Skip to content

webservice to decommission nodes is broken #1345

Description

@jonstout

We're currently relying on update_cache to propagate node turn down events to fwdctl and discovery.

NSO discovery at least needs a notification of this event to stop tracking changes for the decom'd node. NSO fwdctl's update_cache doesn't current read in node state changes.

Database code that gets called is busted:
https://github.com/GlobalNOC/OESS/blob/admin-ui-refresh/perl-lib/OESS/lib/OESS/Database.pm#L5144-L5192

  • Create and publish rabbitmq events to both fwdctl and discovery for nso
  • Update both fwdctl and discovery cached node hash for nso
  • Replace busted Database::decom_node logic (see admin-ui-refresh branch for correct logic)

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions