🚧 Pretty server's error pages in the docker image & git repository (for traefik, k8s, nginx and so on)
Home Page: https://tarampampam.github.io/error-pages/
License: MIT License
Adding middleware to a service or middleware chain beaks whenever the error pages services is down.
To reproduce add the error pages middleware to a service and than stop the error pages service and try to access the other service.
There should be another way to catch this error pages but without breaking other services if error pages service goes down.
I want to only enable SHOW_DETAILS when the request comes from a specific ip range.
Ex. Customers don't really need to know where the request is going inside the k8s cluster. But for debugging, it would be beneficial.
Add an environment variable to allow source-ip ranges for SHOW_DETAILS.
No response
We are trying to use the image on a K8S only IPv6.
I was not successful when using env LISTEN_ADDR, with "[::]" and "::".
Log:
listen tcp4: address :::8080: too many colons in address
No response
No response
It would be a nice option to have and let generator pick random template from available ones? Or maybe specifying more than one in TEMPLATE_NAME and let it randomly pick from preselected templates.
I'm libra and can't decide between ghost and shuffle ;)
Heya,
I thought I'd update my fork to the new version that you've released (doing a test on a different repo before committing to my actual fork rather than creating a branch for now) to keep it up to date and to add a couple more templates, and it seems that the github actions fail on build with the error:
stat /home/runner/work/error-page-test/error-page-test/cmd/error-pages: directory not found
https://github.com/modem7/error-page-test/runs/4361391755?check_suite_focus=true
Is this something you'd be willing to assist with or would this be something outside your remit?
Many thanks!
Use case:
I require 404 status codes to be handled by the application instead of error-pages. To do so I now have the following middleware setup in Traefik config.
http:
middlewares:
errorPages:
errors:
# 404 excluded
status: 400-403,405-499,501,502,504-599
service: error-pages@docker
query: /{status}.htmlError-pages is setup with docker compose:
version: '3.9'
services:
error-pages:
image: tarampampam/error-pages:${ERROR_PAGES_IMAGE_VERSION} # 2.21
security_opt:
- no-new-privileges:true
restart: always
environment:
TEMPLATE_NAME: ${ERROR_PAGES_TEMPLATE_NAME} # cats
SHOW_DETAILS: ${ERROR_PAGES_SHOW_DETAILS} # true
labels:
traefik.enable: true
traefik.docker.network: proxy
traefik.http.routers.error-pages.tls: true
traefik.http.routers.error-pages.rule: HostRegexp(`{host:.+}`)
traefik.http.routers.error-pages.priority: 10
traefik.http.routers.error-pages.entrypoints: websecure
traefik.http.routers.error-pages.service: error-pages
traefik.http.services.error-pages.loadbalancer.server.port: 8080
networks:
- error-pages
- proxyNow the following currently happens:
example.com is up and a user requests a non-existing page, e.g. example.com/test, the user will see the application's specific error page.example.com is down it shows the 404 (cats) error-pages when requesting example.com. But requesting a non-existing page, e.g. example.com/test, the user will now see the following message:Wrong request URL. Error pages are available at the following URLs: /{code}.htmlWhat I would like to see is that it defaults to the default error page (cats 404) instead of this message.
Possible options:
404.html page when at the notfound handler.DEFAULT_ERROR_PAGE / DefaultErrorPage variable and use it at the notfound handler.Note: I'm not familiar with Go.
Hi!
Now the service can automatically determine the locale settings and install the language, I have a question, is it possible to set the desired language through a variable to turn off this check?
For example, I need only English, how can I do it now? Thank u in advance!
Add environment variable for set up language of pages.
No response
In the old version the http status code from a not found site was 404 and now it's 200 (of course with the text being '404 not found').
I'm not sure if this is the correct behaviour in web, because even if we serve a custom status page, I think the status code should still be "the error" (e.g. 404).
Can someone tell me what I did wrong or if this is just the new behaviour?
If this is not the expected outcome, feel free to ask me any system details that are needed, as I'm not sure what is relevant. some listed here:
traefik:v2.5tarampampam/error-pages:2.2.0
PS: Thanks for rewriting this cool project in golang. ✨
Heya,
Out of curiosity, how are you generating the gifs in the readme file?
Thanks!
