I was looking in my NPM at the End of Support Lookup, and there are a lot of my routers that have the Part Number field incorrect. But the Machine Type IS correct.
What is the difference in these two fields?
Machine Type seems to be generated by SNMP. Where does NPM pull the Part Number field from?
How can I edit the Part number field?
Hey All! Now, you’re probably looking at the title of this and thinking “yeah, that’s easy enough, just create a custom chart and get the top 10 already”. Well, this one is going to be a bit different. What if, instead, you want to see the top 10 interfaces, but the interfaces are in separate charts? Perfstack seems the obvious choice, but perfstack only gives you a static list, not a top 10 list. That’s where things get a bit tricky.
But fear not! With the power of SQL, this can be accomplished:
Here I’ve gone and created a view in Orion that contains the top 10 interfaces across the entire environment. Each one of these resources is a custom chart that is based on a custom SQL query. Each query is exactly the same, except there is an offset number that is changed for the first, second, etc. interface. Below are the steps in creating:
Step 1 - Create a new view (Optional, but recommended):
Go to Settings --> All Settings, and click on Manage Views. Click on Add, then create a Summary View called “Top 10 Interfaces”
Step 2 – Add the new view to a menu bar (Optional, but recommended):
Go to Settings --> All Settings, and click on “Customize Menu Bars”. Click on the menu bar you want to add the view to (I used Network_TabMenu), and select Edit.
From the list of available views on the left, find the “Top 10 Interfaces” you created, and drag that to the right.
Now you have the new view under My Dashboards --> Network --> Top 10 Interfaces
Step 3 – Create the widgets:
Click on the view we just created in steps 1 and 2, and on the left side of our blank canvas, click on Customize Page:
Then go to “Add Widgets” in the upper right:
Search for the “Custom Chart” Widget, then drag and drop that widget 10 times to the main part of the screen (5 in each column):
Click “Done Adding Widgets when finished”
Step 4 – Make the charts:
This is the meaty part. On the first chart, click on “Configure this resource” or “edit” (both will take you to the same page):
For “Title”, call this “Interface #1”
Subtitle we can leave blank
Then for datasource, click "Select Datasource", and change the selection method to “Advanced Database Query (SQL, SWQL)”, then change the query type to SQL:
Now for the query, copy and paste the following:
SELECT
(no.caption + ' - ' + interf.interfacename) as [Node Name],
interf.interfacename as [Interface Name],
it.datetime,
it.in_averagebps as [Average BPS In],
it.out_averagebps as [Average BPS Out]
FROM [dbo].[InterfaceTraffic] it
inner join Interfaces interf
on it.interfaceid=interf.interfaceid
inner join nodes no
on it.nodeid=no.nodeid
where it.interfaceID=(
SELECT
it.interfaceid
FROM [dbo].[InterfaceTraffic] it
where it.datetime >= DATEADD(day, -1, GETDATE()) and it.in_averagebps IS NOT NULL and it.out_averagebps IS NOT NULL
group by it.interfaceid
order by(avg(it.in_averagebps)+avg (it.out_averagebps)) desc
OFFSET 0 ROWS FETCH NEXT 1 ROWS ONLY)
So, quick explanation as to what this query does. The first part, up to the “where” statement, pulls the interface traffic information. The magic happens in the where clause. The where clause takes the last 24 hours of performance information, per interface, and averages it out and sorts the top interfaces.
The important piece to this is the LAST line;
OFFSET 0 ROWS FETCH NEXT 1 ROWS ONLY)
What this line does is return just one line, but offset the rows X number of lines. So, for this chart, we take the top result. For the next chart, we will take an offset of 1, so our query is now:
OFFSET 1 ROWS FETCH NEXT 1 ROWS ONLY)
This query will give us the chart for the 2nd interface in the top ten, and so on.
Step 4 – The rest of the chart setup:
Now that our data source is set for the first chart, you can now set the rest of the options.
And that’s it! Hit submit.
Step 5 – Rinse and Repeat:
Repeat steps 3 and 4 for the other ten charts, and make sure you:
Change the offset number for each chart (Interface #2 has an offset of 1, Interface #3 has an offset of 2, etc.)
