Feature Request: Scheduled maintenance for ports

Feature Requests
1

Hi,

Common occurrence for us:

Transit provider sends out maintenance notification for a given link, and we need to reroute traffic from it & suppress alerting for the duration of maintenance.

Few things I could think of to achieve this:

  • Set the whole device under scheduled maintenance.
  • Build individual alerts for each such link, to be toggled on and off, and remove link tracking from other port monitoring alert rules
  • Wait for alert to fire and then acknowledge it.
  • Edit device settings → port settings → toggle Ignore Alert Tag for affected link

None of these are particularly suitable.

We explicitly don’t want to set the whole device under maintenance, as that shuts off alerting for other, valid things.

There would be significant configuration overhead for creating a dedicated alert for each of these links, especially since we have quite a bit of these.

Waiting for alert and then acknowledging it is just dumb.

So the toggle Ignore Alert Tag is I guess the least bad out of these.

Currently Scheduled Maintenance is restricted to be applicable only for:
Locations, Device Groups, Devices

What I would instead want is an actual Scheduled Maintenance for a specific port. Especially since there one could set a timeframe when the alert needs ignoring, so no need to manually remember to set the alerting back on.

So I would request to add Scheduled Maintenance to also apply for:

  • Port Groups
  • Ports

Nice to have would also be:

  • Alert Rule

So that any alert rule could be turned off for a given timeframe.

In my head these sound simple enough, but I have a suspicion it might not be the case.

1 Like
2

This would be immensely useful in any service provider environment. Currently setting interface “maintenance” would involve an external system to toggle the ignore alert tag off and on per interface via api. While this can work, it would be much more useful to be able to simply set a maintenance window for interfaces (manually or via api).

3

we tag our interfaces during maintenances/planned outages etc and then have an alert rule that disregards those ports

ports.ifAdminStatus = "up" AND ports.ifOperStatus != "up" AND ports.ifAlias REGEXP ".*((^|:)T=B).*" AND ports.ifAlias NOT REGEXP ".*DRAIN.*"