So I'm looking to create a trigger action for our remote office edge devices. We are wanting to alert anytime an external interface goes down for multiple remote sites. I'm wondering what would be the best way of going about this?
Here is what I have so far...
Obviously serial interface could mean any connection that is considered serial. How do I lock it down to specific WIC cards?
We did this by adding a custom interface property. In various installations that can be as simple as a Yes/No field called "alertifdown". Then your alert trigger would add another simple condition if "alertifdown" is equal to "yes".
We also added the exact opposite for our cellular backup circuits (alertifup). That way we get a notification if the cell circuit comes up (ie: we start getting charged by the carrier per minute).
Taking that a step further, we also have an "InterfaceCategory" custom field that is either WAN, UPLINK, VOICE or OTHER. So we can have some interfaces (serial or otherwise) alert and go to the voice team, other alerts go to the network team, etc.
For what it's worth.
I believe I have accomplished this in the past with a client. You may want to look in the DB, and find something that is common between the serial interfaces and use that as a trigger item.
Here is an example that may work.
That's somewhat what I'm doing right now. I have it set as node status is down> Interface name starts with Serial 0/1/0 and this triggers an alert after 2 minutes. I was hoping to get more general but it seems like the granular route is the best way.
Thanks everyone for the help!
This might be an obvious answer but why don't you use the built in "Alert me when an interface goes down"?
I use it in conjunction with the "Alert me when a node goes down" and I get email if an interface or node goes down. You could add in a group variable if you want to only have these alerts for certain devices. And the nodes/interfaces are set to poll every 2 mins for status.
We are currently using this out of the box alert, but it does not get as specific as our management team would like. We are an internet database company so statistics are key in every aspect of our days. So, incidents such as node down need more automated details.
What automated details are you wanting? From your alert, all I see is you getting an email when a serial port goes down. What extra details are you getting over the built in "Alert me when an interface goes down"? Or are you adding in extra information into the email part?
This may or may not help but with my interface down alert email I also add in the interface description field so from it I can see what the interface does.
what about to create a special group for your remote sites and trigger an alert when group status change? You can create dynamic rule for your group if needed based on interface/node IP or custom property - especially if you try 10.4 RC with web based custom property editor.
Yes, creating a custom property would resolve the issue, but maintaining this would not be feasible. I have not tried RC 10.4 is it stable or are there many issues? I would like to jump to 10.4 since it has better support for F5 units.
I would just create a custom property named 'type' or something along those lines then in the description put 'WIC'. Then when you come to your advanced alert you could put in a condition that said when custom property type = WIC
SolarWinds solutions are rooted in our deep connection to our user base in the THWACK® online community. More than 150,000 members are here to solve problems, share technology and best practices, and directly contribute to our product development process. Learn more today by joining now.