lvmetad swap memory issue on CentOS 7.6

[rstar]

I have a VM system for DB(postgresql).

The system information as below.

type : VM
CPU : 16 Core, intel Xeon Gold 6240 @ 2.6Ghz
Memory : 16GB (vm.swappiness = 1)
OS : Centos 7.6 1810 64bit


This VM have had around 50% swap memory for over 3 months,but it is incresing rapidly nowadays.(incresed over 20% for 2 weeks)
All parameter of a DB like session,age, etc are normal and any special event is not exist.
In addition, most of Swap memory are on lvmetad process(870m/1860m) in the TOP command.
In other VMs, I cannot see such a large swap memory on lvmetad.

In this situation, I have question.
1. Is it related of DB and lvmetad?
2. Why swap memory of lvmetad is incresing? or I want to know check point

[TrevorH]

7.6 has been out of support for more than 4 years. Update to 7.9 which has many fixes and closes many security vulnerabilities. There are over 13,000 lines in the rpm changelog from kernel-3.10..0-957.el7 to the current kernel-3.10.0-1160.95.1.el7.x86_64. Any of those 13,000 entries could be your problem and you are missing them all.

yum update