server error 500 on multiple endpoints

JayDee

Hi all

I’ve been using the dingz for quite a while without any issues.
However, for some time now (I think after I started using https://github.com/siku2/hass-dingz/releases), I’m starting to get an unresponsive dingz UI.

I can’t access the dingz via the browser as lots of requests timeout or 500. If I curl single api requests, some of them work, like:

curl --location -g -v --request GET  '192.168.1.13/api/v1/shade'
Note: Unnecessary use of -X or --request, GET is already inferred.
*   Trying 192.168.1.13:80...
* Connected to 192.168.1.13 (192.168.1.13) port 80
> GET /api/v1/shade HTTP/1.1
> Host: 192.168.1.13
> User-Agent: curl/8.7.1
> Accept: */*
>
* Request completely sent off
< HTTP/1.1 200 OK
< Pragma: no-cache
< Cache-Control: no-store, no-cache
< Content-Type: application/json
< Content-Length: 2
< Access-Control-Allow-Origin: *
<
* Connection #0 to host 192.168.1.13 left intact
{}

But others like:

curl --location -g -v --request GET  '192.168.1.13/api/v1/button_config'
Note: Unnecessary use of -X or --request, GET is already inferred.
*   Trying 192.168.1.13:80...
* Connected to 192.168.1.13 (192.168.1.13) port 80
> GET /api/v1/button_config HTTP/1.1
> Host: 192.168.1.13
> User-Agent: curl/8.7.1
> Accept: */*
>
* Request completely sent off
< HTTP/1.1 500 Internal Server Error
< Content-Type: text/html
< Content-Length: 0
< Access-Control-Allow-Origin: *
<
* Connection #0 to host 192.168.1.13 left intact

Don’t. These are also reflected in home assistant:

They recover after some time, usually a couple hours, but then usually within a day they return. They usually affect 1-4 out of 8 dingz, but all are affected at different times.

Has anyone else also experienced any of these 500 internal server error issues?

Any hints on how this could be debugged further? Is there a way to access the internal log of a dingz?
Thanks!
Joris

JayDee

Update:

After talking to the iolo (dingz) support team, they pointed me to this API with the following info: the free RAM should be above 50K in order for the dingz to run without issues. After checking mine (I created a script to check, I will post it below), most of them were below 20K.

Iolo support then pointed out, that mystrom, or homekit might take up some resources. After disabling mystrom (just removing them from mystro didn't help), the free ram shot up to around 50K. I also then went on and disabled the cloud (which was added after I installed them, and I don't use any cloud features), the RAM is now at >110K and the issue is solved.

#!/bin/bash
# This script reboots the Dingz devices using their API.

# List of devices with names and IPs
devices=(
    "foo:192.168.1.x"
    "bar:192.168.1.y"
)

# Flag to only check RAM
check_ram_only=false

# Function to display help
show_help() {
    echo "Usage: $0 [options]"
    echo "Options:"
    echo "  --check-ram-only   Only check the free RAM of Dingz devices without rebooting."
    echo "  --help             Display this help message."
}

# Parse arguments
while [[ "$#" -gt 0 ]]; do
    case $1 in
        --check-ram-only) check_ram_only=true ;;
        --help) show_help; exit 0 ;;
        *) echo "Unknown parameter: $1"; show_help; exit 1 ;;
    esac
    shift
done

# Print table header
printf "%-20s %-15s %-10s %-10s\n" "Device Name" "IP Address" "Free RAM" "Action"
printf "%-20s %-15s %-10s %-10s\n" "------------" "-----------" "--------" "------"

# Check the free RAM for each Dingz device
for device in "${devices[@]}"; do
    name="${device%%:*}"
    ip="${device##*:}"
    response=$(curl -s "http://$ip/api/v1/ram")
    free_ram=$(echo "$response" | jq -r '.free')
    action="None"

    if [[ "$free_ram" -lt 60000 ]]; then
        action="Low RAM"
    fi

    if [[ "$check_ram_only" == false && "$free_ram" -lt 60000 ]]; then
        action="Rebooting"
        curl -X POST "http://$ip/api/v1/reboot"
    fi

    printf "%-20s %-15s %-10s %-10s\n" "$name" "$ip" "$free_ram" "$action"
done