Challenges chart not showing full node controller challenges

[hutchwilco]

I'm still having a problem with the challenges chart - in the entire 6-8 months of getting to know Machinaris, I've had no challenges shown at all ( prior to 0.6.9 or 0.6.8 I would get a blank empty challenges chart - after those versions I just get no chart at all).

Recently I installed my first separate harvester - as soon as I got it working, the Challenges chart appeared on the main dashboard but only shows challenges for that separate harvester, it doesn't show any challenges for the fullnode/controller's harvester.

[guydavis]

Sorry to hear that, please try:

docker exec -it machinaris bash
cd /root/.chia/mainnet/log
grep --text -i eligible debug.log.1 debug.log

You should see output like:

This is parsed and inserted into the Machinaris database for display.

To troubleshoot further, please provide:

1. A full screenshot of your Workers page.
2. A full screenshot of the Chia fullnode Worker page (clicked into that row from Workers page).
3. A full screenshot of your Summary page.
4. A full screenshot of your Index page (Machinaris link in top left)
5. A complete copy of your apisrv.log
6. A complete series of screenshots showing your Machinaris fullnode Unraid Docker configuration. Or docker-compose.yml if using that instead.

Thanks!

[hutchwilco]

Thanks Guy! see info below.
The debug.log and ...log.1 output for 'eligible' show successful proofs found ( this is what's confusing to me about this issue. I haven't changed any ports from standard that I know of, timezones are all the same...).

docker-compose.txt
apisrv.log

[guydavis]

Thanks for all those details. Please the following in your docker-compose.yml for the fullnode. Change the hostname: field to something shorter without hyphens. Example:

version: '3.9'
services:
    machinaris:
        image: ghcr.io/guydavis/machinaris
        container_name: machinaris
        hostname: chiafullnode1
        restart: always
...

No need to change the actual hostname of the machine itself, just the hostname value passed into the container when Docker launches.

Then after 15 minutes once restarted, go to the Workers page, and select the stale row with hostname gridshare-ONDA-Chia-D32H-D4-V1-01. Select that stale row and click 'Prune Selected'.

After a bit longer, check your index page again. Hope this helps!

[hutchwilco]

🤦🏻‍♂️ It's always the simplest, dumbest answer, hiding in plain sight isn't it!
I never liked that silly long hostname anyway - someone else set it and I never got round to updating it!

It's worth noting that I didn't have to Prune the old worker (I did remove the container before restarting - not sure if that was what you intended) - anyway, it has automagically updated the worker hostname and the graph is now displaying both harvester and fullnode.
Thanks!