We’ve prepared two reports within Report Writer to report on node downtime:
These reports will work with or without view limitations applied.
Feedback most appreciated.
A combination of both
We would like to improve user experience and for such reason I'd like to better understand if our users prefers INITIALLY to add nodes manually or run product network discovery in order to import devices to NPM
I am by far not an expert SQL person, so I am hoping someone here can help me.
I want to create a table of all our UPS's and display certain stats in it. These stats are collected from Universal device pollers which I already have created.
What I am looking for is something like this:
Node Input V Output V Current Battery % Status
ups1 123 123 10 100 onLine
ups2 123 123 15 100 onLine
Can someone help me get started on this. I know there is a way, I just cant figure it out.
Thank You!
Hi all,
I have done this before a number of years ago, but for the life I can't remember how. I would like to remove the 'Reboot', 'Service Control Manager' 'Real-Time Process Explorer' and others from the node management snap-in of the node summary page. I have looked through the GUI and Thwack but I can't find my answers. I'm running version 11.5.3
Thanks
Before I submit a feature request, has anyone yet tried to get monitoring for Rubrik going? It's a pretty nice backup and recovery solution, and with great power comes great respon... er, need to monitor.
Hi all,
Suppose I'm monitoring two client's network devices. In there LAN, they are using same IP range. If two devices one from each client's LAN having same IP (10.2.2.3). Then how can we add them in solarwinds ? While adding second device with same IP, solarwinds shows error.
And what can be the different option we can opt.
I've made some changes to the original Google Maps that I introduced about this time last year. This version provides a status icon on the map for each unique latitude and longitude value in the database. I'm leaving the old one up as it is better for environments where a large number of sites exist, and would simply be too cluttered with a status icon at each site.
Prerequisites:
1) Obtain Google API Key for Maps v3
2) Create and populate Custom Properties:
Installation:
Copy files to the c:\inetpub\SolarWinds\Orion\GoogleMap\
Update your connection string and API key
Create a view in NPM using the Custom HTML resource, configure iFrame
See the movie (how to install the mod):
Read the book (asp files, readme, notes on GitHub):
https://gist.github.com/BarefootAtomic/a396a12541ff97a2ce1f
Enjoy!
Andrew LaGrone, SCP#1368
I have a node I want to monitor (IP address on interface of already monitored router) but under normal conditions it is down and only comes up if there is a problem. Is there a better way to handle this in solar winds?
I have a product that has multiple UDP, one for the service type and another for the service status. Is there a way to link the 2 UDPs in a single table?
Hi.
I am trying to write to the windows event log on a remote Windows 2012 server as an action in response to an snmp trap and can't get it to work.
I can write to the event log on the NPM server itself no problem, but the remote doesn't work.
I pushed the agent to the remote server and can monitor it from npm no problem.
I ran wireshark on the NPM server. Although I can see it having a conversation with the remote windows server, it all seems to be part of the monitoring process. I can't see it doing anything specifically related to the trap.
I restarted the Solarwinds trap service, no change.
Any ideas how to make this work? Thank you.
I've started to notice that some of our alerts periodically fail to translate variables in some cases. This is usually caused by a misconfigured alert, but recently I noticed it happening on long-standing alerts that generally work. In this specific case, it is a random occurrence. In the past few days we generated 142 alerts for this particular configuration and 4 of the 142 failed to translate the variables. While not a huge number, it does mean that 2.8% of those particular alerts didn't work. That's a high failure rate in a business where the only thing that matters is people trusting their alerting!
So I went digging. For context, we run a single instance SolarWinds environment with 17 additional polling engines across 3 data centers with almost 98K elements and 22K application monitors. We have VMAN integrated and have deployed SRM alongside a lightly used NTA. Now I know that the alerting service is SolarWinds.Alerting.Service.exe and Process Explorer says that is is using about 3% of my total CPU on my 8 vCPU. I also know that my processor queue length runs at about 5 for this server and that most of my CPUs are about 60% utilized at any given time.
Except for processor 2 (or in SolarWinds, processor 3!).
From the screenshot below you can see that SolarWinds.InformationService.ServiceV3.exe is railing this processor. All the time. (See screenshots below)
The SWISv3 exe is 2015.1.1.6134. No, I am not running NPM v12 yet.
Is anyone else seeing similar behaviour on this executable? Anyone else noticing problems with interpreting variables where this service is railing a processor? This SWISv3 service consumes far and away the most amount of CPU time of any service on our primary poller. At the time of this posting, SWISv3 had consumed 65 hours of CPU time. The next closest was our Splunk agent (31 hours) and the next closest SolarWinds process was a BusinessLayerHost that was 4.5 hours.
We're just getting started with NPM12, and have installed:
NPM 12.0 with Cumulative Update 4 contains:
Orion Platform HotFix 4
NPM HotFix 1
NetPath HotFix 1
And also installed SolarWinds-NPM-v12.0-HF2.exe.
I have not see NetPath HotFix 1 listed as a downloadable item on the portal page, so I'm not sure if that is the same as what is listed in SolarWinds-NPM-v12.0-HF1.Readme.txt.
In any case, this problem is not with the the 1st hop - it's with the last hop. I'm seeing very high latency on only that hop, for both the canned Google service as well as internal services I setup across our WAN. The latency is hundreds of milliseconds.
Anyone else see this?
!
We are indeed going through firewalls fairly early in the path, and we do not decrement TTL on the firewall.
=Foon=
Since the release on NPM 12.0 we've been hard at working building the next round of exciting functionality and improvements in existing functionality. I'm pleased to share the following list of items we're working on:
Ongoing Initiatives:
Is there a way to build a report for all active serious and critical alerts? It seems that when I go to the report manager and try to build a report my only option is all active alerts.
