Integrate inspector dashboard json

[gsxryan]

[gsxryan]

Append -p 7777:7777 to the run command

[JeromeGillard]

Any idea where this /app/config/inspector might be?

[gsxryan]

It's inside the storagenode container.

[JeromeGillard]

They seem to build the GO inspector from Storj source code. I guess they run it in the SNO container by binding the inspector binary from the host machine. (Correct me if I am wrong, that's just how I would do that).

As my Telegraf container doesn't have golang binaries, I can't do that and keep being portable with official SNO and Telegraf Docker images.

I've heard the SNO dashboard (web interface) will be issued soon. I hope we will be able to get more easily many information we strive to compute.

I would wait for that release.

[gsxryan]

I imagined capturing within the container, and dumping it somewhere outside the container for access. I do agree with you, this should be tabled until checking out the SNO board.

[gsxryan]

I think the most critical attribute to follow here would be the "Time since last online" metric. Some SNOs report that they get high values here, and the only way to fix it is to reboot the node. If we can monitor this, we could schedule proactive reboots to remedy it, and alerts to monitor how often it occurs.

Alternatively, it likely goes along with low throughput, so that could be monitored instead.

[gsxryan]

Monkit: https://forum.storj.io/t/guide-to-debug-my-storage-node/1372