Non-numeric users don't go well with the pod security policies of kubernetes
/
//This might be useful for futher reference: https://github.com/CollaboraOnline/online/blob/master/docker/from-source/Debian
The container stays unhealthy because the healthcheck fails. This seems to be caused by the change
- Flags --verbose, --debug and --log-json are deprecated, use --log-level and --log-format instead #163
since the healthcheck is still
HEALTHCHECK &{["CMD" "/bin/error-pages" "--log-json" "healthcheck"] "7s" "2s" "0s" '\x00'}
No response
No response
No response
No response
Allow even more privacy friendliness.
It would be great if there was an option to bundle all required resources in the Docker image itself.
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>{{ code }} - {{ message }}</title>
<style>
body {
background-color: #222;
color: #aaa;
font-family: 'Hack', monospace;
}
#error_text {
position: absolute;
top: 50%;
left: 0;
right: 0;
text-align: center;
margin-top: -35px;
height: 30px;
font-size: 2em;
}
</style>
</head>
<body>
<div id="error_text">{{ code }}: {{ message }}</div>
<script>
'use strict';
const setCharAt = function (str, index, chr) {
return (index > str.length - 1)
? str
: str.substr(0, index) + chr + str.substr(index + 1);
};
const $errorText = document.getElementById('error_text');
const text = $errorText.innerText;
let progress = 0;
const scrambleInterval = window.setInterval(() => {
const characters = 'ABCDEFGHIJKLMNOPQRSTUVWXYZ1234567890-=+<>,./?[{()}]!@#$%^&*~`\|'.split('');
let message = text;
for (let i = 0; i < text.length; i++) {
if (i >= progress) {
message = setCharAt(message, i, characters[Math.round(Math.random() * (characters.length - 1))]);
}
}
$errorText.innerText = message;
}, 800 / 60);
window.setTimeout(() => {
let revealInterval = window.setInterval(() => {
if (progress < text.length) {
progress++;
} else {
clearInterval(revealInterval);
clearInterval(scrambleInterval);
}
}, 70);
}, 500);
</script>
</body>
</html>Looks like:
Dockerhub rate limits are a pita, it would be cool if you could push this image to the github container registry as well. It's not really hard to do.
Thanks!
The current translation (l10n) works fine but it bothers me that the l10n.js is added at the bottom of the page, async and deferred. My connection is not slow and I can see the page blinking
I would love to implement this, but I'm a newbie with go
I think a very simple solution would be:
On startup we read the translations, which are already in l10n.js, and store it in a hash
On errorpage.go we initialize the props as usual (props.Message = page.Message()), after this we check the Accept-Header for the language codes, sort by weight and get the first that matches one of the available languages, if the language is not detected or is en we just skip the translation, if different than en we call a function that just returns the value for the key message, just as the l10n already does. Then just for caching we add the lang-code to props
My pseudo code would be something like this:
LiveTranslation.parse('translations.json')
if page, exists := cfg.Pages[pageCode]; exists {
props.Message = page.Message()
props.Description = page.Description()
} else if c, err := strconv.Atoi(pageCode); err == nil {
if s := fasthttp.StatusMessage(c); s != "Unknown Status Code" { // as a fallback
props.Message = s
}
}
selected_lang = LiveTranslation.parse_and_match( string(ctx.Request.Header.Peek("Accept-Language")) )
if selected_lang != "en" && selected_lang != null {
for name, value in props {
props .name = LiveTranslation.translate(value, selected_lang)
}
props.Lang = selected_lang
}I noticed cpu spikes on one of my VM machines, and figured with: docker stats --format "table {{.Name}}t{{.CPUPerc}}t{{.MemUsage}}" that like every 5-10 seconds error-pages the cpu usage spikes to 8-16% while my other containers chill at ~1%.
traefikt0.00%t24.01MiB / 1.929GiB
error-pagest17.05%t2.922MiB / 1.929GiB
adguardhomet0.02%t45.77MiB / 1.929GiB
portainert0.00%t8.27MiB / 1.929GiB
dozzlet0.00%t13.95MiB / 1.929GiB
sp_watchtowert0.02%t2.418MiB / 1.929GiB
apachet0.01%t5.945MiB / 1.929GiB
sp_traefikt0.01%t9.016MiB / 1.929GiB
autheliat0.00%t21.08MiB / 1.929GiB
sp_portainert0.01%t2.32MiB / 1.929GiB
whoamit0.00%t1MiB / 1.929GiB
traefik-bouncert0.00%t18.5MiB / 1.929GiB
honestly no idea why, I closed all http/s connections to the server and still had this issue.
