Cdm probe question on code used to monitor disk with cdm

Products

• DX Unified Infrastructure Management (Nimsoft / UIM)

Issue/Introduction

Request for information on:

1)Share connectivity is checked by attempting to map the drive, using netAddConnection2.
netAddConnection2 is Windows based cmd. How does CA check filesystem availability on Linux side?

2)Linux - The list of mounted filesystems is found using the system call getmntent().
This is linux based function to find mounted drives only. (Get Mount Entries)
The getmntent() and getmntent_r() functions return a pointer to the mntent structure or NULL on failure. – what is the failure?

3)The FS alert received has the following setup – check interval is every 10 min , 2 sample failures.
What operation/process tells it was a failure?
When it stated as failure?
Is there any timeout value in which response should be retrieved?
Is alert being sent when there was 2 failures straight one after another without any pause between re-tries? Or 2 failures out of two in 20min?

 

Environment

Release : 9.2.0

Component : UIM - CDM WITH IOSTAT

Cause

Request for Information

Resolution

1)Q: Share connectivity is checked by attempting to map the drive, using netAddConnection2.
netAddConnection2 is Windows based cmd. How does CA check filesystem availability on Linux side?

A For Windows, this is the API we use to map a remote drive or network folder to monitor by CDM. Now we have implemented a new enhancement to auto detect all the mapped drive in the system and start monitoring them.

For Linux we detect all the mount points and monitor them.

2)Q: Linux - The list of mounted filesystems is found using the system call getmntent().
This is linux based function to find mounted drives only. (Get Mount Entries)
The getmntent() and getmntent_r() functions return a pointer to the mntent structure or NULL on failure. – what is the failure?  

A There  are multiple reasons possible of failure, some might not about the CDM, sort of generic failure for system.

Access right, do not have privilege to call API and many more.

3)Q:The FS alert received has the following setup – check interval is every 10 min , 2 sample failures.
What operation/process tells it was a failure?
When it stated as failure?
Is there any timeout value in which response should be retrieved?
Is alert being sent when there was 2 failures straight one after another without any pause between re-tries? Or 2 failures out of two in 20min?

A: If  we found a FS being monitored by CDM is unmounted, then we receive an Alert.

Number of Samples are averaged at each poll interval. It is applicable for measurable data like Disk Usage etc. In case of missing alert (events), There is no averaging number of configured samples. So reported at first instance.

Every poll interval we check availability of FS if it is configured to be monitored by CDM.