What a good idea - using sysContact!
And keep in mind that Netview 5.1 gives you the nvdbimport
command for setting field values. Since the information has to
be administered external to the devices anyway, you can have
a tabular file with the field information and run nvdbimport on a
regular basis to set/update that field or any others.
Cordially,
Leslie A. Clark
IBM Global Services - Systems Mgmt & Networking
(248) 552-4968 Voicemail, Fax, Pager
---------------------- Forwarded by Leslie Clark/Southfield/IBM on 05/07/99
08:10 AM ---------------------------
Steve Francis <steve.francis@COMMSERV.UCSB.EDU> on 05/05/99 06:57:41 PM
Please respond to Discussion of IBM NetView and POLYCENTER Manager on NetView
<NV-L@UCSBVM.UCSB.EDU>
To: NV-L@UCSBVM.UCSB.EDU
cc: (bcc: Leslie Clark/Southfield/IBM)
Subject: Re: Availability for Ciscos/Integration with NV automated actions
I have implemented a kind of similar thing.
At my last job, various groups were responsible for various classes of devices:
one group of people for NT servers, one for Netware, one for Unix machines, one
for routers, etc.
I use the SNMP syscontact field, which every SNMP agent supports. Then when an
event occurs concerning a device that warrants a page or an email, the
management system runs a script or a rule (however you want to implement it)
that queries the database field for SNMP syscontact. (No use trying to query
the device, as the page may be being generated by the device being down.) If
the field contains the word "autocontact" then it is assumed that the rest of
the field specifies the name of a group.
These group names are aliases for email addresses and pager IDs, so the
appropriate people get contacted for the appropriate device. If the alias does
not exist, or the field does not contain autocontact, then a default group is
notified.
You can set the snmp sys contact via SNMP or in the IOS, with the command
snmp-server contact autocontact router
I also have the script that pages/emails on interface changes collect the local
interface description, which has english descriptions of the interface affected,
and other helpful information (like circuit ID on WAN circuit.s)
You can set this in IOS with the interface subcommand
description Interface to LA, Sprint Ckt ID XXXXXXXX
"Prokott, Joe" wrote:
> Thanks for all the suggestions, but this is not exactly what I am looking
> for.
>
> I am aware of configuring/retrieving RMON and MIB values, but I am trying to
> get notification info. specific to our environment into the Cisco by using
> Cisco commands (if such commands exist).
>
> I am trying to get into the router configuration values such as e-mail
> addresses or pager numbers, etc... directly into the Ciscos. The analyst
> would need to configure this themselves when they configure a new router. I
> am thinking that NetView could then query for this info and place it into
> its database for future automatic actions (i.e., scripts executed by
> NetView). The concept we are trying to achieve here is creating a process
> and distributing the management of specific devices back to the analyst that
> is configuring the device. Our network management group would define the
> process and it would be expected for the analysts to follow this process in
> order to assure availability and proper notification for outages or specific
> events.
>
> I do not think Cisco has such config. statements available in their IOS, but
> it would be nice if they did. Nonetheless, I am wondering if anyone else
> has tried to do something similar to this by using other Cisco
> user-definable MIB variables such as the "locIfDescr" MIB, etc...?
>
> Thanks in advance,
>
> Joe Prokott - West Group
> Network Architect
> 610 Opperman Drive
> St. Paul, MN 55123
> Phone: 651-687-4536
> Fax: 651-687-6946
> E-mail: joe.prokott@westgroup.com
>
> -----Original Message-----
> From: Jeff Fitzwater [mailto:jfitz@PRINCETON.EDU]
> Sent: Wednesday, May 05, 1999 10:47 AM
> To: NV-L@UCSBVM.UCSB.EDU
> Subject: Re: Availability for Ciscos/Integration with NV automated
> actions
>
> The only things you can set and monitor are rmon events and alarms,
> which you set up. You can fire a trap on alarm treshold. Its under the
> rmon command. The event fires the trap when a corresponding alarm is
> triggered.
> Havn't tried it yet. Not sure what the limits are on the mib variables
> you can test.
>
> Jeff Fitzwater
> Princeton University
>
> "Prokott, Joe" wrote:
> >
> > Has anyone configured a system that would allow a Cisco router analyst the
> > ability to configure availability times, notification methods, etc...
> > directly on the router and then have these propagated back to the NV
> > database on a regular interval for automated actions done on the
> management
> > system when certain conditions occur (interface down, interface link down
> > trap, etc...)? Does Cisco have any specific config. commands that allow
> > such a configuration to be defined on the router? Perhaps there are Cisco
> > integrated tools with NV that help do this (although I am not sure if
> Cisco
> > IOS supports such config. statements to allow for this?)?
> >
> > We have a mechanism in place today that does the notification, but it does
> > not take into account any specific configuration defined by analysts on
> the
> > router interfaces themselves. Rather, it just uses the NV collections to
> > determine notification. It has been requested by analysts configuring the
> > routers here to allow them the ability to configure some of these
> parameters
> > on a per interface level and then send this info. back to the NV database
> > for storage and retrieval when particular condtions arise.
> >
> > In doing this, it would offload some of the management functions to the
> > analysts configuring the Ciscos and define a more robust process used for
> > the management of these devices.
> >
> > Thanks,
> >
> > Joe Prokott - West Group
> > Network Architect
> > 610 Opperman Drive
> > St. Paul, MN 55123
> > Phone: 651-687-4536
> > Fax: 651-687-6946
> > E-mail: joe.prokott@westgroup.com
|