This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Deleted Interface Still Appear on Node Child Status

FormerMember
FormerMember

Hi,

We ran a script to bulk delete some interfaces that were in a down status, I have run the configuration wizard and verified they are no longer in the managed rescources section but the parent node of the interfaces still have the child status warning for these interfaces, any ideas?

  • Good to know how the script looks like. Is that Orion API based script using CRUD operations (supported way as it handles quite a lot of operations like notifications for services, moving to deletedInterfaces for DB maintenance, deleting pollers etc...) or just "delete from interfaces where" sql script?

  • FormerMember
    0 FormerMember in reply to Diner

    It was a simple delete from interfaces where sql script, I was told this wouldn't be an issue...

  • That most probably IS the issue. Please check the DB table NodeChildStatus. Simple deleting without running additional related procedures causes leaving a lot of orphaned data in the DB(pollers, statistics, ...). To correct at least the NodeChildStatus issue you can truncate that table. This will be refreshed in few minutes.

  • FormerMember
    0 FormerMember in reply to Diner

    Unfortuently this doesn't seem to have helped as the child status still appears, oddly if I unmanage and then remanage the devices it seems to fix the issue...

  • Yes, this should regenerate that as well. DB maintenance (by default running nightly)  executes stored procedures, deleting the orphaned data, but I'm not sure if it covers all the cases for this way of deletion.

  • What you see as Child Status are value from Nodes table (StatusDescription, GroupStatus,etc.). This values are updated when the interface is added/removed via UI or when status of some monitored interface on node is changed. That's the reason why unmanage/remanage has fixed the child status.

    To recompute child status run this SQL query (use the specific NodeID instead of @nodeId)

    execute NPM_RecomputeChildStatus @nodeId

  • I ran into a similar issue where I deleted interfaces in bulk using SQL. To resolve the Node status, I just "UnManaged" and then "Manage Again" my group of nodes. Be cautious though, this could trigger some alerts depending on how they are configured.

  • the information/busness service logic distributes infomation about polling, etc amongst the processes and internal databases in Solarwinds. deleteing tings directly out of the database can be fraught with unexpected challenges.

    Tip: When I want to bulk-delete things out of the SQL database I don't actually delete them using SQL, instead I use SQL to update a custom property ( typically the 'comment' custom property)

    Then, I use that property in the webUI to select multiple items and delete them.

    This ensures that various things get updated properly, and also, when Solarwinds implements an undelete function I'll have a chance to test it out.