No response
error-pages:
<<: *common-keys-alwaysup # See EXTENSION FIELDS at the top
image: tarampampam/error-pages:2.20.0 # Using the latest tag is highly discouraged. Please, use tags in X.Y.Z format
container_name: error-pages
command:
- "--log-level=error"
- "serve"
networks:
- traefik
volumes:
- ${APPDATA_DIR:?err}/error-pages/mytemplates:/opt/templates:ro # my templates
environment:
# TEMPLATE_PATH: /mytemplates/matrix.html
TEMPLATE_NAME: matrix # set the error pages template
SHOW_DETAILS: true
labels:
traefik.enable: true
traefik.http.services.error-pages.loadbalancer.server.port: 8080
com.centurylinklabs.watchtower.monitor-only: trueNo response
maybe you can tell me, what could cause this
thanks a lot and I really like this project
I want to dynamically set the document.domain in my custom template based on the mode I am in. I use different domains for development and production mode. Currently this is the solution I have:
<script>
{{ if show_details }}
document.domain = 'local.mydomain.com';
{{ else }}
document.domain = 'mydomain.com';
{{ end }}
</script>This is not perfect since I actually have 3 states: development, staging and production, each with their own domain. The current solution just handles development and production with the help of the show_details flag since this is turned off in production I can use it to set the domain accordingly.
I searched existing issues and also had a peak inside the source code but did not find anything the would allow me to set a custom environment variables that are populated inside the template like so:
...
extraEnv:
DOMAIN: mydomain.comthat would be available inside the html template:
<script>
document.domain = '{{domain}}';
</script>If there is a similar existing solution I have missed please let me know.
I am trying to upgrade to 2.0.0 of this docker image, great job re-writing it in Go!
I would have thought it would have been a seamless upgrade however on 2.0.0 I am getting the message. Hi there! Error pages are available at the following URLs: /{code}.html. If I rollback to 1.8.0 it works fine.
error-pages-644fdd7c94-g5pzr error-pages 10.69.152.237 - - [05/Oct/2021:13:43:13 +0000] "GET / HTTP/1.1" 404 3526 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/94.0.4606.61 Safari/537.36" "x.x.x.x, 172.70.110.112"
error-pages-6b4cc76777-4mt4l error-pages {"level":"info","ts":1633441272.6022592,"msg":"HTTP request processed","useragent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/94.0.4606.61 Safari/537.36","method":"GET","url":"/","status_code":404,"connection_close":false,"duration":0.000023755}
You can see my use of error-pages with the middleware and ingressroute here:
https://github.com/onedr0p/home-cluster/tree/main/cluster/apps/networking/error-pages
Is a requirement of the new version supposed to force us to annotate each ingress with using this middleware? If so, can that be changed to be global again?
Thanks!
As you can see here
The issue is that the project tinifier is publishing the page to the wrong docker repository.
You should force am update from this repo after the fix
No response
No response
No response
No response
Thanks for making these! I am wondering if it's all possible to support a default backend when using ingress-nginx.
When using ingress-nginx in kubernetes you can set a defaultBackend
an example of a default backend is below, using Go is not a hard requirement. It just acts on HTTP headers as described in the table on this project.
https://github.com/181192/custom-error-pages
I totally get not wanting to support this, but I'd rather have it built in here instead of forking your repo.
Thanks!
When the auth cookies for my website are present and I try to a non existing page I get "Too big request header" instead of the error page. When I clear said cookies I see my error page as expected.
No response
No response
I am not sure if this is a traefik issue, It does only happen with the error page container, accessing the normal webpage is fine with and without the auth cookies set.
Some services (Syncthing in my case), have a popup authorization prompt rather than something on the page itself. I'm not familiar with what this authorization method is, but it's the same thing my router uses for its admin UI. When my error-pages range includes status code 401, error-pages intercepts and presents the unauthorized page, never giving the popup the chance to... pop up. Right now I've just set my range to 402-599 to avoid this, but I thought you should know in case there's something that can be done to fix it.
I deployed error pages by building it in my docker image of openresty. I manage to get the error page ghost working. The only thing that fails for me is the loading of the font.
I get these error messages in console:
Do you know if there is anything I can do to make the font load properly?
No response
No response
No response
No response
Hi,
Thank you for this simple docker container. I use it with Traefik and it works flawless.
