| Title: | FDDI - The Next Generation |
| Moderator: | NETCAD::STEFANI |
| Created: | Thu Apr 27 1989 |
| Last Modified: | Thu Jun 05 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 2259 |
| Total number of notes: | 8590 |
Hi all
I am trying to set up some alarms in MCC to detect "usual" error
conditions in a customer's network. With those alarms, operators
should be able to detect failures and to report them, either to the
network manager or to DEC.
The network is quite simple, two buildings connected by a double ring,
a DECconcentrator 900 in each building, and some VAXes and DECbridge900
connected to each concentrator.
I have installed RFC1512 and DEC ELAN MIBS, and I'm not sure of which
variables to poll. My first choice was:
fddimibSMTECMState
fddimibSMTStationStatus
fddimibMACRMTState
fdimibPORTConnectState
Is this enough? Are there any more status variables that I should poll?
The description of the possible values in the MIB document is quite
cryptic. Can anybody comment a little bit on the meaning of each one,
or give a pointer to any document that I could give to the operators ?
Thanks a lot in advance for any help
Luis Gonzalez Iglesias - Systems Integration Madrid
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 1471.1 | ISIDRO::LUISGON | Mon Oct 17 1994 12:03 | 2 | ||
Anybody out there?
| |||||
| 1471.2 | More specifics needed to be helpful | 56821::B_CRONIN | Tue Oct 18 1994 16:29 | 16 | |
The variables you have mentioned are very low level in the FDDI.
I suspect that you would prefer more high level info, but I don't
want to type it in until I am sure what you would like!
Could you dexribe the types of problems that the user will want to
see, and then we can spend some time on the specifics. For instance,
I assume that you will want to know if the ring wraps, or if a
port goes down. Do you have any other idea of what things you'd
like to see?
If you really want to go done into the bowels of the FDDI we can
do that, but my suspicion is that you don't need or want that much
detail.
| |||||
| 1471.3 | ISIDRO::LUISGON | Tue Oct 18 1994 18:57 | 33 | ||
Well, in this case the customer has few, or I would say none, specific
requirements. The questions are:
- Which are the most usual and important problems to monitor (you
know, those 5% that happen 95% of the times) ?
- How do they map to MIB variables ?
My suggestions (probably incomplete):
- The ring is not working at all or there is a loss of connectivity
between concentrators
- There is connectivity in the ring but a potentially serious failure:
- The ring wraps
- ???
- One of the nodes, or a bridge connected to a concentrator port has
lost connectivity
- One of the nodes, or a bridge connected to a concentrator port is not
working properly, or has a bad comms performance
- ???
All they want is to be sure that the operators will NOTICE that
something is wrong, as quickly as possible, and in the simplest
possible way, so they can report it somebody knowledgeable.
As an example, for DECnet nodes they are alarming on loss of
connectivity, send failures in the ethernet lines, area connectivity
and changes of state of WAN circuit and lines.
Regards
Luis
| |||||