jeremiahmcclain ✭✭✭

I appreciate you working on this, however SW tech support told me that you cannot compare against all rows in the table without using row ID and comparing them separately. I could create discrete alerts for all of the various things i need to alert on but that is an enormous number of alerts. And, really, a product like…

I hope this works, the UnDP should be attatched. [View:/cfs-file.ashx/__key/CommunityServer.Components.UserFiles/00.00.00.98.08/ciscoEnvMonSupplyState.UnDP]

So what is the plan to fix this? APM 3.0?

Any insight would be appreciated, thank you.

So, is this a defect with the WMI architecture or an issue with Orion?

Here is what the db looks like for the node in question. All the nodes look to be this way.

That worked, thanks!

Ok, now i know what is going on, SW is treating the 802.11a and 802.11b/g radios as separate interfaces. I verified this in our Cisco Wireless Control System. No mystery now, but i would be nice to change this behavior or not include them. 

The page is blank except for the text "Nodes with Problems" and a help button. Here is the URL if you would like to test: http://<server>/Orion/Resource.aspx?Resource=/Orion/NetPerfMon/Resources/Summary/ProblemNodes.ascx

I did a quick capture, and it does hit the bufferElMiss OID, but not any of the others as far as I can tell.

In this case, we are monitoring ciscoEnvMonSupply state with a UnDP. We are getting the table and the value type is a raw value. The device I testing against is a Cisco 6509. I would like to have the alert query each row in the table against a custom property which is set to the expected value. For testing purposes, I…

Thank you Denny, that is exactly what I needed.

I am having a tough time reconciling my numbers as far as the interfaces go. Right now I have 1412 interfaces showing in the database details page. I counted the number of interfaces on all nodes except the WLCs and came up with 309. The WLCs show 569 APs and 397 rogue APs. Adding all those up i still come up short with…

Yes, it is a 4-core, single proc.

well, that didn't work, i tried just using ${SQL:<query>} with the query from report and that doesn't work either. I am stuck.

I am looking at the NetPerfMon db. The manual is vague as to what those settings mean, so I am looking for clarification.

Here is my MIB walk for one of the devices in question, looks fine to me. I also took a look at the buffers hourly and detail tables and there are no negative numbers.

This raises another question, the UnDP is set to raw value, neither rate nor status, does this make a difference in the alert. There is no "field raw value" available in the alert.

The reset condition is a copy from the trigger but field status is set to equal in both conditions. However, I think I see your point. So I should set the trigger for the group conditions to any and the reset to all? 

As a follow-up, it looks as though rogue aps are counted as interfaces as well, is that true?

Good to know someone else is seeing this issue.

created a report and the report shows the value of the custom property. Here is the SQL statement: SELECT Nodes.chassisSlotConfig_threshold AS chassisSlotConfig_threshold FROM Nodes Should i just use ${SQL:<query>} and use the query above?

Thanks for the answer, I now see the APM compnents I guess I missed them the first time. Someone should tell SW tech support though, they told me there was no way to add APM components to the map. I am looking at nesting maps, and that will help. I would like to be able to add colored shapes to group things together.…

I periodically get this message as well. The first time I called SW tech they told me to run the config wizard, which I did three times, and the issue was resolved. It came back a second time, ran the wizard again, and its been stable for a few days. Not that this helps you, but you are not the only one.

I had this problem as well, in my case it was happening on devices that I was not collecting CPU/Mem. 

Just noticed, the subject should say Problem Nodes. The Problem Areas page works fine.

I had to create a UnDP using the laLoadInt oid in the ucd mib to get those values.

 We are looking to do the same thing (on VMware server). I have contacted SW directly, and they did not specify any issues with running Orion as a VM. Aside from the usual system requirements. You may still want to talk to them about MS virtual server specifically.

Looking at the release notes for SP2 it looks to be fixed in that release. I haven't moved to SP2 so I can't verify.

We have, seemingly, a different problem. We get syslog relayed from syslog-ng. The admins do not want to recompile to add source address spoofing. We need SW to parse the source address from the message and log as that IP. Is that a possibility given your scenarios?