I used another before (very simular to this thought) which has antoher template called "noise"
How can I add this to this repo? I tried to add template, updated tests.yml to add folder and did a docker build.
Step 9/16 : RUN ./generator/generator.js -c ./config.json -o ./out ---> Running in 08d7c027a654 SyntaxError: Unexpected token { in JSON at position 422 at JSON.parse (<anonymous>) at Object.<anonymous> (/src/generator/generator.js:32:32) at Module._compile (node:internal/modules/cjs/loader:1092:14) at Object.Module._extensions..js (node:internal/modules/cjs/loader:1121:10) at Module.load (node:internal/modules/cjs/loader:972:32) at Function.Module._load (node:internal/modules/cjs/loader:813:14) at Function.executeUserEntryPoint [as runMain] (node:internal/modules/run_main:76:12) at node:internal/main/run_main_module:17:47 The command '/bin/sh -c ./generator/generator.js -c ./config.json -o ./out' returned a non-zero code: 1
I need to say, I'm new to building docker images (but running docker containers for quit some time now) and I'm not a programmer
Can you help me out building it or how to add this template to this repo.
Thanks you for your reply,
Ralph
Fetching a specific error with /403.html (and anything not 404) returns as HTTP/1.1 200 OK but with the correct template content. Instead fetching / with a X-Code: 403 works as expected.
docker run --rm -it -p 1234:8080 ghcr.io/tarampampam/error-pages:2.24curl -v -H "X-Code: 403" localhost:1234 > /dev/null -> < HTTP/1.1 403 Forbiddencurl -v localhost:1234/403.html > /dev/null -> < HTTP/1.1 200 OKN/A, defaults are used.{"level":"info","ts":1682500829.970848,"msg":"We will use the requested template","name":"ghost"}
{"level":"info","ts":1682500829.971214,"msg":"Server starting","addr":"0.0.0.0","port":8080,"default error page":"404","default HTTP response code":404,"proxy headers":[],"show request details":false,"localization disabled":false}
{"level":"info","ts":1682500852.084728,"msg":"HTTP request processed","useragent":"''","method":"GET","url":"/403.html","referer":"","status_code":200,"content_type":"text/html; charset=utf-8","connection_close":false,"duration":0.000796617,"request_headers":["Host: localhost:1234","User-Agent: ''","Accept: */*"],"response_headers":["Content-Type: text/html; charset=utf-8","X-Robots-Tag: noindex"]}
{"level":"info","ts":1682500968.6593013,"msg":"HTTP request processed","useragent":"''","method":"GET","url":"/","referer":"","status_code":403,"content_type":"text/html; charset=utf-8","connection_close":false,"duration":0.000545044,"request_headers":["Host: localhost:1234","User-Agent: ''","Accept: */*","X-Code: 403"],"response_headers":["Content-Type: text/html; charset=utf-8","X-Robots-Tag: noindex"]}No response
Hello there!
Firstly let me thank you for creating and keeping alive this project, it's such a simple yet useful thing
I have been having issues with getting error-pages to work with my traefik configuration.
I will use pastebin for configs as they take some place. I will use example of Traefik + Authelia + Nginx + Error Pages
If I use provided labes from the main page (adjusting entrypoints etc), I just get Traefik's 404 everywhere even traefik.domainname.com that worked before now returns traefik's 404.
With my base setup if I go to traefik.domainname.com it redirects to authelia login page and then back to traefik once logged in, it's visible in middlewares section on pastebin.
With the config I attached in pastebin, which is basically the same, domainname.com loads just fine, traefik.domainname.com returns errorpages 404 also domainname.com/anything returns errorpage's 404 even if the file exists, for example index.html. If used just domainname.com, it correctly returns default index.html from nginx configuration but does not put it in the website bar.
https://pastebin.com/mQFXjQCV (configs)
http://prntscr.com/26klxo7 (not working)
http://prntscr.com/26klxy9 (working correctly)
http://prntscr.com/26klzg5 (traefik log)
http://prntscr.com/26km0mm (error pages log)
Once I remove errorspage from middleswares (dynamic file) it all returs to normal. I have tried setting priority to 2 or 11 in errorspages.
I would love to get a direction where to look for, logs shows just that client is hitting 404. If something is missing I will gladly upload more, though it might take me about a day.
Thank you for your time.
