marcrobinson · Observability Detective · ✭✭✭✭✭

If you want it to have it contain information related to that app - consider creating a custom table widget, or a perfstack report and add the perfstack to the page.

check this thread. https://thwack.solarwinds.com/t5/NPM-Documents/Orion-URLs-for-Firewall-Whitelisting/ta-p/613706

Try the percent symbol. 

Mark, Thanks for keeping us posted on this issue. I am still sitting on the sidelines waiting for the hotfixes to cool down. I have a multi-module, multipoller installation and fear that I will fall into the same pain that you are experiencing. Especially with the new installer, which I love otherwise. Thanks Marc Robinson

@"deiberts" Good catch on the black box monitors. I checked another insight app - active directory and noticed the same upward shift in chart data. 

Hey @"johnwilly" That is a great question. Unfortunately, not the best answer. If you use default settings then you cannot name the disappeared AP. I learned that the db has a setting to remove them by default. You will need to go to YourOrionServer/.../Global.aspx and scroll to the Wireless section. There is an entry call…

in your code sample, it looks like there is a space at the end, between the swql variable and the two closing brackets. I can't see any other issue, especially since it appeared to grab the Manual_Scan job name in your test. We are still doing OK using the TOP 1 filter for the Discovery Successful and Discovery Failed…

I’d recommend getting current if you can – just pay special attention to the windows OS version requirements and the sql version requirements. Also – lots of shiny new toys, er features, since 12.0. Good Luck! Marc

sorry for not replying sooner. There is a table that stores the configs with xml wrappers.I believe that also has the template name. I will continue to poke around for the template and post what I find.

Creating a feature request for loggly still appears disabled. Here is the same view for SAM, with the links to submit a feature request.

Glad I could help a bit. If I can fix the user defined swql variable, I will post it. I do not believe that this one is in the named variable lists. Then you can add it to your email notifications. 

I agree with Superfly99. I have stopped my planned upgrade until this settles down. IPAM is too highly used to chance it. But I REALLY want NPM 12.2 and the new SAM and NCM updates…. This will get fixed, so wait if you can.

I cannot select it as an option though. See image below, Loggly feature request forum is grayed out.

Here is the bad version. 

Thanks for the tip. I was able to notice that the template was not logging in. I switched it from normal to always for authenticate/login. THat did the trick.

I do plan to upgrade to NPM 12.2. But I am waiting for next round of fixes for IPAM 4.6 to get posted. Here is that thread: IPAM 4.6 is broken .

Currently testing an alert based on the event that is logged when a thin ap disappears. THis is NPM 12.3. There is no reset condition, as it logs disappearances only. The logic is shown below. This has worked pretty good so far, but I have not completely switched to it.

Hey lynchnigel, Good Catch, and thanks for that tip! I never even thought about the complex conditions. And I am at NPM 11.5.3 (soon to be 12). THe alert I am using was a Wireless Access point. See below, since its a WAP alert, targeting the thin AP's, your solution should be good. I will test and post back my results.

I think mute and perfstack were released in orion 2017.1 and NPM 12.1. I was able to find release notes for NPM 12.1: NPM 12.1 Release Notes . Looks like these features were released together.

@mesverrum I dont know why I never thought of it before, but I have been beating my head against the wall on a swql, not a sql query (Which works) to get the devicetemplate for NCM nodes. Device Templates are not in the NCM nor Cirrus entities. I didnt look at the CLI entity until seeing your post. Thank you! I am loving…

The information on the net name can be found in the Key-value pairing in the below swql entities. There ought to be a better method to do this, but this is where I have found that information. The query should generate a list of the services and their net names. The joins bring the two entities that have information on the…

A big part of the issue with WMI has to do with how security was setup in wmi and windows. You can script/gpo a response to permit non-admin accounts to grab data. But you will need to account for variations in operating systems. 2008 and 2016 will require subtle tweaks for example. If you want event logs, that is another…

Is there a thread in NPM for this? I would like to follow this discussion.

I was looking forward to the IPAM 4.6 upgrade until I read this. It sounds a bit like the growing pains from the NPM 11/11.5 upgrades. As much as I was planning an NPM 12.2 multi-module/multi-poller upgrade, I think it may be wiser to wait until this gets resolved. ding​ - can you follow up on this mega-thread if a new…

Here is a swql version. But it looks like it triggers on any difference, up or down that meets the threshold. I am still sorting that bit out. SELECT v.FullName, v.VolumeID, AVG(vh.PercentDiskUsed) as DISK_Used FROM Orion.Volumes v JOIN Orion.VolumeUsageHistory vh on v.VolumeID = vh.VolumeID WHERE vh.datetime> ADDHOUR(-2,…

I am on 12.1 HF1, Orion 2017 HF3, and this suggestion worked on a poller also. Disabled the cleaner, copied the files (from the backup repo in my case), and ran config wizard. I was using the scalability installer to actually install everything on a new poller. Thanks for posting this solution!

We had a problem with it returning the same profile name every time. This custom variable has been working for us, but it is not elegant. I am using NPM 12. I call the top 1 job (single job) and use a where clause to reorder and filter by successful jobs and timestamp in descending order. ${N=SWQL;M=SELECT Top 1…

This appears like it might work. It should only alert when the controller is online and the ap is offline. I briefly tested it and it looks ok. I make no guarantees.

I know this is a bit late, but were you ever able to test this alert logic?Specifically if the controller went down, would the thin AP's still alert?

Looks like removing the view limitations and resetting them is the answer for now. Thanks for looking into this.