1 of 1 people found this helpful
age_count attribute is not available for clustered software nodes. As documented here
"A Cluster node can be destroyed either manually or automatically. Cluster nodes are automatically removed when the last remaining Host node or MFPart is unlinked. In some cases, particular patterns might not create a cluster until hosts are detected. In these cases, the pattern will remove the Cluster node when there is only one host linked to it. For vCenter cluster nodes, cluster nodes are removed when a VMware cluster no longer exists."
OK .. agreed with you Vaibhav..
However it may be my misteake pointing to the cluster to much . It may have shifted focus of this question.
Look into my query .. in fact I am looking for the Software Instance Age Count .. when it is "clustered"..
according to the BMC documentation: (Software Instance node - BMC Discovery 11.2 - BMC Documentation )
Not shown in the UI
The number of consecutive successful (positive) or failed (negative) times the
Software Instance has been seen during host scans.
.. therefore I have expected some real results ..
Am I right? .. or , as asked before, am I overlooking something obvious ..
Thanks a lot for your help
2 of 2 people found this helpful
You seem to have skipped the bit on the same page:
age_countattribute of the (first order) SI node contains information about when the SI node was last confirmed by its maintaining pattern.
A first order Software Instance is one which models a piece of software running directly on the OS.
Which is not the case - it is running on the cluster.
It does not want to go away because something like SQL Server is run in failover mode. Hence, assuming a two member cluster, it will be running on one member but not the other and you would not want the aging to go negative when everything is fine just because it is running on the other member of the cluster.
Hello Andrew ..
Yes, that makes sense. I have chilling suspicion that this MUST be an underlying logic here. As always, appreciate your clean cut here.
Andrew .. since we are on that subject. On the final note:
Should we presume then that if ANY software instance exists with a non-integer age_count then it should be treated the same as a SI with age_count > 0?
Best; Kris C