Edit 1:
I have found my traefik labels were somehow missing - "traefik.http.middlewares.redirect-to-https.redirectscheme.scheme=https" but it does not help the issue I am having right now.
Would be fun if there was a way to switch to a different error page template after x amount of time (e.g. every 24 hours) when it's set to random.
The docker image for 2.27.0 failed to build. Some sort of golang issue?
https://github.com/tarampampam/error-pages/actions/runs/8309184333
No response
No response
No response
No response
Hello,
first of all, thank you very much for this great project. I like to use this project in different situations. I would like to ask if it would be possible to extend the tool with the following feature:
I am interested in more usecases for simple static-pages which from my ops point of view could run best directly integrated in the "defaultbackend" so this tool.
Currently I am establishing the /.well-known/security.txt on the loadbalancers. See background. This is a static TXT which I would like to read via configmap or similar.
Why here? Because I like the logging, the lightweight implementation and the deployment without filesystem-access.
The way of request-handling could be triggered via 'static-path' or better a request header Static-Page: foobar (e.g. set by the loadbalancer).
maybe also an ENV variable with a comma separated list of static pages as path or header-value (value=filename) to include.
I'm not sure if this is beyond the scope of this tool from your point of view, I think it would be a good extension.
Would there be interest in such a feature?
kind regards
More an FYI for others:
Just been doing some research and it seems that if you want to apply these to any container that does exist, you'll need to create new middleware in traefik.
Easy enough by following:
https://dille.name/blog/2021/03/14/using-traefik-error-pages-to-handle-unavailable-services/
And applying the middleware to all your services.
A good way to test the guide above:
httpbin: # A simple HTTP Request & Response Service.
container_name: httpbin
image: kennethreitz/httpbin
restart: always
networks:
pihole:
ipv4_address: 'static.ip.address.here'
security_opt:
- no-new-privileges:true
labels:
- "traefik.enable=true"
## HTTP Routers
- "traefik.http.routers.httpbin-rtr.entrypoints=https"
- "traefik.http.routers.httpbin-rtr.rule=Host(`httpbin.$DOMAINNAME`)"
- "traefik.http.routers.httpbin-rtr.tls=true"
## Middlewares
- "traefik.http.routers.httpbin-rtr.middlewares=chain-no-auth@file"
- "traefik.http.routers.httpbin-rtr.middlewares=traefik-error-pages"
## HTTP Services
- "traefik.http.routers.httpbin-rtr.service=httpbin-svc"
- "traefik.http.services.httpbin-svc.loadbalancer.server.port=80"https://www.modem7.com/books/docker/page/current-docker-compose — you can see how I've implemented multiple middlewares in each container.
Hi
I have the following error when restarting the container after initial start.
error-pages | mkdir: can't create directory '/opt/html/nginx-error-pages': File exists
error-pages | /docker-entrypoint.sh: set pages for template 'ghost' as default (make accessible in root directory)
Solved by chaging docker-entrypoint.sh line 16, from
mkdir /opt/html/nginx-error-pages;
to
mkdir -p /opt/html/nginx-error-pages;
First of all thank you for your wonderful package!
I would like to use your pages (for instance, the l7-dark template) but with a custom 404 page.
Would it be possible to implement an environmental variable, such as CUSTOM_404 that, if not set, has a default to False?
Then, in the docker container, is just a matter to set CUSTOM_404 to true and implement the correct mapping via a volume entry..
Something like
error-pages:
image: tarampampam/error-pages:latest
ports:
- 8080:8080
volumes:
- /path/to/local/404.html:/custom/404.html:ro
environment:
TEMPLATE_NAME: l7-dark
CUSTOM_404: true
labels:
traefik.enable: true
Thank you!
No response
Hi,
The wiki needs to be updated based on the findings in this discussion: #275
Traefik v3 introduces some change to regex handling and the current wiki version of tarampampam/error-pages suggests an invalid configuration option.
Try using the official documentation with Traefik v3
# Wrong
traefik.http.routers.error-pages-router.rule: HostRegexp(`{host:.+}`)
# Good
traefik.http.routers.error-pages-router.rule: HostRegexp(`.+`)N/AN/A
It would be awesome to be able to define error pages content by env variables.
It'd be sooo nice to be able to do something like:
# docker-compose.yml
version: '3.4'
services:
error-pages:
image: tarampampam/error-pages:latest
environment:
CUSTOM_503: |
<body>
<div>My custom error page</div>
</body>Problem
I'm currently setting up this container with the example provided in the README, but when looking at my Traefik logs, I see the following error log popping up when I up my containers.
