Unable to connect to the admin dashboard

[Guillaume Rossolini]

Hi,
I am using a rpi zero w with the latest buster lite.
Your latest commit allowed the setup to finish without any errors:

pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ docker-compose up -d
Creating network "pi-hole-monitoring_influxdb" with the default driver
Creating influxdb ... done
Creating sql_influx ... done
Creating grafana    ... done
Creating telegraf   ... done

But I can't access the web interface and neither can I curl it:

pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ sudo netstat -peanut | grep LISTEN | grep 3000
tcp6       0      0 :::3000                 :::*                    LISTEN      0          178138     12125/docker-proxy  
pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ curl -I http://127.0.0.1:3000
curl: (56) Recv failure: Connection reset by peer
pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ sudo netstat -peanut | grep LISTEN | grep 3000
tcp6       0      0 :::3000                 :::*                    LISTEN      0          178138     12125/docker-proxy

After a few minutes, netstat reports that nothing is listening on port 3000.
I can docker-compose down && docker-compose up -d again, but to the same effect.

Please let me know how I can provide more info.

[Brandon McFarlin]

Can you run docker-compose up without the -d and see what the output says?

[Guillaume Rossolini]

Sure, here you go:

pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ git status
On branch master
Your branch is up to date with 'origin/master'.

nothing to commit, working tree clean
pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ docker-compose down
Stopping sql_influx ... done
Stopping grafana    ... done
Stopping telegraf   ... done
Stopping influxdb   ... done
Removing sql_influx ... done
Removing grafana    ... done
Removing telegraf   ... done
Removing influxdb   ... done
Removing network pi-hole-monitoring_influxdb
pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ docker-compose up
Creating network "pi-hole-monitoring_influxdb" with the default driver
Creating influxdb ... done
Creating sql_influx ... done
Creating telegraf   ... done
Creating grafana    ... done
Attaching to influxdb, sql_influx, grafana, telegraf
influxdb exited with code 139

[Brandon McFarlin]

Looks like the pi zero might not have enough memory to start the stack. Try running
influx by itself with docker-compose run influxdb bash. If that runs and gives a shell, then the pi zero is failing because it runs out of memory. If not, let me know.

[Guillaume Rossolini]

I'm not sure how to tell (this is after a reboot):

pi@raspberrypi-gr:~ $ free -m
              total        used        free      shared  buff/cache   available
Mem:            432         102         178           5         151         276
Swap:            99           4          95
pi@raspberrypi-gr:~ $ cd Pi-Hole-Monitoring/
pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ docker-compose run influxdb bash
pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ 
pi@raspberrypi-gr:~/Pi-Hole-Monitoring $ free -m
              total        used        free      shared  buff/cache   available
Mem:            432         134         133           9         165         239
Swap:            99          19          80

[Brandon McFarlin]

Your memory is fine. It's a buster lite issue. Someone else had the same issue on DietPi which essentially is Stretch lite. The lite version of the OS does not come with everything required to start this.