Best practices questions

15 messages Options
Embed this post
Permalink
Norm Shultry

Best practices questions

Reply Threaded More More options
Print post
Permalink
Hello,

I'm considering using OpenNMS for my company which is composed of several divisions that I would like to organize, somehow.

Does OpenNMS have any facility for grouping systems into "divisions" or some other partitioning scheme?

Does OpenNMS support an environment that heavily utilizes DHCP for assigning IP addresses?

Thanks,
Norm

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
Alexander Finger

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
Hi Norm,


I'm considering using OpenNMS for my company which is composed of several divisions that I would like to organize, somehow.

OpenNMS has several ways of separating Nodes into "Categories". Which way works best for you depends on circumstances.

One way to create categories is based on IP-Addresses, if that's feasible. These are the "SLA Categories".

Another way to group Nodes logically are the Surveillance Categories. A node can be member of no, one or many Surveillance Categories. It's fairly easy to use those later on in the notifications, allowing you to bother the "right" people with the "right" information.

In the next release of OpenNMS you will have the option to present the nodes as well to people grouped by Surveillance Categories.

 

Does OpenNMS have any facility for grouping systems into "divisions" or some other partitioning scheme?

Does OpenNMS support an environment that heavily utilizes DHCP for assigning IP addresses?


We depend very much on the IP of a Node. If Nodes swap IP it's a tough call. In general, if speaking of servers and routers and so on, you won't change their IPs.
For PC-Networks you will want to think about what you want to monitor, down to the point of considering to ignore the "popular" used IPs.

Again, what I say might be entirely wrong in your network :-)

Hope that helps, nonetheless

Alex

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
Norm Shultry

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
Hi Alex, thanks very much for the points -- they do help!

As far as categorizing goes, I'll check out the Surveillance Categories as that sounds like the best fit at the moment.

With respect to the dynamic IP address, I do understand the issue involved in getting a unique key for each node (and I've seen this with the two other packages I've looked at). From a design standpoint, though, why don't NMS systems use the MAC address as the unique identifier?

Norm

On Mon, Nov 2, 2009 at 1:57 PM, Alexander Finger <[hidden email]> wrote:
Hi Norm,


I'm considering using OpenNMS for my company which is composed of several divisions that I would like to organize, somehow.

OpenNMS has several ways of separating Nodes into "Categories". Which way works best for you depends on circumstances.

One way to create categories is based on IP-Addresses, if that's feasible. These are the "SLA Categories".

Another way to group Nodes logically are the Surveillance Categories. A node can be member of no, one or many Surveillance Categories. It's fairly easy to use those later on in the notifications, allowing you to bother the "right" people with the "right" information.

In the next release of OpenNMS you will have the option to present the nodes as well to people grouped by Surveillance Categories.

 

Does OpenNMS have any facility for grouping systems into "divisions" or some other partitioning scheme?

Does OpenNMS support an environment that heavily utilizes DHCP for assigning IP addresses?


We depend very much on the IP of a Node. If Nodes swap IP it's a tough call. In general, if speaking of servers and routers and so on, you won't change their IPs.
For PC-Networks you will want to think about what you want to monitor, down to the point of considering to ignore the "popular" used IPs.

Again, what I say might be entirely wrong in your network :-)

Hope that helps, nonetheless

Alex

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss


------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
Alexander Finger

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink

With respect to the dynamic IP address, I do understand the issue involved in getting a unique key for each node (and I've seen this with the two other packages I've looked at). From a design standpoint, though, why don't NMS systems use the MAC address as the unique identifier?


That would indeed be an option if you could always get it. Given you can snmp in your switch and on your hosts, that's not an issue.

As soon as you need to monitor a "remote" Network or can not access the ARP Tables on the switches, you won't be able to get the MAC Address of an interace though.

So the answer is that in IP-based networks there's always an IP address available to identify an interface using IP (Layer 3) - you might not be able to access the Layer 2 though.

Kind regards
Alex

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
Norm Shultry

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
Hi Alex,

Thanks very much for the explanation. I can see the issue now.

Best,
Gil

On Mon, Nov 2, 2009 at 2:43 PM, Alexander Finger <[hidden email]> wrote:

With respect to the dynamic IP address, I do understand the issue involved in getting a unique key for each node (and I've seen this with the two other packages I've looked at). From a design standpoint, though, why don't NMS systems use the MAC address as the unique identifier?