{"entryPointName":"websecure","level":"error","msg":"middleware \"error-pages-middleware@docker\" does not exist","routerName":"maintenance-secure@docker","time":"2021-12-19T19:10:24+01:00"}
Maybe I'm missing something obvious, but I just can't seem to find it.
Config
docker-compose.yml
# Custom error pages for downed containers
error-pages:
image: tarampampam/error-pages:${ERROR_PAGES_VERSION}
container_name: error-pages
depends_on:
- traefik
environment:
TEMPLATE_NAME: custom
labels:
- "traefik.enable=true"
- "traefik.http.routers.error-pages-router.rule=HostRegexp(`{host:.+}`)"
- "traefik.http.routers.error-pages-router.priority=10"
- "traefik.http.routers.error-pages-router.tls=true"
- "traefik.http.routers.error-pages-router.entrypoints=websecure"
- "traefik.http.routers.error-pages-router.middlewares=error-pages-middleware@docker"
- "traefik.http.services.error-pages-service.loadbalancer.server.port=8080"
- "traefik.http.middlewares.error-pages-middleware.errors.status=400-599"
- "traefik.http.middlewares.error-pages-middleware.errors.service=error-pages-service@docker"
- "traefik.http.middlewares.error-pages-middleware.errors.query=/{status}.html"
- "traefik.docker.network=proxy"
networks:
- proxy
volumes:
- /var/www/html/error-pages/error-pages.yml:/opt/error-pages.yml
- /var/www/html/error-pages/custom.html:/opt/custom.html
- /var/www/html/error-pages/custom:/opt/templates/custom
# Application maintenance page
maintenance:
build:
context: docker/apache
args:
APACHE_VERSION: ${APACHE_VERSION}
container_name: maintenance_proxy
depends_on:
- error-pages
security_opt:
- no-new-privileges:true
restart: 'no'
volumes:
- /var/www/html/error-pages/custom/503.2.html:/usr/local/apache2/htdocs/index.html:ro
labels:
- "traefik.enable=true"
- "traefik.http.routers.maintenance-secure.entrypoints=websecure"
- "traefik.http.routers.maintenance-secure.rule=Host(`${APP_DOMAIN}`, `www.${APP_DOMAIN}`)"
- "traefik.http.routers.maintenance-secure.service=maintenance-service"
- "traefik.http.routers.maintenance-secure.tls=true"
- 'traefik.http.routers.maintenance-secure.middlewares=error-pages-middleware@docker'
- "traefik.http.services.maintenance-service.loadbalancer.server.port=80"
- "traefik.docker.network=proxy"
networks:
- proxytraefik.yml
api:
dashboard: true
log:
format: json
level: ERROR
entryPoints:
web:
address: :80
http:
redirections:
entryPoint:
to: websecure
websecure:
address: :443
providers:
docker:
endpoint: "unix:///var/run/docker.sock"
exposedByDefault: false
watch: true
network: proxy
file:
directory: /dynamic
watch: trueVersions
Currently, configuring error pages in Docker containers requires manual adjustments to the entrypoint script and ensuring proper permissions for mounted directories. Users need a streamlined way to mount volumes for error page templates, HTML, and configuration files, allowing easy customization and configuration without modifying the container's filesystem or rebuilding the image. Additionally, users should have shell access with docker exec to manage files if needed
Implement support for Docker volumes in the docker-entrypoint.sh script and Dockerfile to allow users to mount local directories to the container. This can be achieved by modifying the entrypoint script to handle symbolic links and permissions, and updating the Dockerfile to ensure proper setup. The following changes are suggested:
/config/templates, /config/html, and /config/error-pages.yml are writable by the container user.The following is a proposed Dockerfile and docker-entrypoint.sh script to support volume mounting and to support shell access:
# use alpine instead of scratch
FROM alpine:latest AS runtime
# Copy the entrypoint script
COPY docker-entrypoint.sh /usr/local/bin/
# Make the script executable
RUN chmod +x /usr/local/bin/docker-entrypoint.sh
# Install recommended packages
RUN apk add --no-cache bash rsync nano
ENTRYPOINT ["/usr/local/bin/docker-entrypoint.sh"]# syntax=docker/dockerfile:1
# this stage is used to build the application
FROM docker.io/library/golang:1.22-bookworm AS builder
COPY ./go.* /src/
WORKDIR /src
# burn the modules cache
RUN go mod download
# this stage is used to compile the application
FROM builder AS compiler
# can be passed with any prefix (like `v1.2.3@GITHASH`), e.g.: `docker build --build-arg "APP_VERSION=v1.2.3@GITHASH" .`
ARG APP_VERSION="undefined@docker"
WORKDIR /src
COPY . .
