Automatic dependency calculation

Running NPM v11.5?

Yep.

UI Issue- thanks for flagging sir. Setting also found under Settings->Polling Settings

Is there a good doc on the implications of doing this?  For instance, while I'm all for the tool doing the work, I'd really like to know when and if it makes changes in dependencies so that perhaps I can approve or deny those dependencies.

I'd hate for something to be missed because it was marked as dependent on the wrong upstream object.

Outside of the admin guide, we don't have further documentation. I believe a KB is in the works. If auto-dependencies are enabled, we we will recalculate dependencies every topology polling interval.

Auto-dependencies are displayed on Manage Dependencies grid as "Calculated Automatically" under "Origin." If a calculated dependency is not accurate, it can be deleted and it will move under the "Ignored Dependencies" list. These dependencies will not be recalculated. If you have a current manually-defined dependency, any auto-deps that would duplicate will be discarded.

rob.hock

Any news on the KB?

I'm also looking for more information on auto dependencies...

While were at it, this might be a good thread to ask about implied dependencies. I have always understood that certain dependencies were automatic in that they did not have to be defined by the user. Examples would be interfaces on a node or applications on a server. In each case, Orion would automatically place those interfaces or applications into "unreachable" status if the parent node was down or unreachable.

Am I correct in this, and are there any other implied dependencies I have missed?

Thanks.

While were at it, this might be a good thread to ask about implied dependencies. I have always understood that certain dependencies were automatic in that they did not have to be defined by the user. Examples would be interfaces on a node or applications on a server. In each case, Orion would automatically place those interfaces or applications into "unreachable" status if the parent node was down or unreachable.

Am I correct in this, and are there any other implied dependencies I have missed?

Thanks.

I have found that my Application Monitors (like chkdsk) will go into Critical status even if the node is down, so that part has not been true for me.

Testing with NPM v12 and SAM v6.3 I have found that applications do indeed go into Unreachable status automatically when the server is down. However, when a switch is down, the interfaces are NOT automatically placed into Unreachable status.

Solarwinds? This seems abnormal. Comments?

Here is an image of what I am seeing using a manually defined dependency. Why are the interfaces on the down node not showing as "unreachable"?

Mine do show as unreachable on a down node:

SolarWinds? Is this a bug in Platform 2016?

Are they unreachable or can they still be reached somehow?  Sometimes having snmp working but icmp blocked can cause something that looks like that.