That would indeed be an option if you could always get it. Given you can snmp in your switch and on your hosts, that's not an issue.

As soon as you need to monitor a "remote" Network or can not access the ARP Tables on the switches, you won't be able to get the MAC Address of an interace though.

So the answer is that in IP-based networks there's always an IP address available to identify an interface using IP (Layer 3) - you might not be able to access the Layer 2 though.

Kind regards
Alex

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss


------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
douglasstvnsn

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
The other thing here is that a multi-homed host may have the same MAC across multiple interfaces.

Check out Sun systems... ;)

>From an L2 perspective, once you egress the local LAN, your IP assumes a MAC of a Router.

Spectrum does use an IP Address/ MAC Address combination as a unique key.

HTH,

Dougie!!!


-----Original Message-----
From: Norm Shultry <[hidden email]>
To: General OpenNMS Discussion <[hidden email]>
Sent: Tue, Nov 3, 2009 3:14 pm
Subject: Re: [opennms-discuss] Best practices questions

Hi Alex,

Thanks very much for the explanation. I can see the issue now.

Best,
Gil

On Mon, Nov 2, 2009 at 2:43 PM, Alexander Finger <[hidden email]> wrote:

With respect to the dynamic IP address, I do understand the issue involved in getting a unique key for each node (and I've seen this with the two other packages I've looked at). From a design standpoint, though, why don't NMS systems use the MAC address as the unique identifier?


That would indeed be an option if you could always get it. Given you can snmp in your switch and on your hosts, that's not an issue.

As soon as you need to monitor a "remote" Network or can not access the ARP Tables on the switches, you won't be able to get the MAC Address of an interace though.

So the answer is that in IP-based networks there's always an IP address available to identify an interface using IP (Layer 3) - you might not be able to access the Layer 2 though.

Kind regards
Alex

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this
page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
Les Mikesell

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
[hidden email] wrote:
>   The other thing here is that a multi-homed host may have the same MAC
> across multiple interfaces.
>
> Check out Sun systems... ;)
>
>  >From an L2 perspective, once you egress the local LAN, your IP assumes
> a MAC of a Router.
>
> Spectrum does use an IP Address/ MAC Address combination as a unique key.

How does that work when your routes change and the mac comes from
different router?

--
   Les Mikesell
    [hidden email]





------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
David Hustace

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
In reply to this post by Alexander Finger
Some javascript/style in this post has been disabled (why?)

On Nov 2, 2009, at 1:57 PM, Alexander Finger wrote:


In the next release of OpenNMS you will have the option to present the nodes as well to people grouped by Surveillance Categories.

Also, you may also provision nodes as groups.  These "groups" will become more and more prevalent as 1.8 matures.


We depend very much on the IP of a Node. If Nodes swap IP it's a tough call. In general, if speaking of servers and routers and so on, you won't change their IPs.
For PC-Networks you will want to think about what you want to monitor, down to the point of considering to ignore the "popular" used IPs.

If you are directly provisioning nodes, then you can change the IP and maintain the node's integrity.  Also, now in 1.8, there is the notion of DNS as a source for provisioning.  The node-label (i.e. hostname from the A record) is used as the key and the zone or zone filter is the source/group name.  Now, if the host IP address changes, then the next DNS import will change the IP address of the node.  So, if you are running dynamic DNS, changes to the IP address related to the node will be reflected in OpenNMS on the next import.



David Hustace
The OpenNMS Group, Inc.


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
David Hustace

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
In reply to this post by douglasstvnsn
Some javascript/style in this post has been disabled (why?)

On Nov 3, 2009, at 4:24 PM, [hidden email] wrote:

Spectrum does use an IP Address/ MAC Address combination as a unique key.

That's why spectrum rocks ;)  Actually, outside of some of the performance and maintenance related issues of Spectrum, it was a pretty cool tool.  Wasn't it Cabletron that owned the patent to every network management related concept in the universe!

-D 

David Hustace
The OpenNMS Group, Inc.


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
Michael Danko

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
Yes, it was. There was a point where Cabletron was such a big supplier  
of equipment (all layers) they had to take all sorts of steps to prove  
their technologies were interoperable with others.

