Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials
Store

Email alerts not returning node detail

Hi guys, I'm fairly new to SW and in this case; SAM, and am troubleshooting an installation someone else setup.

The issue I'm having with the SMTP email alerts is that they're coming through without any node details presumably mean to be resolved via DNS, like:

Subject: Error on ${N=SwisEntity;M=Caption} at 08 May 2016 13:34

An issue has been detected at 08 May 2016 13:34 on ${N=SwisEntity;M=MachineType} device named ${N=SwisEntity;M=Caption} (IP: ${N=SwisEntity;M=IP_Address}, DNS: ${N=SwisEntity;M=DNS})

View full device details here: SHFVPTHSOLAR01:80/.../ContainerDetails.aspx
View full alert details here: SHFVPTHSOLAR01:80/.../View.aspx
Click here to acknowledge the alert: SHFVPTHSOLAR01:80/.../AckAlert.aspx

I'm not sure if they've configured the SWQL incorrectly or that SW isn't able to pull the details through for some other reason...

The syntax from the alert associated with the email above under 'Trigger Actions':

An issue has been detected at ${N=Alerting;M=AlertTriggerTime;F=DateTime} on ${N=SwisEntity;M=MachineType} device named ${N=SwisEntity;M=Caption} (IP: ${N=SwisEntity;M=IP_Address}, DNS: ${N=SwisEntity;M=DNS})

View full device details here: ${N=SwisEntity;M=DetailsUrl}.

View full alert details here: ${N=Alerting;M=AlertDetailsUrl}

Click here to acknowledge the alert: ${N=Alerting;M=AcknowledgeUrl}

This message was brought to you by the alert named: ${N=Alerting;M=AlertName}

The node is monitored by the polling engine ${N=SwisEntity;M=Engine.ServerName}

Find more posts tagged with

Accepted answers

All comments

aLTeReGo

The following KB article should help get you pointed in the right direction.

https://support.solarwinds.com/Success_Center/Network_Performance_Monitor_(NPM)/Configuring_website_URL_for_Alerts_and_Reports

djmarco

Hey- thanks for the article. I've been through it, but it appears to be talking about editing the host-name in the URL which in our case appears to be correct? (see below)

The part I'm concerned with is where the email alert displays lines where there's no detail: An issue has been detected at 08 May 2016 13:34 on ${N=SwisEntity;M=MachineType} device named ${N=SwisEntity;M=Caption} (IP: ${N=SwisEntity;M=IP_Address}, DNS: ${N=SwisEntity;M=DNS})

aLTeReGo

In this scenario you would be replacing the hostname with the fully qualified domain name of your server since hostname resolution does not appear to be working properly in your environment. E.G. the DNS search domain is not configured, the Orion server is in a different domain, etc.

neomatrix1217

So what your saying is that the email sent to you shows the variable device named ${N=SwisEntity;M=Caption} instead of device named Matrix? Have you tried removing the variable and re-adding it or creating a new alert and see if the variables get populated correctly.

pratikmehta003

Are you getting only for this alert?

I had faced similar issue when we integrated Solarwinds to another tool for forwarding events. I had approached to customer support and what I was told that due to some global un-acceptable characters, this becomes an issue. For ex: an extra space or some special characters.

djmarco

Well I've managed to get permissions to the SQL Websites data table and edit the server names field as per AlterEgo's advice- thank you for that...

neomatrix- thanks for your feedback; based on your advise I've re-created the alert from scratch too. So after carrying out these steps I've triggered the 'Simulate' function and it does appear to now be formatted correctly but of course had no content, so I've left it active in the live environment now to see what comes through...

Thanks all- I'll pop an update on here as soon as I see some results.

Marcus.

Re: Email alerts not returning node detail

muckman

Using the alert variable wizard you can select an object to populate values in the valid variables. This helps you identify the variables that are not valid. Try selecting your node in question and see if you can populate the values before selecting the variable.

Screenshots of how your entire alert definition is configured would be helpful for troubleshooting.

djmarco

Hi thanks for this- I've included here some screenshots of the original alert configuration which I've since tried to replicate and re-apply. In the re-created version-all I managed was a kind of chopped down version which although didn't have the same errors, only alerted on the fact there was an issue, and then included links to the alerts and nothing else. The issue is that links are useless outside of their network of course, so I really need the email body to contain more info about the node and the actual issue as described above.

Then editing the email body, followed by the fully email body code:

View full device details here: ${N=SwisEntity;M=DetailsUrl}.

View full alert details here: ${N=Alerting;M=AlertDetailsUrl}

Click here to acknowledge the alert: ${N=Alerting;M=AcknowledgeUrl}

This message was brought to you by the alert named:

${N=Alerting;M=AlertName}

The node is monitored by the polling engine

${N=SwisEntity;M=Engine.ServerName}

j_a_catlin

Based on what you are looking for in the alert you may be better off changing this alert to I want to alert on Group Member instead of Group. After making that adjustment it should work for what you are looking for.

muckman

That's exactly what I suspected. Your alerting scope is Groups which doesn't have access to node information.
I just tested Group Member which also doesn't have the variables you need.

When applying variables you can use the output preview by changing the test object.

This will show you which variables will work.

j_a_catlin

I guess the bigger question that even comes into this is what are trying to alert on? Groups can contain more than nodes, so that is one issue that arises here. If the group only contains nodes, are you wanting to alert on nodes that belongs to this group, or are you wanting to alert on the group status itself? If it is the groups status, it will not report the individual node info the group because it can't be quite sure which node you want information for. It would only be able to report on the group that contains that node.

If it contains more than nodes and you want to alert on any group member, you can't get as specific as you are trying to get since that data is node specific. You can use items like group member name, group member status, details url,

If the group is all nodes and you want to alert on a node that has one of the statuses listed in the trigger action and belongs to the group, then you may be better off setting a custom property that IDs it as a part of the group or potentially using a custom SQL query.

muckman is correct that you should test the variables that are available to you when creating the message. Just because variables work for one alert type, do not mean that there viable in another alert type.

If you give us an idea of what potential object types this could alert for and the overall idea of the alert, it may help in us being able to identify what alert setup would work best for you.

djmarco

Ok guys, just want to say thanks for sticking with me here... turns out teh group contains.... other groups... which when drilling down further do indeed contain critical nodes. So at the level below this one, the group statuses can be in states such as down/warning/shutdown/critical statuses. So i need to the trigger to be on those group status if poss! (WM-NOC is the parent group)

alert groups1.JPG

alert groups2.JPG

j_a_catlin

To get the kind of info you are looking for and put that data in an alert that triggers on groups, you will need to build out a SQL variable instead of using a default variable. To think through all of the possible combinations of what could potentially come up with the alert, it could be a pretty complex SQL query. It would need to filter to the most recent device in the group that changed status and change the data returned based on the objecttype that triggered the status change. It can be done, but it may not be worth the effort in the long run.

I would suggest creating a custom property on the nodes for that group - like a true false for WM-NOC and setting it to true(or a custom property that has a list of your groups if you need this for more than one) on those critical nodes that are part of that group. This would allow you to create the same alert targeting a node in the group and get you the data you want. It would also help if you need to do any form of reporting on the nodes specifically in this group as well.

When targeting a group for the alert, the alert has no ability natively to ID the specific node with the problem and can't grab its data natively. It can only grab the group's data natively.

djmarco

Thanks Jeff- looks like I need to back to the IT guys who looks after the IT for this contract and try to figure out a solution to this based on what you've very helpfully suggested here.

Really appreciate it- will feed back here asap.