Hello,
Does anybody knows what is the maximus number of nodes and interfaces that can be monitored by Orion?
Does de limit depends on the size of the database or some other external component?
Regards!
The limit is on a per server basis. Its the total number of elements. After 12000 elements you need to think about adding another polling engine. Total elements is the total number of nodes + interfaces + volumes.
Assuming you have an unlimited license, Orion won't stop if you go over 12k elements, but polling will slow down a little on all the elements in order to keep up. The DB is something else that needs to be maintained to keep Orion working at an acceptable level.
Zak Kahl
Loop1 Systems.
www.loop1systems.com
I can add to this by saying I have recently worked in a very massive environment that took advantage of a large number of additional pollers to handle the extreme polling needs. As Zak, mentioned, the DB will have to have some impressive specs to perform adequately in such cases.
Sohail Bhamani
Loop1 Systems
I agree and disagree with SJA on this one. If you went with HP OV or Tivoli, you would first be paying a ridiculous amount of money up front and year over year for maintenance. You will also then be forced to procure professional services for not only the initial setup and config, but also for any on going changes and config until such time as you become knowledgeable enough to handle this your self. The learning curve is very steep.
SW Orion will definitely be cheaper and the learning curve isnt as extreme. It will require more hardware and licensing, but the long term cost projection will still be significantly lower than those "big boy" monitoring applications. The days of those apps ruling the massive monitoring world has been over for some time.
I agree that you can add additional pollers to extend your polling capacity.
Sohail Bhamani
Loop1 Systems
Well
HP OV is much more expensive and it NOT better den solarwinds.
IF talk about "system scale" you need to monitor SNMP and you have a MPLS network with 10,000 Router/switch + 40,000 interfaces.
NPM is getting better from 10.1 but you need to know "Ninja tricks" to monitor 50,000 elements.
:-)
I currently help maintain a system with 4 polling engines and we are around 30k elements (and growing). There are some hurdles especially since we upgraded from 10.4, but for the most part the system runs smoothly. The licensing for each additional polling engine is a little expensive. Like other said as long as you have a powerful SQL server you should be fine. (that seems to be the case with any orion product)
Strongly object to HP OV and similar. You will be wasting your time and money.
I have trailed Solarwinds but found it is not to be sufficient functionally.
With so many different Solarwinds products I have found it very hard to create the solution which acts like a single system.
Maintenance will also be a great overhead.
I have ended up writing my own NMS - Enigma NMS. http://netsas.com.au
Largest deployment to date is at Queensland Government where single server manages over 4000 nodes with over 200,000 interfaces a.k.a elements.
All l stats are 1 min and never rolled.
You can check out following links.
Enigma NMS Introduction | NETSAS
In Enigma you don't have to worry about maintaining the OS or Database.
All following features are built into the same and single product.
SolarWinds solutions are rooted in our deep connection to our user base in the THWACK® online community. More than 195,000 members are here to solve problems, share technology and best practices, and directly contribute to our product development process.