Spectrum, back in the day, had some really awesome options for ISV's,  
say if you were had a product that was a systems/network controller,  
they had some pretty tight integration packages. No one but a very  
select few will ever have seen some of the things they deployed and  
how truly awesome it was (minus the overhead and whatnot you  
mentioned, we had to turn ours off)

This is actually one of the reasons I'm an OpenNMS zealot. While auto-
discovery through specifying IP ranges or through SNMP trap receipt is  
great and all, those features aren't really something that scales with  
day to day operations. With that said, the provisioning groups feature  
is something that provides so much functionality that may go  
overlooked -- it allows for things like provisioning and change  
management systems to drive your network as they should, rather than  
relying on hindsight when things move around. It's super enterprisey!

Gonna get wordy since we're talking best practices...

A MAC/IP pair will never be consistent enough to use it for something  
like a primary key on the node table given the scenario of an IP  
change. A MAC or IP would also turn up fail in the event of a hardware  
swap coupled with a re-ip'ing (happens ALL the time at the same time).  
The Asset Management system baked into ONMS could sort of do this, but  
it can get sort of confusing after changing out hardware later.

I'd reckon to say the only human understandable CONSISTENT identifier  
for any sort of equipment is knowing that it's the X Numbered Y Device  
at Site Z, in which X, Y, and Z will wildly vary by business type. You  
could, OpenNMS does, and many vendors ask for things like building  
number, floor number, rack number, and rack position, but I always  
feel completely ridiculous explaining that there's only one room, one  
floor, a few racks and fewer rack positions at my remote sites. When  
you're talking monitoring, things like port numbers, interfaces, etc.,  
come into play, but when something completely fails, it's the X  
Numbered, Y Device at Site Z.

If you want OpenNMS to track XYZ, it can, and do a fine job at that.  
But it can't, and never could come close to, make assumptions about  
whats in your head.

So, to:

1.) Reduce Human Error via input or field device change out/
misconfiguration
2.) Document Changes
3.) Properly arrange for ONMS to know your changes and when they're to  
happen
4.) Always have your eyeballs on the right device with little or no  
delay (see #3)
5.) Make sure your end users and/or customers are happy people

It's critical to have an awesome grasp on your inventory and change  
management procedures. The XML node import and Asset Management  
systems built into ONMS aren't as quick, impressive, or flashy as auto-
discovery or adding IP's manually, but it's the real deal as far as  
enterprise management.

I'm not currently using the Asset system, but I do currently use the  
XML importer. Our network people could probably get some use out of  
the AM component, but everything I'm monitoring is either one or two  
different units. The XML schema can be a little hard to follow for  
people who don't use it, but if you create one via the WebUI, it will  
drop it in opennms/etc for you. You define set physical locations,  
then the nodes within, the services monitored, etc., etc.

The magic, and if you only read two lines of this long rant, is this:

If you have a consistent value (external key) for your device set in  
the XML, and you make changes to any other data, OpenNMS will still  
equate that external key to your Node ID in the DB, and after clicking  
"import' after making changes, you have NOTHING to worry about. Keep  
your change management, ticketing, etc., properly maintained, and  
OpenNMS will handle it all for you.

I keep meaning to do something to make this easier for people, but it  
always falls through the cracks.

- Mike


On Nov 3, 2009, at 8:12 PM, David Hustace wrote:

>
> On Nov 3, 2009, at 4:24 PM, [hidden email] wrote:
>
>> Spectrum does use an IP Address/ MAC Address combination as a  
>> unique key.
>
> That's why spectrum rocks ;)  Actually, outside of some of the  
> performance and maintenance related issues of Spectrum, it was a  
> pretty cool tool.  Wasn't it Cabletron that owned the patent to  
> every network management related concept in the universe!
>
> -D
>
> David Hustace
> The OpenNMS Group, Inc.
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008  
> 30-Day
> trial. Simplify your report design, integration and deployment - and  
> focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july_______________________________________________
> Please read the OpenNMS Mailing List FAQ:
> http://www.opennms.org/index.php/Mailing_List_FAQ
>
> opennms-discuss mailing list
>
> To *unsubscribe* or change your subscription options, see the bottom  
> of this page:
> https://lists.sourceforge.net/lists/listinfo/opennms-discuss


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
Rob McKrill

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
In reply to this post by David Hustace
Some javascript/style in this post has been disabled (why?)
Re: [opennms-discuss] Best practices questions This may be off topic, but in reading through the “DNS Importing” wiki page and attempting to implement it in my installation I ran into an issue with discovery actually using IXFR instead of AXFR like the wiki page says to test.  Which way is this intended to work?  Should it work as described in the wiki page where it should grab the entire zone every time and update nodes as it sees changes or using incremental (IXFR) queries and changing only what ONMS’ current version of the zone file vs. the master DNS zone shows?

