Aug 03 1990 15:05 | 25 |
|
The Iconic Map reads the NODE4 subclass information from the dictionary
when the user double-clicks into the first node4 entity. Once this read in,
the double-click code checks the memory resident class information to see
if NODE4 has any children. If there aren't any, it displays the error message
"...no child entities..."
It sounds like the memory resident NODE4 class information was corrupted.
Exactly how is unclear to me. This type of corruption may also explain why
the map datafiles gets bogus code values since the class information is
written to the file (see note 220).
I need more info about your environment...
You mentioned you are running IFT V1.1. The most recent kit is version X1.1.
Do you remember the sequence of operations you performed before the accvio?
Is your namespace distributed? (i.e., you have more than one clearing house for
the same namespace?)
Is the bug reproducable?
/Jim
|
| 231.2 | | CCIIS1::ROGGEBAND | _ �hili��e _ | Mon Aug 06 1990 05:20 | 34 |
|
Jim, here is more Info, though there isn't much available.
�The Iconic Map reads the NODE4 subclass information from the dictionary
�when the user double-clicks into the first node4 entity. Once this read in,
�the double-click code checks the memory resident class information to see
�if NODE4 has any children. If there aren't any, it displays the error message
�"...no child entities..."
In this case (as in most), NODE4 does have child entities
�You mentioned you are running IFT V1.1. The most recent kit is version X1.1.
The kit I installed is X1.1 (The most recent).
�Do you remember the sequence of operations you performed before the accvio?
No, I was "playing around". I have a hierarchy of three domains,
The third level has the nodes on it.
�Is your namespace distributed? (i.e., you have more than one clearing house for
�the same namespace?)
No, I have only one server and one clearinghouse. DECmcc is running
on the french ACT network which is not part of Easynet.
�Is the bug reproducable?
I haven't been able to reproduce it so far, I'll let you know if it
comes back.
Cordialement,
Philippe.
|
| 231.3 | It's back... | CCIIS1::ROGGEBAND | _ �hili��e _ | Mon Aug 20 1990 04:45 | 14 |
|
The bug (or a different ont) has just re-appeared, but in a somewhat
different fashion. I cannot fire up the Iconic Map PM, when I enter:
$ MANAGE/ENTER/INTERFACE=DECWINDOW
%MCC-F-PM_BUG, A PM bug has been unearthed
That's all I get. It worked ok on friday night before I left, and
unless a virus got into my station, I haven't got a clue of what
might be causing this.
As a temptative cure, I will re-install the kit.
Philippe.
|
| 231.4 | Were the logicals mcc_icons,mcc_maps,mcc_uil set? | TOOK::F_MESSINGER | | Mon Aug 20 1990 09:13 | 7 |
|
Just as a first stab, was the display set properly?
e.g. set display/node=yournodename/cre
Fred
|
| 231.5 | yes | CCIIS1::ROGGEBAND | _ �hili��e _ | Mon Aug 20 1990 09:34 | 16 |
| Some progress has been made :
I can now fire up from the system account, but I'm back to the
situation described in the basenote when I try to access one of
my domains, so I guess my DNS setup has somehow got fouled up. I have
tried to deregister all entities and delete my domains, but I'm
getting messages such as "Invalid entity structure" or "No such
entity" when I try to clear up the contents of my domains.
The problem is that DNS$CONTROL isn't that much help tracing MCC
directory contents... (I know, that's a DNS group concern. Hey,
when will we be able to manage DNS with MCC ?)
I'll let you know if I come up with anything.
Philippe.
|