bscherer

Other than drinks after a big project on occasion, nothing... Am I a BOFH and didn't know it?

Mostly Solarwinds products, but a few nitch products too. OCI,Netbeez and i'm sure a handful of security tools I have no idea about.

When switching to the Orion VMAN polling, does historical data from the appliance move over also? 

Nothing too exciting, just dumb luck. I was taking some classes with someone that would end up being my future coworker and he recommended me for a position. I was young, cheap and not beaten down by the world yet. 

Read the release notes and it looks like their is a sync wizard. I'll RTFM next time before commenting

I assumed you didn't assume, so i am the worse assumer in this case.

I actually did both at the same time in order to limit downtime. I had new 2016 VMs ready to go, I just re-ip'd them after shutting down old servers and ran installers. *If you don't rename them to match old pollers, will need to update names in SQL.

Yeah you are correct, i threw this into SWQL studio and it doesnt work. I was going off of mesverrums query. CPULoad is not a column in the Orion.CPULoad table, you want to use the AvgLoad column. try this query. ReReading the thread i think you wanted additional info. You can modify this to convert values from bytes to GB…

That is what I usually do, or you could convert it to one line. I prefer to keep it multiline for readability I would do something like this, assuming powershell. $swis = connect-swis -swishost OrionServer -trusted # or use -username and -password if your account does not have permissions $swisQuery = " select n.caption,…

I am going to assume the volume definitely exists in windows. 1. SNMP Monitoring a. Is this a mount point? If so, you need to use WMI to monitor mount points b. Not a mount point, run snmpwalk to see if volume appears(in your Orion install directory) 2. WMI Monitoring a. From your polling engine, open powershell b. run the…

I did not have that issue, because the account with limitations are very uniform. With the exception of the nodes they can see. I am going to run into the same thing adding additional limitations for vman/srm in appstack.

You need SRM to get down to the LUN level

This is exactly how i do it and we have a lot of clusters here. Individual Nodes - Monitor Local resources Cluster IPs - Monitor Cluster resources. If you do any type of automated discovery, you will need to put something in place to do cleanup. Otherwise you will have cluster resources assigned to whatever the active node…

You just need to query the CPULoad view. Not much changes in the query. At least I think my sqlfu is correct here select n.caption, avg(cpu.avgload) as avgCPULoad,min(MinLoad) as minCPULoad, max(MaxLoad) as maxCPULoad, avg(cpu.TotalMemory) as TotalMemory, min(MinMemoryUsed) as minMemUsed, max(MaxMemoryUsed) as MaxMemUsed,…

something like this select n.caption, avg(cpu.cpuload) as avgCPULoad from orion.cpuload cpu left outer join Orion.Nodes N on n.nodeid = cpu.nodeid where daydiff(datetime,getdate())<7 group by n.caption

I dont use auto-discover, but.... Maybe do a sql agent job or ps script that performs cleanup? Or a Orion action that triggers a cleanup script? Probably some more elegant solutions, just off the top of my head.

Its been a while since i configured any *nix servers. In your snmpd.conf, i believe it is where you define the communities and ip/subnets that have access. I think the line looks something like this:rocommunity communityname subnet

Are there any conflicting permissions that may be restricting your? Multiple group memberships?

I do not recall the all in one installer being available for the main polling engine at the time. But that would definitely be the way to go now and is how I did my last set of upgrades

A couple hours. With the new installers though, probably even faster.

We had the same thing, but luckily in 11.5 it is very easy to remove the storage tab and all access to multiple user accounts

aLTeRego, There seems to be a limitation of 3 limitations on an account. How do i configure limitations for each object past that? I really do not want any of the vman/srm data to be visible to accounts with limitations. Thanks, --Ben

Also, if you want to programmatically delete an agent, i would use the SDK. something like this. This is powershell btw, but the same thing could be done via REST also. $swis = connect-swis -swishost OrionServer -trusted or $swis = connect-swis -swishost OrionServer -username user -password password $agentObj =…

This looks interesting, i will test it out. The only issue i foresee is that LEM will see all events as coming from one system.

Thank you! This makes much more sense. Do you know if it is possible to add more than 20 items at a time to the vman charts?

I would probably change it to only do the CPU/MEMORY pollers. Some of the pollers in my list are module specific. $snmpPollers = @( "N.Cpu.SNMP.HrProcessorLoad", "N.Memory.SNMP.HrStorage" ) $wmiPollers = @( "N.Cpu.WMI.Windows", "N.Memory.WMI.Windows" )

Dipak, Thank you for the response. If i manually run "List Resources" on a node; CPU/Memory is there, but unchecked. Checking it enables normal collection. I just want to avoid having to manually enabled it on all of these servers or removing/readding the nodes. Thanks --Ben

It uses powershell remoting, so you need to have that configured on the target node.

SWQL studio is your friend for stuff like this. At least for me. You are going to need to query Orion.Nodes and Orion.volumes for all this data. In powershell i would do something like this. $swisQuery = " select n.caption,n.cpuload,n.percentmemoryused,v.caption,v.VolumePercentUsed,v.VolumeSize, v.VolumeSpaceUsed,…

Yes, but their are views so you do not need to worry about that. When using SWQL it is not an issue either.