Insane Node Memory & CPU Consumption

Could you please show the result of the command

docker ps

and

docker stats

This is the docker ps output:

and this is the docker stats output:
Schermata 2020-03-23 alle 09.01.48

Please, remove all watchtowers:

But the main issue seems that your storagenode doing a lot of work to move the deleted data to the trash.
Please, check the disk and network activity too. You may see almost zero network activity and high disk activity.

I’m running my node on QNAP using a VM connected via ISCSI where run docker, the network trash on my QNAP is switched off so I can’t understand how can mode the data to the trash.

Is it possible move the container directly on my QNAP container without using a VM ?

How I can connect my folder via ISCSI directly from container ?

thx

The storagenode have an own trash folder. It moves deleted data for the time when your node was offline or unable to delete pieces immediately. The deleted data will be in trash for the next 7 days.

The iSCSI will always consume more CPU and RAM and will be slower than a local connected storage.
If your QNAP is able to run a docker, you can run a container on your QNAP instead of VM.

Also, you can try to use the native app for QNAP:

Yes I can run the docker directly from QNAP but all my data are on ISCSI folder and I don’t know how can attache the ISCSI directly on a container, should I have to rsyc all data on a folder and the run the container from that ?

If you have enough space on your NAS, you can rsync the iSCSI storage to the folder on your NAS several times while your storagenode is running, then stop and remove the storagenode and rsync one more time with a --delete parameter, then run storagenode container directly on your NAS with correct paths to the identity and data.
You should specify the data path to the folder with storage folder inside (do not select the storage as folder for data), otherwise your node will start from scratch and will be disqualified.