Change the name of the Charts for Interface 2, 3, etc.
That should do it! From this point, if you want to change the query to include a custom property, or maybe do volumes or cpu performance, you can. The real part of this is the where clause where you look up your information ahead of time, then offset/fetch. Let me know in the comments if you have any questions!
Recently in a conversation with some Thwack MVPs there was some lamenting over PerfStack™ color palette. I had a few minutes so I decided to look into it. This is a quick write up from the results of that conversation.
I found the path for the cascading style sheets is here: \inetpub\SolarWinds\ui\modules\perfstack\css
At the moment I do not understand the difference between perfstack.css and the perfstack.min.css - how they interact or if only one is necessary. The kicker for a simple search and replace is the values in perfstack.min.css don't exactly match what is in perfstack.css. An example is .color-1 rgba(181, 34, 0, 0.8) will be in perfstack.css and .color-1 rgba(181,34,0,.8) will be in perfstack.min.css. The spaces are removed and the leading zero on the saturation is dropped. In addition, perfstack.min.css has no CR/LFs, so it is one line. It would be nice to understand the relationship between the two files. Is perfstack.css used as the base and the perfstack.min.css was created after the web server started? Otherwise, it can be a little painful but you only have to do it once.
I thought it would be neat to be able to create several sets of files based on some "top x" list of web color schemes then post them with an article saying, if you want to use this color palette use this perfstack_Iowa_State_.css file set and replace your originals. Or perhaps write a PowerShell script that would take a file containing a color palette of your design and create the PerfStack™ color palette files for you. Alas, I do not have the time right now to do either of these things.
So using the site http://paletton.com and data from the Iowa State University web site I created a scheme based on Iowa State University. There are five basic colors from ISU, but there are 17 in the css. I used the basic colors and had paletton give me the rest of the color palette to create the two files. After making a backup of the original file and storing it in a safe place, simply replace perfstack.min.css on the web server with the perfstack.min.IowaState.css file. Do the same with perfstack.IowaState.css and perfstack.css.
The files use a combination of HEX and RGBA color definitions. The colors are paired up in different classes so that the legends in PerfStack™ and the charts match up.These color changes will effect all your PerfStack™ charts.
I hope to expand upon this as time permits. [I used Iowa State since my boss went there, my undergraduate school didn't have a football team.]
Hi,
How can I setup NPM/NTA to monitor for P2P behavior on my network and identify the source?
Hi there
How are other users recording asset tag information for multiple switches recorded as one node in solarwinds?
i can get to the devices once im into the stack but im keen on recording each individual device as a custom properties so that info can get ported back to our asset database as it does via an API for a single device node
Im still using NPM 12.0.1 too
Cheers]
Gary
I have a few type 2 services with limited visibility beyond my handoffs. I need to be able to trigger an alarm when bandwidth TX or RX goes to zero packets per second for say 5 minutes. This would help our NOC be more proactive about a possible issue upstream of our network. We have snmp and network elements set up in NPM already and actively monitor and have alarms for errors and high bandwidth usage on standard ethernet interfaces... Works great. Just struggling with how to make the happen for no usage. Hope one of you out there has done this and can share your method.
NPM 12.3 has shipped and we're hard at work building the next release. Here's what we're working on, in no particularly order.
When troubleshooting cloud-based audio/video communication tools (BlueJeans, Skype, Zoom) for packet loss, jitter etc within your LAN and WAN environments, which SolarWinds tools do you use, and how? Would it be interesting to you to be able to get an end to end view of the call path (user -> LAN -> WAN -> Cloud tool and back) and associated path metrics?
After upgrading NPM from 12.2 to 12.3 I've had an issue with using a filter on the "Volumes with High Percent Usage" resource. We've been using this filter for years to keep a few dozen volumes that it is acceptable to have a volume with high percentage usage out of this table. There must have been a change between these reversions and this has rendered my filter that relies on a custom property value useless. I see that there is a "CustomProperties" SWQL entry under the resource but I have not figured how to use filter with my custom property value.
Does anyone use "CustomProperties" to filter volumes? Can you please provide an example of how you are tying in your custom property value into the filter?
