PBSE_NODE_BAD_CURRENT_AOE should not cause node deletion at server startup

Description

Hello,

We recently had pbs_server print out a bunch of these at startup:

02/24/2015 21:09:18;0001;Server@xyz;Svr;Server@xyz;No such file or directory (2) in remove_node_topology, unlink of topology file for node xxxxxx failed
02/24/2015 21:09:18;0001;Server@xyz;Svr;Server@xyz;setup_nodes, could not create node 'xxxxxx', error = 15149

It was our own fault that we had several nodes with current_aoe set to a value that wasn\'t in resources_available.aoe, but error 15149 (PBSE_NODE_BAD_CURRENT_AOE) shouldn\'t be reason enough to delete a node at server startup. Please change the server to instead unset current_aoe for the offending node (and log accordingly).

Please provide a code diff when this change is made - I\'ll backport it to whatever version we have running. Or even a simple outline of which functions to change and how could be fine.

-Customer

Acceptance Criteria

None

Status

Assignee

Unassigned

Reporter

Former user

Severity

None

OS

None

Start Date

None

Pull Request URL

None

Components

Priority

High
Configure