How to setup OS level metrics on remote monitored db

[tomskopek]

Hi there,

I have two digital ocean droplets, one is running pgwatch2 via docker image, and this is connected to the other which is running the production db.

I'd like to get the DATADIR metric showing:

As I understand, this relies on the get_psutil_disk wrapper, and I'm struggling to understand the documentation recommended approach for getting this set up.

1. From my understanding, I am supposed to run these instructions on the monitored db, is that correct?

2. The production db droplet does not have pgwatch2 installed, therefore it does not have the /etc/pgwatch2/metrics/00_helpers/... scripts. What is the recommended approach? Should I just git clone that 00_helpers folder, and run them?

3. As per these instructions I also tried the “Auto-create helpers” checkbox on the pgwatch2 Web UI. I tried checking it, and updating the existing connection, and then I tried deleting the connection and recreating it with it checked. The user I used was the pgwatch2 user created as explained here. Neither attempts seemed to work, as the DATADIR disk space left is still not working.

I would be grateful for any guidance on how to get this working!