nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

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
Store

Polling and Statistics

dan-uk

I Am trying to set up orion version 7.8.5 for the first time.

Without wanting to appear lazy, I've read through many posts on here and I still wanted to confirm that I understand what the polling and statistics settings actually relate to, and how they affect the graphs.

So... the 'polling' tab controls how often orion checks that the device is still up, and collects packet loss info etc, which it stores in the sql database and uses to plot the graphs, I take it that the more often you poll a device the more likely you'll spot issues OR does it, just take stats produced internally by the switch/router. In other words what does increasing/decreasing the polling frequency actually do ?

Same question for the Statistics tab, I take it that it collects stats as a snapshot of what the interface had at that moment or is it just collecting stats from the router which the router itself has complied? I read in a previous post that cisco devices produce internal stats every 5 mins so setting the 'statistics collection' to less than 5 mins is pointless. In other words whats does increasing/decreasing the statistical collection frequency actually do in terms of graphs. How does this differ to running a bandwidth guage from the Engineers toolset and collecting stats every 3 secs. ?

One other question was about the Detail + and Detail - buttons, these seems to affect the bandwidth utilisation quite dramatically on some graphs, e.g, the default detail shows 6Mbps throughput over a short period, and then increasing detail ramps it up to 8Mbps.. Probably a simple answer, is this just showing a peak to 8Mbps at some point, a point which was averaged out in the graph with less detail ?

Apologies for the newbie questions, any advice most welcome.

Regards

Dan

Find more posts tagged with

Accepted answers

All comments

dan-uk

Hi, Can anyone help me here as i'm still having a issue understanding what creates the utilisation charts, Statistics collection or polling.

Gustelj

Hi,

perhaps I can add my prespective of understanding the Orion terms "polling" and "statistics collection". Orion in general gathers data from managed devices through protocols ICMP and SNMP. Message types can be echo (ping) for ICMP and GET (or get next or get bulk) for SNMP. The term "polling" refers to sending an echo request to managed device (thus the response time and % packet loss is measured/calculated). The "statistics collection" refers to retrieving (getting) the data stored in MIB variables in the managed device by using the SNMP (e. g. free memory, CPU load, network traffic characteristics etc.).

Maybe someone with more deep insight in Orion operation could correct or add something.

FormerMember

I take it that the more often you poll a device the more likely you'll spot issues OR does it, just take stats produced internally by the switch/router. In other words what does increasing/decreasing the polling frequency actually do ?

ICMP polling is a regular ping. If we get no reply on a ping, we go into a fast polling mode and ping it a number of times to see if it's really down. In that case, polling more frequently will identify a box that's become unavailable more quickly; however, we poll every 2 minutes by default, which is frequent enough for most people.

Wrt to statistics, we collect a counter value from each device. In that case, polling more frequently won't give you better data. It's just like the odometer in your care. If you're trying to calculate miles per hour with only the odometer, looking at the counter more frequently won't improve the accuracy of your calculation. Checking more frequently will be more likely to pick up short-lived deviations.

dan-uk

Thanks for your responses everyone, so just to confirm, with regard the Statistics collection, the counter you are collecting is the 5 min input / output rates displayed using a "SH INT" command or is this an internal counter not available to view from the CLI on the router ? Basically I wanted to know what MIB variable is being read that makes up the plot points on the graphs. I'm currently reviewing Orion as we are looking possibly to replace HP Openview and Ciscoworks. Thanks again for your help

Dan

Digeridoo

Hi,

After much reading and clicking I am still trying to find confirmation of the default i.e. as newly installed...

"default poll interval for nodes" (& interfaces) in the 'Orion Net. Perf. Mon. settings'

Our Orion servers CPU gets hammered regularly and am sure polling time is a reason, but want to check the default before adjusting.

Thanks

FormerMember

Default status polling for nodes and interfaces is 120 sec.

Default statistics polling for nodes is 10 min.

Default statistics polling for interfaces is 9 min.

borgan

The polling and stats collection intervals can be adjusted on the Settings button in the toolbar of System Manager. Look on the Polling and Statistics tabs. These are global settings. Both of these intervals can also be set on any individual Node, Interface or Volume on the respective Details screen. Hope this helps.

Digeridoo

Thanks for quick reply Denny

Gustelj

Thanks for your responses everyone, so just to confirm, with regard the Statistics collection, the counter you are collecting is the 5 min input / output rates displayed using a "SH INT" command or is this an internal counter not available to view from the CLI on the router ? Basically I wanted to know what MIB variable is being read that makes up the plot points on the graphs. I'm currently reviewing Orion as we are looking possibly to replace HP Openview and Ciscoworks. Thanks again for your help
Dan

The values from sh int and in MIB variables should be the same, providing the same time frame and that there is no bug. Which MIB variable is actually used by Orion I don't know. Possible MIB variables are from obsoleted RFC2233 and RFC2863, which are implemented in new Cisco IOSes (IF-MIB, OLD-CISCO-INTERFACES-MIB ...).

torrielw

We have adjusted our polling frequencies on all of devices based on device criticality. We are experiencing these manually adjusted frequencies being changed by the system is there a remedy to this?