I am learning Network Performance Monitor (need help)

noobie

I am tring to monitor our networks servers and I want to know if it ever goes down for any reason. I have about 18 servers that I manage. I was told the NPM is a great tool to run as a service while i am out of the office. Is this true? Also, my main readon for posting here is to find out where I can learn the basics of NPM. I have purchased the engineer toolset. Basically I want to leave an application running to monitor all servers and if something goes down i want to be notified. I dont want to be bum rushed monday morning. Thanks in advance for your help.

erica.gill

Hi noobie,

Nice name! I think your question might be more for the Engineer's Toolset but you mention NPM and are in the NPM area so I'll try answer a bit for NPM. I'm going to do it in the order I'd have liked when starting fresh and getting to grips with it all.

For what you are saying and for myself alerting is one of the main features of any monitoring system. It was always handy to know in advance if Monday was going to be rough. If you can putting a bit of time into watching http://www.solarwinds.com/resources/webcasts/solarwinds-level-2-training-advanced-alert-manager.html, it will be well worth it, the webcast can give some ideas on what you can do with alerting if nothing else.

When looking for more information on absolutely anything in Orion, http://www.solarwinds.com/NetPerfMon/SolarWinds/default.htm#href=OrionAGTitlePage.htm is a great browsable, indexed and searchable resource.

Along with alerting, reporting is very important, (it is how we prove we are good at what we do!) http://www.solarwinds.com/documentation/Orion/docs/UnderstandingOrionReportWriter.pdf covers reporting in a lot of detail. It also covers the report scheduler and putting charts in reports, these can add some real polish to your reports. The default reports should cover most scenarios and the "Content Exchange" link at the top of the page will allow you to get a few more.

The customer portal is a very handy collection of all your software links and licenses, it also has links to all your current and previous support cases. What you might not know about is the Additional Components link on the left hand side where depenting on your products you find additional resources, like agents installers for LEM or STM. The Event Log Forwarder is a handy tool if you've access to it. Have a browse around, you might find something of interest.

What can be hard to give general advice on is organising your nodes and groups in Orion. This is very variable and the best people to organise these are the people who best know the environment and have to work with the devices and Orion day in day out. The best advice I can give here is to just make it easy on yourelf at first. Then tweak things as you get used to it and when you have a better feel for things have a look at http://www.solarwinds.com/documentation/Orion/docs/Groupsanddependencies.pdf and see how you can fit Orion around your needs.

Of course time is often a luxury and if you are planning groups and dependencies in advance some general advice would be to remember groups in Orion are logical groupings and like AD domains and OU designs you aren't tied to physical or geopraphical constraints.

The Manage Nodes page is very powerful and through it you can bulk view, sort and edit your nodes. Often times I finding myself planning to run a SQL query to make changes and when I stop and think I realise the Manage Nodes page will do it more easily. The two key things are the drop down on the left, where you can sort by Vendor, status, SNMP version/ICMP nodes, Polling Engine, etc. Then hidden away on the very right of the menu bar is two greatr then symbols ">>". Clicking on this will allow you to adjust the columns displayed and to sort based on thoe columns. There is also a little drop down in the middle of the page, just above the menu bar that you can use to change the Manage Nodes page into a "Manage Interfaces Page".

To come back to alerting a little, when troubleshooting alerts with customers we don't always have test nodes we can bring up or down, what we often do instead is in the web console adjust the nodes IP address to an IP address which isn't reachable in the environment and we can from Orion's perspective take a node down without impacting the production environment or the device.

In the Advanced Alert Manager there are a few little things you can do to make life easier. When you first open the alert manager rather then clicking straight away on "Configure Alerts", try clicking on "Active Alerts" to see all your currently active alerts. This "Active Alerts" view also has the ability to turn of alert actions, just ticking the "Disable All Actions" button will stop all alert emails being sent out until it is unticked again. This can be very handy when undertaking planned work and the "Unmanage Scheduling Utility" doesn't cover all you need.

Often in alerts you may only have an email action, this can be fine for most scenarios but it can make it hard to track false alerts, delayed emails and so on. Having a "Log Alert to NetPerfMon Log" action in your alerts can make later tracking of alerts easier.

From the alerts video, people are can get familiar with using alert macros, copying alert actions but I've found that having the exact same alert reset action as the alert trigger action can make telling the reset action from the trigger action a little tricky. For tracking and easy differentiation I often add a full stop at the end of the subject or some other punctuation mark someplace in the reset action's text.

Two programs I find can be under used are the Syslog Viewer and the Trap Viewer. Some customer use these heavily but others just use the web site to review syslogs and traps. If you've never used these before, open them and have a look, you may find the alerting options there of interest. If you've the Event Log Forwarder then you can tie this in with the Syslog Viewer's alerting.

Hope this helps some.

All the best,

Erica