Before installing Linux my hard disk was parted as follows:
500MB (NTFS) reserved by system
100GB (NTFS) primary partition
100GB (NTFS) primary partition
152.5GB free space

How i configured volumes during installation of Linux:
1) I created logical partition a 2.5 GB (ext4) and configured the mount point "/boot";
2) From the entire free space (150GB) i created an encrypted partition;
3) Using LVM i created a volume group "M_V_G" from an encrypted partition;
4) In this group i created logical volumes and assigned the mount points "/root", "/swap" and "/home";

Installation was successful, system works fine but... after selecting "Kali" in GRUB i see the following:
WARNING: Failed to connect to lvmetad. Falling back to device scanning.
Volume group "M_V_G" not found
Cannot process volume group M_V_G
Please unlock disk sda6_crypt: *I entered password*
WARNING: Failed to connect lvmetad. Falling back to device scanning.
Reading all physical volumes. This my take a while...
Found volume group "M_V_G" using metadata type lvm2
WARNING: Failed to connect to lvmetad. Falling back to device scanning.
3 logical volume(s)in volume group "M_V_G" now active
Then system is loaded

And so... the question: can i get rid of WARNINGs?

Check if you have enabled service lvmetad. Enable it if not.
Check why it does not run if fails.
If nothing helps, disable use_lvmetad in /etc/lvm/lvm.conf
I still not understand the benefits of using lvmetad, everything works well even without it.

any reason why you made boot bigger as 128MB?

do you need swap?

keeping home and root separated is up to discussion. i do not do that. the free space is wasted and needed usually on the other partitions. just bad practise.

naming lvm containers in rubbish is not good practise, name it properly, so you have an easier life later.

why do you kali=

Sure. adapt the boot process

I also made a custom init script. the gentoo shipped one has after many years the same faults. i saw it because i saw guys asking the same questions. i did not bohter reporting back, as 90 percent of my bugs on bugs.gentoo.org or bugs.kernel.org are not read / done something to fix it.

it is expected and usual to write / adapt your own init script when using luks.

and dealing with binary distros with luks is something i would not recommend

--

sorry to write this, but it seems you did not understood the principle of lvm and luks. i would not call something sda6_crypt

Normally, each LVM command issues a disk scan to find all relevant physical volumes and to read volume group metadata. However, if the metadata daemon is running and enabled, this expensive scan can be skipped ... This can save a significant amount of I/O and reduce the time required to complete LVM operations, particularly on systems with many disks.


Disabling lvmetad solves the problem, but affects the performance of the commands (I don't know how important it is for me now). If I'll do not find anything better - I use this method, thanks

Did according to the guide (there it was stated that is necessary 1.5Gb, since it is not important for me and I can afford 2.5Gb - made so there were no problems)

Perhaps in the future for hibernation.
Again, the free space in this case is not a problem

For example?

Because of the tools that will be useful in system administration

Can you share the materials/links that will help me in adapting the init script?

Because you are not at the stage of studying the above