Receiving alert from Exchange_monitor Probe, without seeing a problem on that server, Niscache is abnormally large.

Products

• DX Unified Infrastructure Management (Nimsoft / UIM)

Issue/Introduction

• In this case below alert from exchange probe was continuously received, however there were no issues from the services running on exchange server. 

Alert:- (services_check): Find services status - communication error, Nimsoft Alarm from: server_name 

 

• During research we noticed in the exchange_monitor log that services_check started but failed to lookup ntservices probe:

 exchange_monitor: services_check... 

... and failed with

 exchange_monitor: nimNamedSession: failed lookup ntservices (2) 

• Similar messages were showing up also for ntevl, processes and perfmon probes.
• Lookups into the controller start to fail when the alarm is seen.
• On checking the niscache directory we saw >100,000 files.
• A very large niscache directory can adversely impact the robot operation/connection anytime information is retrieved from it. 

Environment

UIM version - 8.4.0 Database used : Microsoft SQL exchange probe 5.20 Exchange Server 2010 Perfmon Ver:- 1.52 Ntservices Ver:- 3.24 Ntevl Ver:- 4.22 Processes Ver:- 3.75

Cause

• The errors are due to the abnormally large niscache.
• The large niscache is caused by the perfmon probe

Resolution

• In such cases it is recommended to clear the niscache on that robot.
• See KB000044669 How to clear niscache on an individual robot
• This will restore the functionality for the probes
• To prevent future recurrence of this issue upgrade the perfmon probe to 1.63 or higher

Additional Information

There is a KB Article KB000034717 niscache - what it is, what it does