Too many Database connections

[ghunjan]

Current connection : 146

A value outside the normal range can indicate a heavy load on the DB, which can affect DB and overall Manager Performance  - Normal range: <90 

What I have done so far:

1. Update Max_Connection to 250 in the NSM_INSTALL_PATH/MySQL/my.ini

2. Verified Config of NSM Manager and MySQL is installed

Intel Xeon CPU E5-2697 v2 @ 2.7 GHz 

RAM - 32 GB

OS - Windows Server 2008 R2 Standard with SP1 - 64 bit

NSM Manager version - 9.1.7.75

Sensor model : M-4050

Sensor Software: 9.1.3.13

Any help will be greatly appreciated 

Thanks,

Gagandeep

[petermason]

Hey Gagandeep,

Is this for the same environment with 1 manager and 2 sensors?

Where are you seeing this error? Is it appearing in the Health Checks?

Are you seeing the error continuously or is it intermittent?

What volume of alerts are you generating per sensor and what is the size of your DB?

Have you run tuning on the DB recently?

Regards

Peter

[ghunjan]

Hello Peter,

Yes, this is for the same environment.

Correct, it is appearing on the Health Check.

I would say, I see it continuously.

DB size is 1 GB with largest Table: 394 MB

Alert Count - 234,421 - Oldest Alert: Mar 01 03:00:05 2019
Newest Alert: May 30 14:50:07 2019

DB is set to autotune Every Saturday for All Tables.

Last Tuning Time:  2019-Jun-01 03:01:59 GMT-04:00

Thanks,

Gagandeep

[petermason]

Hey Gagandeep,

I think I had this problem back when the Heath Checks were introduced to NSP, it seemed to be because I was running all of the health checks at the same time. 

If you just select the Database Connections check on it's own does it still give a high volume of connections?

Regards

Peter

[ghunjan]

That's interesting.

Indeed DB connection count came low when selecting only DB Connections. 

Thanks for the tip. 

Cheers

Gagandeep

[ghunjan]

Hi Peter,

I'm not sure if McAfee Developer/Engineering is aware about this issue.

Cheers

Gagandeep

[petermason · Accepted solution]

Hi Gagandeep,

I looked back through my old SR's and see that I reported this to them previously.

There was no solution to the SR at the time.

I normally just run thee health checks one group at a time, Summary, Database and Connectivity.

Regards

Peter

[ghunjan]

Thank you so much for that information.