Hello,
I always had the challenge of creating a historical alert report, none of the inbuilt report were correct and not showing the actual alerts, Unfortunately I couldnt find any information in Thwack as well.
I wanted a daily report Which will have all the alerts triggered, Acknowledged etc.. I created the below query and it is very useful for me to generate quick reports on the historical alerts.Since I also wanted the custom properties to be included in the report I have to fetch the information from 2 tables.
The information which I wanted was in [dbo].[AlertHistoryView] and [dbo].[Nodes].
Create a new web report, Select custom table and use SQL Query. Select the required columns you want to see in report.
Query
SELECT TOP 200 b.CUSTOMPROPERTY1,b.CUSTOMPROPERTY2,b.CUSTOMPROPERTY3,b.CUSTOMPROPERTY4,a.* FROM
[dbo].[AlertHistoryView] a,[dbo].[Nodes] b
Where
a.relatednodeid = b.nodeid and
a.EventTypeWord in('Triggered' ,'Acknowledged')
and a.timestamp >= getdate() - 1
order by a.timestamp asc
Note
Replace the CUSTOMPROPERTY1,2,3,4, with required custom properties name.
you can also select more eventypype if you want.
get date () -1 will give last 24 hours, change 1 to any desired days you wanted to see the alerts
Order by desc or asc.
Attached sample report output,Incase if some one has a better query please share.
SELECT TOP 200 b.customername,b.nodeclassification,b.noderole,b.region,a.* FROM
[dbo].[AlertHistoryView] a,[dbo].[Nodes] b
Where
a.relatednodeid = b.nodeid and
a.EventTypeWord in('Triggered' ,'Acknowledged')
and a.timestamp >= getdate() - 1
order by a.timestamp asc
New
SELECT TOP 10
NodesData.Caption
,NodesData.IP_Address
,NodesData.Vendor
,AlertHistory.Message
,AlertHistory.TimeStamp
,NodesCustomProperties.NodeClassification
,NodesCustomProperties.NodeRole
,NodesCustomProperties.AssetTag
FROM dbo.NodesData
,dbo.AlertHistory
,dbo.NodesCustomProperties
WHERE AlertHistory.TimeStamp >= GETDATE() -1
Please note:
Change Select top 10 to a higher value depending on your setup.
Add/Modify custom properties marked in bold, You can copy the line and add as much as custom properties in this.
Modify Get Date -1 to pull out the number of days old alerts...
We are using NPM 11.0.1 monitoring OSPF neighbor events on our Cisco equipment. Several times now this event was logged:
12/5/2014 2:22 AM The neighbor 10.225.3.198 on Node ourhost.com went down.
When checking the switch for corresponding neighbor events on that date and time, none were found. Why is this alert being generated in Solarwinds when the L3 switch/router does not log any OSPF events?
Any help would be appreciated
Thanks!!
Hi everyone.
I am trying to create a custom report to give me a list of sites with availability less than 95% in yesterday I have selected node name, IP address and availability under fields. The challenge comes when I try to filter fields specifying records that contain only records where availability is less than 95. The result however comes out showing a lot of sites that are not in this category and every site is listed with availability of 0.00%.
SELECT TOP 10000 Convert(DateTime,Floor(Cast((DateTime) as Float)),0) AS SummaryDate,
AVG(ResponseTime.Availability) AS AVERAGE_of_Availability
FROM
Nodes INNER JOIN ResponseTime ON (Nodes.NodeID = ResponseTime.NodeID)
WHERE
( DateTime BETWEEN 42643 AND 42644 )
AND
(
(Nodes.Branch_Type = 'My City') OR
(Nodes.Branch_Type = 'My State')
)
GROUP BY Convert(DateTime,Floor(Cast((DateTime) as Float)),0),
ORDER BY SummaryDate ASC, 6 ASC
Where Nodes custom property named “Branch_Type” fill with 'My City' and 'My State' are required nodes.
Shell be very thankful,
Jawwad~
Hello,
We just install in a test environment the NPM V12.0.
On our MS SQL Infra we use a DR facility that use itself the Full revory model.
We just discover that Solarwinds don't support the Full Recovery Model....
We know that NPM 12.1 RC have now a hight availability mode.
So, what are the best practice of the HA implementation without using the Full Recovery mode of MSSQL ?
Does the Solarwinds HA just cover the application servers (pollers) and not the DB side ?
Cyril
Solarwinds Admin
Good afternoon (t)whackers
I have been building a test environment. The purpose of this environment is for the following objectives:
- To understand and implement a new solarwinds install.
- This includes a clustered instance of SQL
- To test the following applications:
- LEM
- Failover environment.
There are other things I know I need to include, but the late nights have taken their toll and my brain is cloudy today. But to surmise, I want to know how to install and implement a new solarwinds server because of the following (I am also seeking advice on this, so any advice would be appreciated.)
- We have about 2000 + vms/physical servers on a single server with solarwinds installed. (windows 2k8, 12gb ram NPM 11.5) I want to install 12 and start a fresh. Now, in my unbiased opinion, that are some things which are really good about it, but there are other things that are really bad. This is why I am unsure if it is feasible to either start again, or migrate all of the data.
- The SQL is on a separate server (windows 2k8, 8gb RAM)
- We use the following Orion applications:
- SAM
- WPM
- IPAM
- IVIM
- DPA
- NTA
When I try to add a node in my solarwinds test environment I am met with the follow error (please see picture) so, I am wondering what I am doing wrong. It is fine in the old, live environment, but in the test it fails on the validation. I know it's not the community string as they were copied.
It is either an issue with adding the node to two solarwinds, or something that I did in the config.
Any help would be appreciated.