dhanson

You could also use custom interface properties to match up to this, but that might actually be even more work than using port descriptions.

Although it is very likely you've already solved your problem, one of the other reasons SolarWinds may react this way is a conflict between configured SNMP versions. For instance, if the node is configured for SNMPv2 and SNMPv3, and you're only polling in SolarWinds for SNMPv2, the SNMPv3 messages will be considered a bad…

Should be under "Settings", Then in the "Nodes & Group Management" section, there's an option to "Manage Custom Properties". In this window, select the property you want to edit, click "Edit", and add/remove values as needed. -HTH!

My query gives you the total number of polls in your [InterfaceTraffic_Detail] table that are in excess of 70% utilization. As Craig Norborg said, how this translates is based on your poll rate, and your retention settings. My statistics poll rate is 1 minute (small environment) and my data retention is 30 days of per…

Another caveat that would prevent crossing between SQL and SWQL: CTE's. SQL works fine with CTE's, but SWQL does not. There are also functions in SWQL that provide ease of use that also don't cross back and forth between the two (like for instance, any of the ${variables}). Then there's PIVOTs, which SWQL does not support.…

The "Name" heading was not a selectable option regardless of the filter. Nor was the assign dates or refresh available. Ironically enough, Jiri was right, it was a permissions issue, and I needed to log out and log back in to enable the correct permissions. I appreciate your assistance, though!

My account has full admin privileges. Anything else you can suggest?

Why not using nested conditions? Something that logically flows like... trigger alert when all of the following apply: Trigger alert when all of the following apply: custom node poller unique name is equal to Status_UNDP_1 custom node poller RowID = # custom node poller current value is equal to <value> Trigger alert when…

I'm running 15.0(2)SE6 on a stack with no excessive discards. Fully populated with WS-C3750X-12S-S and WS-C3750X-48T-E stacked. Can you do "show mem overflow" and see if there are any results? Also, how is your CPU utilization? I'd absolutely recommend a TAC case.

Hah! I never spotted that. I had always dealt with the custom chart without modifying the selected data source. Thanks, sir!

Seems like a similar issue that alok.malvi‌ had to deal with. I'm under the impression that the Alert Log might be homed elsewhere in 11.5, or may no longer populate, and I haven't upgraded yet (too many reports of failures with 11.5, 11.5.1, and 11.5.2) so I don't know for sure. If you want to hunt around in the database…

In that case... Events are significantly different than alerts. There's a separate "Event Log". You can have alerts write to the NPM Event Log as a trigger action. When an alert is triggered, it is written to the Alert Log automatically. Since you are using 11.5, it is possible there is an alternate log file related to…

Alright, so report writer. The application right? not web-based? This is the query straight from my Report Writer, and an image of the results. If you yield no results from this, I'd definitely review the tables and column names in 11.5. It'd be pretty cool of you to let me know if you find any discrepancies, because…

There aren't any CP prereq's. I built this on NPM 11.0.1, so if you're using 11.5+, some of the table structure has changed. I'd review the table names to make sure they're accurate. Also, are you attempting this in SWQL Studio or Database Manager? Or as a Custom Query resource on a page?

This is my comment.

I take it the particular device doesn't have a DRAC or ILO port? No management interface that can be connected in a different VLAN? Otherwise, you're stuck putting a NIC from your poller into the VLAN. Another option is to make the VLAN routable, but put a policy denying any traffic aside from management traffic. Not an…

well, you could always use a TACACS or RADIUS servers for authentication. Typically those applications may have a reporting function. If not, you could poll for sessions (basically find the OID that presents the results of a "show user") and compare them against an approved list. That might take some manual labor (in…

I'd assume that the concern for NTA install as a VM was based on the bandwidth requirements, however NTA isn't super bandwidth heavy. You will need a third server, though, for NTA's dedicated flow storage database. We've run our 3 servers (with a similar scope) as a VM with no problems. In fact, we use a shared SQL…

Well, I can attest to the need for database experience to enhance the value of reports. I've also observed that reports, whether from the server-based application or from the web, can have very different limitations (like only one of them can be graphed, and only the other can be added to a node's detail page in a…

A workaround for this is to create a UnDP for the WLC's to pull their IOS. It won't appear inside the "Node Details" tile, but you can create a custom poller tile on the Node Details pages that can show the results of the custom UnDP poller. Here's how: 1. Create the UnDP. - On the server, open the Universal Device Poller…

This should answer most of your questions: thwackCamp 2015 FAQ On the individual items in the schedule, you can RSVP to each individually. Make sure to pay attention to the times, as some of the courses occur at the same time! Don't wanna double book yourself. HTH! Edits: Perhaps you wanted some clarification on the…

Why not plug in the whole SQL string and try? ${SQL:Select DATEADD(hh,-4,LastSystemUpTimePollUtc) FROM [dbo].[Nodes] WHERE NodeID = ${NodeID}} -HTH

This post might help you get started. Look at any post about creating UnDP's and they should be able to help you get the details you want from your system. Not sure about integrating video though, that might be a tough one. When Good OIDs Go Bad: Customizing Device Pollers

They might be right about the memory leak. Any server monitoring to observe with? You should be able to tell pretty quick. Also, review any Custom SQL Queries or reports you have running or scheduled to make sure they aren't looping, and creating an endless query (or a temp table too massive for your system to handle).

Well, you can get general up/down alerting, CPU/Mem statistics and (if you have NTA) netflow collection without custom pollers. If you want something to alert on outside of the SNMP trap viewer rules, then you'll have to establish some custom pollers and set them up for alerting, although in my experience with Riverbeds,…

You can create a SQL alert like below. This one, as configured, would trigger if any one node had a status other than 1 (1 is "up"). By the way, the AND statement, that entire row is optional. If you only want to alert on specific locations, and not apply these rules to others, you'd have to define this AND statement.…

Have you looked into creating a report with the information you require? The web-based reporting feature should allow you to add custom resources covering custom time periods, and be reported on at custom intervals. So, you can have unique resources in a report, that all cover the last 12 hours, and be distributed every 12…

Ok, so before you get terrified by the length of this reply, just know that it's definitely easier than it looks, and seriously, the amount of action required, you should be able to get this whole thing set up in like an hour or two, and once you get it done the first time, you should be able to repeat it in less than 30…

Well, if you write to the event log both when the event triggers and resets, then you can diff the timestamps in the event database. It might take some advanced SQL to write a query that will successfully filter these based on NetObjectID as pairs of Alert and Reset, but I'm sure it is possible. I can't mentally compose a…

Your DBA is right, you're going to have to do a join to compile the lists if you want to pull them from more than one table, but the problem is that the Events and Alerts tables have nothing in common, really. Then you can use a UNION if you want to collect Event Messages that include it as well, but it is extremely…