Thanks for any help!
I am trying to use Solarwinds bandwidth monitoring to monitor the bandwidth ports for the Trendnet TI-PG1284i switch but it is not capturing the data. I tried with Orion managed nodes and SolarWinds IP monitor. It is not gathering the data but I do get response times and other registers. The SNMP community string is correct. The bandwidth usage is non existent and I know this to be inaccurate. Do you have any idea of what could be the problem? Please help on this because I really need to know how to capture the bandwidth usage on these type switches.
How can you set an alert to all groups except one? I have a lot of locations so creating conditions for each location beside the one I don't want to monitor would take a while. So how can I create a condition that says monitor all nodes except these nodes?
FYI: If you want to monitor the temperature of HP Proliant servers and you've installed SNMP and the Compaq Insight Management software. You can create a Custom MIB Poller using OID (1.3.6.1.4.1.232.6.2.6.8.1.4).
cpqHeTemperatureCelsius: This is the current temperature sensor reading in degrees Celsius. MIB value type should be set to Raw Value, and type is set to none.
Unfortunately, you can't yet perform math on the Custom MIB Poller to convert the value to Fahrenheit.Hi All,
Want some help in creating alerting on custom pollers that i have created for hardware monitoring. below screenshot is from a single device where i want to alert different hardware parameters like Fan, Power, Temp etc..
Can some guide me how exactly i can do it? I tried couple of method but getting confused....
Ex: i want to alert on temp sensors going above a value of 45.. OR i want to alert on Power supply when its not normal( in this case value will be anything apart from 2. value of 2 signifies its OK).
The other thread is closed so I figured I would start a new one I usually get more help here than actually contacting support.
So same issues as before but instead of the server not responding in 36 hours or so it took maybe a week but it is the SAME issues.
1. Server stopped sending alerts out sometime around 11AM on the 4th.
2. Logged onto server and opened Orion service manager and both the module engine and the administration service were going back and forth between running and stopping.
3. Orion could not connect to SQL
4. I have some alerts that at are going out but not sure if they are legit or not.
5. After the reboot I notice that a good chunk of my nodes interfaces are 'unknown' this looks like it fixes itself but again something else going on.
I have applied the 'hotfix' that you all pushed out to try to fix this.
I have done the change from streaming to buffered
I have done the registry change for the ports
The only thing I have not done is revert the snap shots back to June 14th prior to the update so Solarwinds is stable again.
At this point I am going to schedule a task in VM Ware to reboot the server every night. That is pretty much the only way I will know Solarwinds will actually work.
Hi All,
I want the 'Node Name' as a hyperlink so that a single click on the node name can take me to its details page. I want this in a custom table on a page view. PFB the screenshot of a seemingly available option 'Allow HTML tags', but not sure how to use it.
Can someone please advise on this.
Many Thanks!
Under Edit Node/Resources/Volume Utilization:: why is Physical Memory/Virtual Memory Included?
And what is the diff between CPU & Memory and Volume Utilization/ Physical and Virtual Memory selections?
Our NPM 11.5.2 environment reported a significant change in average CPU utilization for one of our Windows Server 2008 VMs monitored via WMI (past few weeks shown below). However, if you log into this server directly it still shows <5% CPU. Also, the reported CPU increase happened immediately following the weekly scheduled reboot of our SolarWinds servers.
We have multiple NPM 11.5.2 environments, so I setup two of the others to monitor the same server - one via WMI and one via SNMP. They both accurately showed <5% CPU while our original SolarWinds environment still showed around 40%. I went back to our original SolarWinds environment and changed the polling interval from 10 minutes to 1 minute for more granularity. As soon as I made this one simple change, it started reporting the correct value (shown below just before 12:00 PM).
I'm concerned about why this happened because it calls the data for our many servers into question. Anyone else encountered this or have some theories?
Problem: DBAs love to add/change drives and drive configuration on the fly without anything resembling Change Management and then blame us when a drive fills up/fails.
Question: How can we ensure that any new drive/volume additions are included in monitoring?
It would seem that selecting "All Volumes", then deselecting selected items, that future added disks would be included... yes?
Example here setting up a new node Resources to monitor:
