NAS queue backed up with "Command 'hubpost_bulk' not in command-list" message in logs

Products

• DX Unified Infrastructure Management (Nimsoft / UIM)

Issue/Introduction

NAS queue backed up and stops processing alarms and show a message in the logs" Command hubpost_bulk not in command-list in probe logs.

Problem is also seen post hub upgrade to 7.x, where some probes stop working

Example from Logmon profile using queue mode:
Apr 13 20:05:57:465 [1992] logmon: RREQUEST: hubpost_bulk <-10.10.15.31/48002  h=236 d=9520 
Apr 13 20:05:57:465 [1992] logmon:  data   bulk_size=1 bulk_list=PPDS(9467) qsize=1 
Apr 13 20:05:57:465 [1992] logmon: Command 'hubpost_bulk' not in command-list 
Apr 13 20:05:57:465 [1992] logmon: SREPLY: status = 11(command not found) ->10.10.15.31/48002 

Example from the nas probe logs
Jul 30 12:23:14:097 [8924] nas: RREQUEST: hubpost_bulk <-10.10.10.91/48002 h=271 d=95007 
Jul 30 12:23:14:097 [8924] nas: Command 'hubpost_bulk' not in command-list 
Jul 30 12:23:14:097 [8924] nas: SREPLY: status = 11(command not found) ->10.10.10.91/48002 
Jul 30 12:23:14:199 [8924] nas: RREQUEST: hubpost_bulk <-10.10.10.91/48002 h=271 d=95007 
Jul 30 12:23:14:199 [8924] nas: Command 'hubpost_bulk' not in command-list 
Jul 30 12:23:14:199 [8924] nas: SREPLY: status = 11(command not found) ->10.10.10.91/48002 
 

Environment

Release:
Component: UIMHUB

Cause

This is happening as Hub 7.x tries to disallow a bulk size of 1 and imposes a "floor" of 100 whenever it thinks the consuming probe supports that.
Please upgrade NAS to latest supported version & hotifx level as well.

e.g., logmon was trying to make a connection to the hub. Because the hub was trying to force the connection to use a minimum bulk size of 100 it was causing a problem. By changing the hub to allow it to default back to 1 resolved the issue.

Resolution

To resolve this issue two new keys need to be added to the hub in raw config:

bulk_size_floor = 1

default_bulk_size = 1