Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials

How many elements do you monitor?

I was just curious on how many elements other companies monitor with their Orion server? I guess I'm wanting to know the limit of the SLX version before I should go to a 2nd poller. I'm running at 2500 elements. The site says between 6000-10000. Just wanted to know what everyone else has seen with larger installs then mine.

Thanks.

Find more posts tagged with

Accepted answers

All comments

pancamo

Network Elements 1411 Nodes, 4280 Interfaces, 2080 Volumes, 7771 Total Elements

We are using (1) 4-way 3Ghz, 4Gb Mem box poller/sqlserver. The bottle neck appears to be the local disks which we are moving to a SAN soon.

dp

quote:Originally posted by BryanBecker

I was just curious on how many elements other companies monitor with their Orion server? I guess I'm wanting to know the limit of the SLX version before I should go to a 2nd poller. I'm running at 2500 elements. The site says between 6000-10000. Just wanted to know what everyone else has seen with larger installs then mine.

Thanks.

ftsang

We currently have 8000+ elements being monitored in our environment. We have 2 polling engines running on 3.2 GHZ servers with 4 GB of memory on each server. We don't have any performance issues with Orion.

It's hard to gauge how many elements each polling engine can handle because another factor to consider is how often you plan to poll them.

www.solarwinds.net/.../Scalability_Options.htm

The info that solarwinds has is based on 9 minute stats collection. So if you're polling more frequently, the chart needs to be adjusted accordingly. we're doing 5 minutes in our environment.

pancamo

What is your Polls/Sec for status and Stats?

Can you post your database setting?

Here is ours, and we are missing polls which causes gaps in our graphs:

Database Details

Web Server USHOUSOLWNDS01
Operating System Windows 2000 Advanced Server
OS Version 5.0.2195
Service Pack 4.0

Database Server ushousolwnds01
Catalog Solarwinds2
Database Engine Microsoft SQL Server
SQL Version Microsoft SQL Server 2000 - 8.00.760 (Intel X86)
Dec 17 2002 14:22:05
Copyright (c) 1988-2003 Microsoft Corporation
Standard Edition on Windows NT 5.0 (Build 2195: Service Pack 4)

Security Model User Name \ Password
User Name SolarWindsNPM
Connection Timeout 20 seconds

Network Elements 8319 Elements
Nodes 1502 Nodes
Interfaces 4580 Interfaces
Volumes 2237 Volumes

Alerts 33 Alerts
Events 120461 Events
Pollers 13873 Pollers
Polling Engines 1 Polling Engines

Oldest Event 22-May-04 12:00 AM
Oldest Daily Traffic Stat 07-Jan-04 12:00 AM
Oldest Hourly Traffic Stat 22-May-04 12:00 AM
Oldest Detailed Traffic Stat 21-Jul-04 12:00 AM

Last Database Archive 18-Nov-04 01:15 AM
Next Scheduled Database Archive 19-Nov-04 01:15 AM
Database Archive Time 01:15 AM

Retain Detail Stats 120 Days
Retain Hourly Stats 180 Days
Retain Daily Stats 5000 Days
Retain Events 180 Days

quote:Originally posted by ftsang

We currently have 8000+ elements being monitored in our environment. We have 2 polling engines running on 3.2 GHZ servers with 4 GB of memory on each server. We don't have any performance issues with Orion.

It's hard to gauge how many elements each polling engine can handle because another factor to consider is how often you plan to poll them.

www.solarwinds.net/.../Scalability_Options.htm

The info that solarwinds has is based on 9 minute stats collection. So if you're polling more frequently, the chart needs to be adjusted accordingly. we're doing 5 minutes in our environment.

ftsang

we have 2 polling engines the polls/sec for status is 78 and stats is set to 65.

We do 5 minute stat collections but we do get some missed polls. If we bring up 5 minute interval graphs, we'll see some gaps but 10 minute interval graphs show up fine. Solarwinds support has never had a real answer to the issue of missed polls other than to use the polls per second tuning tool, which we do use.

Here's my stats:

Database Details

Web Server PRDB-EM04
Operating System Windows 2003 Standard Edition
OS Version 5.2.3790
Service Pack 0.0

Database Server PRDA-DB99
Catalog NetPerfMon
Database Engine Microsoft SQL Server
SQL Version Microsoft SQL Server 2000 - 8.00.760 (Intel X86)
Dec 17 2002 14:22:05
Copyright (c) 1988-2003 Microsoft Corporation
Standard Edition on Windows NT 5.2 (Build 3790: )

Security Model User Name \ Password
User Name SolarWindsNPM
Connection Timeout 20 seconds

Network Elements 8411 Elements
Nodes 755 Nodes
Interfaces 7656 Interfaces
Volumes 0 Volumes

Alerts 24 Alerts
Events 13579 Events
Pollers 16476 Pollers
Polling Engines 2 Polling Engines

Oldest Event 20-Oct-04 12:00 AM
Oldest Daily Traffic Stat 02-Apr-04 12:00 AM
Oldest Hourly Traffic Stat 20-Oct-04 12:00 AM
Oldest Detailed Traffic Stat 12-Nov-04 12:00 AM

Last Database Archive 19-Nov-04 02:16 AM
Next Scheduled Database Archive 20-Nov-04 02:15 AM
Database Archive Time 02:15 AM

Retain Detail Stats 7 Days
Retain Hourly Stats 30 Days
Retain Daily Stats 365 Days
Retain Events 30 Days

BryanBecker

Dan,

I see you keep your stats alot longer then the norm...like daily at 120 days instead of the default 7. Have you experience any issues with your server or database when keeping this much detailed info? Especially when your have 4 times the elements then I have.

I would like to keep longer detailed and hourly stats but I do not want to sacrifice the perfromance of my server.

B

pancamo

As long as you can keep the netperfmon service from peeking to 100%, you'll be ok. If the service goes to 100%, we miss polls.

We finally have set our Polls to 50/sec and status polls to 100/sec and we are ok for now.

i'm not sure if setting the daily to 120 will make a difference in anything other than DB size.

dp

quote:Originally posted by BryanBecker

Dan,

I see you keep your stats alot longer then the norm...like daily at 120 days instead of the default 7. Have you experience any issues with your server or database when keeping this much detailed info? Especially when your have 4 times the elements then I have.

I would like to keep longer detailed and hourly stats but I do not want to sacrifice the perfromance of my server.

B

mcobb

I am running 2 polling servers(Compaq DL380 G4)with 25,265 elements placed accross the 2 pollers. They both report to a SQL server (DL580, Quad P4 3.2 8gig RAM). All are on Gig over ethernet. I have never had a problem with the program as far as expansion. I have it set to poll all interfaces every 30 seconds and all nodes every 60 seconds. The program will only be as good as the servers that they run on...

Mithrilhall

13,000 + elements (about 500+ volumes) on a single polling engine. I actually have Orion and MS SQL on the same box (I'm hoping to fix that soon and get another polling engine).

pblatz

Polling 7400 elements on a single Orion Server with DB on a seperate system.

Using the Poller Tuner calculation I've adjusted the registry settings for 75 ICMP/sec and 100 SNMP/sec.

The dual proc Orion system runs at a constant 40% CPU utilization.

Expansion goal is to have dedicated Web host and second poller when we reach 10000 elements.
Also looking at changing retension of stats.

Biggest issue I have is Web response.

Network Elements 7381 Elements
Nodes 458 Nodes
Interfaces 6905 Interfaces
Volumes 18 Volumes

Alerts 14 Alerts
Events 29624 Events
Pollers 14329 Pollers
Polling Engines 1 Polling Engines

Oldest Event 22-Feb-06 12:00 AM
Oldest Daily Traffic Stat 16-Jan-06 12:00 AM
Oldest Hourly Traffic Stat 22-Feb-06 12:00 AM
Oldest Detailed Traffic Stat 17-Mar-06 12:00 AM

Last Database Archive 24-Mar-06 02:15 AM
Next Scheduled Database Archive 25-Mar-06 02:15 AM
Database Archive Time 02:15 AM

Retain Detail Stats 7 Days
Retain Hourly Stats 30 Days
Retain Daily Stats 365 Days
Retain Events 30 Days

Thanks,
P. B.

Success requires three bones--wishbone, backbone and funnybone. Kobi Yamada

jtimes

quote:Originally posted by pblatz

Biggest issue I have is Web response.

I have found that removing the All Nodes View from the Network Summary page (Home) speeds up the Web response significantly. I also created a custom view called Regions with Node List in column 1, the Search Nodes and Search Interfaces in column 2, then placed it on the Menu bar and set it to open in a new window.

Back to the Topic:

Originally, we had one Orion server, as soon as we got to around 7k to 8k elements we started dropping polls (thats about the time we went to full blown Orion setup)

4 Polling Engines
1 Hot Stand-By
2 Clustered SQL servers
80+ clients concurrently logged in

Network Elements 15491 Elements
Nodes 6107 Nodes
Interfaces 9231 Interfaces
Volumes 153 Volumes

Alerts 23 Alerts
Events 94390 Events
Pollers 26022 Pollers

Retain Detail Stats 30 Days
Retain Hourly Stats 90 Days
Retain Daily Stats 366 Days
Retain Events 366 Days

The database is around 15GB

Default Poll Interval for Nodes 240 Seconds
Default Poll Interval for Interfaces 240 Seconds
Default Poll Interval for Volumes 300 Seconds
Rediscover Resources 60 Minutes

John J. Times

Quick Links

All Categories
Recent Posts
Activity
Unanswered
Groups
Help
Best Of