Monitoring strategy

[cult]

Hy everyone,

I need help to setup a monitoring strategy to find the bootle neck of my problem. I use Proxmox for my home server and when i watch a movie (4k hdr) it start to stuttering after 30 to 60 minutes.

Here my setup :

1g Network end to end cable

1 VM with ubuntu server 20.04 and Open Media Vault
8G RAM, 1 HDD 7TB Share with SMB (EXT4)

Android Nvidia Shield 2017, using Kodi latess version

I test my network with iperf3 and everything look fine. I will try to boost my VM RAM to 16g tonight, but it's strange that my friend use synology with 4g RAM and it butter smooth.

That been said, my question is how can i setup a monitoring strategy on my stack to find the problem ?
When i look the RAM with free -h, it's looked fine. when i look the Proxmox dashboard, the VM eat all the RAM. So how can i monitoring the system in order to see when the machine is struggling and what part is the bootle neck ? I dont thing the CPU is the problem because it under 2% most of the time for all the server, but i dont whan to exclude it.

Thanks in advance and sorry for my poor English writing.

 

[Dunuin]

First you should try to watch that movie from another machine to verify that your virtualized OMV is the problem and not your Shield.
Then you didn't told us how you are watching the movie. DLNA? Plex/Jellyfin/Emby app or webui? If you just open the file over the network instead of using a streaming protocol/software how do you access it (FTP/SMB/NFS)?

I like running htop and iostat 60 2on my host and inside the guest to see thats going on. Both need to be installed first ("htop" and "sysstat" packages).