The reason I ask is because the name servers I query allow AXFR but not IXFR and the errors I am seeing is:

Discovery.Pinger] Discovery: Error reading URL: dns://mydnsserver/myzone: Problem getting input stream: org.xbill.DNS.ZoneTransferException: server doesn't support IXFR

Any suggestions?


Thank you,

Rob


On 11/3/09 7:56 PM, "David Hustace" <david@...> wrote:


On Nov 2, 2009, at 1:57 PM, Alexander Finger wrote:


In the next release of OpenNMS you will have the option to present the nodes as well to people grouped by Surveillance Categories.

Also, you may also provision nodes as groups.  These "groups" will become more and more prevalent as 1.8 matures.


We depend very much on the IP of a Node. If Nodes swap IP it's a tough call. In general, if speaking of servers and routers and so on, you won't change their IPs.
 For PC-Networks you will want to think about what you want to monitor, down to the point of considering to ignore the "popular" used IPs.

If you are directly provisioning nodes, then you can change the IP and maintain the node's integrity.  Also, now in 1.8, there is the notion of DNS as a source for provisioning.  The node-label (i.e. hostname from the A record) is used as the key and the zone or zone filter is the source/group name.  Now, if the host IP address changes, then the next DNS import will change the IP address of the node.  So, if you are running dynamic DNS, changes to the IP address related to the node will be reflected in OpenNMS on the next import.

http://www.opennms.org/wiki/DNS_Importing


 
David Hustace
The OpenNMS Group, Inc.



------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
Norm Shultry

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
In reply to this post by Alexander Finger
Hi,

I'm reviewing some of our requirements it looks like it would be easiest for me for keep all the divisions separate -- that is, running in separate OpenNMS instances. 

Is it possible to configure the installation this way (several OpenNMS instances on one physical server and have separate PostgreSQL databases for each division but running on the same PostgreSQL server?)

Thanks,
Norm

On Mon, Nov 2, 2009 at 1:57 PM, Alexander Finger <[hidden email]> wrote:
Hi Norm,


I'm considering using OpenNMS for my company which is composed of several divisions that I would like to organize, somehow.

OpenNMS has several ways of separating Nodes into "Categories". Which way works best for you depends on circumstances.

One way to create categories is based on IP-Addresses, if that's feasible. These are the "SLA Categories".

Another way to group Nodes logically are the Surveillance Categories. A node can be member of no, one or many Surveillance Categories. It's fairly easy to use those later on in the notifications, allowing you to bother the "right" people with the "right" information.

In the next release of OpenNMS you will have the option to present the nodes as well to people grouped by Surveillance Categories.

 

Does OpenNMS have any facility for grouping systems into "divisions" or some other partitioning scheme?

Does OpenNMS support an environment that heavily utilizes DHCP for assigning IP addresses?


We depend very much on the IP of a Node. If Nodes swap IP it's a tough call. In general, if speaking of servers and routers and so on, you won't change their IPs.
For PC-Networks you will want to think about what you want to monitor, down to the point of considering to ignore the "popular" used IPs.

Again, what I say might be entirely wrong in your network :-)

Hope that helps, nonetheless

Alex

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
John A. Sullivan III

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
My guess would be one would do it by editing opennms-datasources.xml
before running the installation routine - John

On Wed, 2009-11-04 at 14:39 -0500, Norm Shultry wrote:

> Hi,
>
>
> I'm reviewing some of our requirements it looks like it would be
> easiest for me for keep all the divisions separate -- that is, running
> in separate OpenNMS instances.
>
>
> Is it possible to configure the installation this way (several OpenNMS
> instances on one physical server and have separate PostgreSQL
> databases for each division but running on the same PostgreSQL
> server?)
>
>
> Thanks,
> Norm
>
> On Mon, Nov 2, 2009 at 1:57 PM, Alexander Finger
> <[hidden email]> wrote:
>         Hi Norm,
>        
>                
>                
>                 I'm considering using OpenNMS for my company which is
>                 composed of several divisions that I would like to
>                 organize, somehow.
>        
>         OpenNMS has several ways of separating Nodes into
>         "Categories". Which way works best for you depends on
>         circumstances.
>        
>         One way to create categories is based on IP-Addresses, if
>         that's feasible. These are the "SLA Categories".
>        
>         Another way to group Nodes logically are the Surveillance
>         Categories. A node can be member of no, one or many
>         Surveillance Categories. It's fairly easy to use those later
>         on in the notifications, allowing you to bother the "right"
>         people with the "right" information.
>        
>         In the next release of OpenNMS you will have the option to
>         present the nodes as well to people grouped by Surveillance
>         Categories.
>        
>          
>        
>                
>                
>                 Does OpenNMS have any facility for grouping systems
>                 into "divisions" or some other partitioning scheme?
>                
>                
>                 Does OpenNMS support an environment that heavily
>                 utilizes DHCP for assigning IP addresses?
>                
>                
>        
>         We depend very much on the IP of a Node. If Nodes swap IP it's
>         a tough call. In general, if speaking of servers and routers
>         and so on, you won't change their IPs.
>         For PC-Networks you will want to think about what you want to
>         monitor, down to the point of considering to ignore the
>         "popular" used IPs.
>        
>         Again, what I say might be entirely wrong in your network :-)
>        
>         Hope that helps, nonetheless
>        
>         Alex
>        
>        
>         ------------------------------------------------------------------------------
>         Come build with us! The BlackBerry(R) Developer Conference in
>         SF, CA
>         is the only developer event you need to attend this year.
>         Jumpstart your
>         developing skills, take BlackBerry mobile applications to
>         market and stay
>         ahead of the curve. Join us from November 9 - 12, 2009.
>         Register now!
>         http://p.sf.net/sfu/devconference
>         _______________________________________________
>         Please read the OpenNMS Mailing List FAQ:
>         http://www.opennms.org/index.php/Mailing_List_FAQ
>        
>         opennms-discuss mailing list
>        
>         To *unsubscribe* or change your subscription options, see the
>         bottom of this page:
>         https://lists.sourceforge.net/lists/listinfo/opennms-discuss
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________ Please read the OpenNMS Mailing List FAQ: http://www.opennms.org/index.php/Mailing_List_FAQ opennms-discuss mailing list To *unsubscribe* or change your subscription options, see the bottom of this page: https://lists.sourceforge.net/lists/listinfo/opennms-discuss
--
John A. Sullivan III
Open Source Development Corporation
+1 207-985-7880
[hidden email]

http://www.spiritualoutreach.com
Making Christianity intelligible to secular society


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
David Hustace

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
In reply to this post by Rob McKrill
Some javascript/style in this post has been disabled (why?)

On Nov 4, 2009, at 1:33 PM, Rob McKrill wrote:

Discovery.Pinger] Discovery: Error reading URL: dns://mydnsserver/myzone: Problem getting input stream: org.xbill.DNS.ZoneTransferException: server doesn't support IXFR

Any suggestions?

I think this thread is getting overloaded...

Rob, good observation, i just checked the code, and yes, I did it wrong.  Please open a bug and I'll get it fixed.


-D

David Hustace
The OpenNMS Group, Inc.


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss
Johan Edstrom

Re: Best practices questions

Reply Threaded More More options
Print post
Permalink
This thread looks like a few Wiki pages and a few bugzillas...

On Nov 4, 2009, at 6:32 PM, David Hustace wrote:

>
> On Nov 4, 2009, at 1:33 PM, Rob McKrill wrote:
>
>> Discovery.Pinger] Discovery: Error reading URL: dns://mydnsserver/myzone:
>>  Problem getting input stream: org.xbill.DNS.ZoneTransferException:  
>> server doesn't support IXFR
>>
>> Any suggestions?
>
> I think this thread is getting overloaded...
>
> Rob, good observation, i just checked the code, and yes, I did it  
> wrong.  Please open a bug and I'll get it fixed.
>
>
> -D
>
> David Hustace
> The OpenNMS Group, Inc.
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008  
> 30-Day
> trial. Simplify your report design, integration and deployment - and  
> focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july_______________________________________________
> Please read the OpenNMS Mailing List FAQ:
> http://www.opennms.org/index.php/Mailing_List_FAQ
>
> opennms-discuss mailing list
>
> To *unsubscribe* or change your subscription options, see the bottom  
> of this page:
> https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Johan Edstrom

[hidden email]

They that can give up essential liberty to purchase a little temporary  
safety, deserve neither liberty nor safety.

Benjamin Franklin, Historical Review of Pennsylvania, 1759






------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss