healthchecks widget responds with ENETUNREACH error #4845

ilium007 · 2025-02-27T22:21:07Z

ilium007
Feb 27, 2025

Description

API calls from healthchecks widget are not working, reporting ENETUNREACH error. Can successfully reach healthchecks.io API from within container using curl.

I am not using a HTTP proxy.

homepage version

v0.10.9 (b5f4daa, Jan 7, 2025)

Installation method

Docker

Configuration

services.yaml

 ---
 # For configuration options and examples, please see:
 # https://gethomepage.dev/latest/configs/services

 - Health Checks:
     - healthchecks.io:
         icon: healthchecks.png
         href: https://healthchecks.io/projects/xxxx/checks/
         description: healthchecks.io service checks
         widget:
             type: healthchecks
             url: https://healthchecks.io
             key: xxxx
             uuid: # optional

Container Logs

homepage  | [2025-02-27T22:15:06.721Z] error: <credentialedProxyHandler> HTTP Error 500 calling https://healthchecks.io/api/v3/checks/
homepage  | [2025-02-27T22:15:07.192Z] error: <httpProxy> Error calling https://healthchecks.io/api/v3/checks/...
homepage  | [2025-02-27T22:15:07.194Z] error: <httpProxy> [
homepage  |   500,
homepage  |   AggregateError
homepage  |       at internalConnectMultiple (node:net:1114:18)
homepage  |       at internalConnectMultiple (node:net:1173:5)
homepage  |       at Timeout.internalConnectMultipleTimeout (node:net:1632:3)
homepage  |       at listOnTimeout (node:internal/timers:571:11)
homepage  |       at process.processTimers (node:internal/timers:512:7) {
homepage  |     code: 'ENETUNREACH',
homepage  |     [errors]: [ [Error], [Error], [Error], [Error] ]
homepage  |   }
homepage  | ]
homepage  | [2025-02-27T22:15:07.196Z] error: <credentialedProxyHandler> HTTP Error 500 calling https://healthchecks.io/api/v3/checks/

Browser Logs

API Error: Unknown error
URL: https://healthchecks.io/api/v3/checks/
Raw Error:
{
    "code": "ENETUNREACH"
}

Troubleshooting

Installed curl in container and confirmed API call works fine

shamoon · 2025-02-27T23:30:23Z

shamoon
Feb 27, 2025
Maintainer

It has nothing to do with the credentialedProxyHandler, because it works fine. ENETUNREACH is very google-able, it is a network error.

As the template emphatically requests, please do all of the troubleshooting steps and post their output if you need further assistance.

3 replies

ilium007 Feb 27, 2025
Author

I have done what the template has asked me to do.

I have executed a curl request from within the Docker container - this works fine.

I have also executed the curl request from the host - it works fine.

What else am I supposed to check?

shamoon Feb 27, 2025
Maintainer

I have done what the template has asked me to do.

I mean, no you haven't? Here is the template, for emphasis:

and the steps linked to: https://gethomepage.dev/more/troubleshooting/#service-widget-errors , there are 4 of them.

And I had asked that you post the actual command and output.

It's possible that curl can still work and homepage cannot reach the url, but it is extremely rare. In general, as I said, this widget is not broken as is very easily demonstrated:

So the question remains what about your network or homepage container is preventing this because it is very unlikely there will be anything to be changed on the homepage side of things.

ilium007 Feb 28, 2025
Author

I mean, no you haven't? Here is the template, for emphasis:

I had added all the troubleshooting to the other unanswered "ENETUNREACH" discussion post. I have since deleted the post and created a new discussion. My bad.

ilium007 · 2025-02-27T23:58:53Z

ilium007
Feb 27, 2025
Author

All 4 troubleshooting steps completed. I have no other network connectivity issues with other Homepage widgets.

What other information can I supply?

services.yaml:

cat homepage/config/services.yaml
---
# For configuration options and examples, please see:
# https://gethomepage.dev/latest/configs/services

- Health Checks:
    - healthchecks:
        icon: healthchecks.png
        href: https://healthchecks.io/projects/xxxx/checks/
        description: healthchecks.io service checks
        widget:
            type: healthchecks
            url: https://healthchecks.io
            key: xxxx
            uuid: # optional

- Management Links:
    - IT Wiki:
        href: https://xxxx
        description: xxxx IT Documentation
        icon: wikijs
    - Proxmox Virtual Environment:
        href: https://pve.mgt.xxxx.com.au
        description: Proxmox Hypervisor
        icon: proxmox
        widget:
            type: proxmox
            url: https://192.168.100.10:8006
            username: api-ro@pam!homepage
            password: xxxx
            node: pve01
    - Proxmox Backup Server:
        href: https://pbs.mgt.xxxx.com.au
        description: Proxmox Backup
        icon: proxmox
        widget:
            type: proxmoxbackupserver
            url: https://192.168.100.11:8007
            username: api-ro@pbs!homepage
            password: xxxx
    - Unifi:
        href: https://unifi.mgt.xxxx.com.au
        description: Unifi Network Application
        icon: unifi
    - Apprise:
        href: https://apprise-api.mgt.xxxx.com.au
        description: Apprise API Configuration
        icon: apprise
    - Guacamole:
        href: https://guacamole.mgt.xxxx.com.au
        description: Apache Quacamole
        icon: guacamole
    - Wireguard Portal:
        href: https://wg-portal.mgt.xxxx.com.au
        description: Wireguard management portal (remote users)
        icon: wireguard

Confirmed all services with a widget DO have a unique name.

Docker container ping test:

xxxx on debian01 /data/docker                                                                                                                                                                                       10:52:40
➜ docker exec homepage ping healthchecks.io
PING healthchecks.io (188.40.122.95): 56 data bytes
64 bytes from 188.40.122.95: seq=0 ttl=47 time=305.637 ms
64 bytes from 188.40.122.95: seq=1 ttl=47 time=305.593 ms
64 bytes from 188.40.122.95: seq=2 ttl=47 time=305.729 ms
64 bytes from 188.40.122.95: seq=3 ttl=47 time=305.624 ms
^Ccontext canceled

Host curl request:

xxxx on debian01 /data/docker took 990ms                                                                                                                                                                            10:49:09
➜ curl --header "X-Api-Key: xxxx" https://healthchecks.io/api/v3/checks/
{"checks": [{"name": "OOL", "slug": "ool", "tags": "mikrotik", "desc": "", "grace": 120, "n_pings": 123, "status": "down", "started": false, "last_ping": "2025-02-27T22:37:53+00:00", "next_ping": null, "manual_resume": false, "methods": "", "subject": "", "subject_fail": "", "start_kw": "", "success_kw": "", "failure_kw": "", "filter_subject": false, "filter_body": false, "badge_url": "https://healthchecks.io/b/2/6524c006-94eb-4b52-bf55-de651c5cdd81.svg", "unique_key": "xxxx", "timeout": 60}, {"name": "RB5009", "slug": "rb5009", "tags": "mikrotik", "desc": "", "grace": 120, "n_pings": 268, "status": "up", "started": false, "last_ping": "2025-02-27T23:49:44+00:00", "next_ping": "2025-02-27T23:50:44+00:00", "manual_resume": false, "methods": "", "subject": "", "subject_fail": "", "start_kw": "", "success_kw": "", "failure_kw": "", "filter_subject": false, "filter_body": false, "badge_url": "https://healthchecks.io/b/2/9feb2652-6e8d-4793-8485-49bbcadcf6ca.svg", "unique_key": "xxxx", "timeout": 60}]}%

Container curl request:

xxxx on debian01 /data/docker took 990ms                                                                                                                                                                            10:51:00
➜ docker exec -it homepage sh
/app # curl --header "X-Api-Key: xxxx" https://healthchecks.io/api/v3/checks/
{"checks": [{"name": "OOL", "slug": "ool", "tags": "mikrotik", "desc": "", "grace": 120, "n_pings": 123, "status": "down", "started": false, "last_ping": "2025-02-27T22:37:53+00:00", "next_ping": null, "manual_resume": false, "methods": "", "subject": "", "subject_fail": "", "start_kw": "", "success_kw": "", "failure_kw": "", "filter_subject": false, "filter_body": false, "badge_url": "https://healthchecks.io/b/2/xxxx.svg", "unique_key": "xxxx", "timeout": 60}, {"name": "RB5009", "slug": "rb5009", "tags": "mikrotik", "desc": "", "grace": 120, "n_pings": 270, "status": "up", "started": false, "last_ping": "2025-02-27T23:50:44+00:00", "next_ping": "2025-02-27T23:51:44+00:00", "manual_resume": false, "methods": "", "subject": "", "subject_fail": "", "start_kw": "", "success_kw": "", "failure_kw": "", "filter_subject": false, "filter_body": false, "badge_url": "https://healthchecks.io/b/2/xxxx.svg", "unique_key": "xxxx", "timeout": 60}]}/app #

2 replies

shamoon Feb 28, 2025
Maintainer

Thanks. One thing I have seen help others with this error is to disable ipv6 in docker (I believe it cannot be disabled per-container).

    sysctls:
      - net.ipv6.conf.all.disable_ipv6=1

ilium007 Feb 28, 2025
Author

Ok - I will test but I use and rely on IPv6 in Docker

ilium007 · 2025-02-28T00:02:20Z

ilium007
Feb 28, 2025
Author

Removed all other service widgets and still errors:

0 replies

ilium007 · 2025-02-28T00:17:06Z

ilium007
Feb 28, 2025
Author

Disabling IPv6 in the docker-compose.yml file allows the widget to work.

docker-compose.yml (in case it helps someone else):

services:
  homepage:
    image: ghcr.io/gethomepage/homepage:latest
    container_name: homepage
    environment:
      - PUID=1000
      - PGID=1000
      #- LOG_LEVEL=debug
    volumes:
      - ./homepage/config:/app/config
    #ports:
    #  - 3000:3000
    networks:
      - fp-vlan100
    sysctls:
        - net.ipv6.conf.all.disable_ipv6=1
    restart: unless-stopped

networks:
  fp-vlan100:
    ipam:
      driver: default
      config:
        - subnet: 172.16.239.0/24

Is this worthy of a bug / issue request? I rely heavily on IPv6 across my networks due to ISP CGNat on IPv4.

1 reply

shamoon Feb 28, 2025
Maintainer

I think this is highly specific to some people's network / DNS / I dont know what setup.

Two things if youre willing to try:

Try the current dev image. FYI, it contains another required change if youre running via the public internet (run with env HOMEPAGE_ALLOWED_HOSTS=yourhost)
If that doesnt work, try the image tagged feature-disable-ipv6 with env HOMEPAGE_PROXY_DISABLE_IPV6=true

This comment was marked as off-topic.

Sign in to view

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

healthchecks widget responds with ENETUNREACH error #4845

{{title}}

{{editor}}'s edit

{{editor}}'s edit

Replies: 5 comments 6 replies

This comment was marked as off-topic.

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{editor}}'s edit

{{editor}}'s edit

{{title}}

Select a reply

healthchecks widget responds with ENETUNREACH error #4845

ilium007 Feb 27, 2025

Description

homepage version

Installation method

Configuration

Container Logs

Browser Logs

Troubleshooting

Replies: 5 comments · 6 replies

This comment was marked as off-topic.

shamoon Feb 27, 2025 Maintainer

ilium007 Feb 27, 2025 Author

shamoon Feb 27, 2025 Maintainer

ilium007 Feb 28, 2025 Author

ilium007 Feb 27, 2025 Author

shamoon Feb 28, 2025 Maintainer

ilium007 Feb 28, 2025 Author

ilium007 Feb 28, 2025 Author

ilium007 Feb 28, 2025 Author

shamoon Feb 28, 2025 Maintainer

ilium007
Feb 27, 2025

Replies: 5 comments 6 replies

shamoon
Feb 27, 2025
Maintainer

ilium007 Feb 27, 2025
Author

shamoon Feb 27, 2025
Maintainer

ilium007 Feb 28, 2025
Author

ilium007
Feb 27, 2025
Author

shamoon Feb 28, 2025
Maintainer

ilium007 Feb 28, 2025
Author

ilium007
Feb 28, 2025
Author

ilium007
Feb 28, 2025
Author

shamoon Feb 28, 2025
Maintainer