# arguments to pass on each go tool link invocation
ENV LDFLAGS="-s -w -X gh.tarampamp.am/error-pages/internal/version.version=$APP_VERSION"
# build the application
RUN set -x \
&& CGO_ENABLED=0 go build -trimpath -ldflags "$LDFLAGS" -o ./error-pages ./cmd/error-pages/ \
&& ./error-pages --version \
&& ./error-pages -h
WORKDIR /tmp/rootfs
# prepare rootfs for runtime
RUN set -x \
&& mkdir -p \
./etc \
./bin \
./opt/html \
&& echo 'appuser:x:10001:10001::/nonexistent:/sbin/nologin' > ./etc/passwd \
&& echo 'appuser:x:10001:' > ./etc/group \
&& mv /src/error-pages ./bin/error-pages \
&& mv /src/templates ./opt/templates \
&& rm ./opt/templates/*.md \
&& mv /src/error-pages.yml ./opt/error-pages.yml
WORKDIR /tmp/rootfs/opt
# generate static error pages (for usage inside another docker images, for example)
RUN set -x \
&& ./../bin/error-pages --verbose build --config-file ./error-pages.yml --index ./html \
&& ls -l ./html
# use alpine instead of scratch
FROM alpine:latest AS runtime
ARG APP_VERSION="undefined@docker"
LABEL \
# Docs: <https://github.com/opencontainers/image-spec/blob/master/annotations.md>
org.opencontainers.image.title="error-pages" \
org.opencontainers.image.description="Static server error pages in the docker image" \
org.opencontainers.image.url="https://github.com/tarampampam/error-pages" \
org.opencontainers.image.source="https://github.com/tarampampam/error-pages" \
org.opencontainers.image.vendor="tarampampam" \
org.opencontainers.version="$APP_VERSION" \
org.opencontainers.image.licenses="MIT"
# Install recommended packages
RUN apk add --no-cache bash rsync nano
# Import from compiler build stage
COPY --from=compiler /tmp/rootfs /
# Copy the entrypoint script
COPY docker-entrypoint.sh /usr/local/bin/
# Make the script executable
RUN chmod +x /usr/local/bin/docker-entrypoint.sh
# Use an unprivileged user
USER 10001:10001
WORKDIR /opt
# Set environment variables
ENV LISTEN_PORT="8080" \
TEMPLATE_NAME="ghost" \
DEFAULT_ERROR_PAGE="404" \
DEFAULT_HTTP_CODE="404" \
SHOW_DETAILS="false" \
DISABLE_L10N="false" \
READ_BUFFER_SIZE="2048"
# Healthcheck
HEALTHCHECK --interval=7s --timeout=2s CMD ["/bin/error-pages", "--log-json", "healthcheck"]
ENTRYPOINT ["/usr/local/bin/docker-entrypoint.sh"]
CMD ["--log-json", "serve"]#!/bin/sh
# ANSI Escape Code for Colors
reset="\033[0m"
white_fg_strong="\033[90m"
red_fg_strong="\033[91m"
green_fg_strong="\033[92m"
yellow_fg_strong="\033[93m"
blue_fg_strong="\033[94m"
magenta_fg_strong="\033[95m"
cyan_fg_strong="\033[96m"
# Normal Background Colors
red_bg="\033[41m"
blue_bg="\033[44m"
yellow_bg="\033[43m"
# Function to log messages with timestamps and colors
log_message() {
# This is only time
current_time=$(date +'%H:%M:%S')
# This is with date and time
# current_time=$(date +'%Y-%m-%d %H:%M:%S')
case "$1" in
"INFO")
echo -e "${blue_bg}[$current_time]${reset} ${blue_fg_strong}[INFO]${reset} $2"
;;
"WARN")
echo -e "${yellow_bg}[$current_time]${reset} ${yellow_fg_strong}[WARN]${reset} $2"
;;
"ERROR")
echo -e "${red_bg}[$current_time]${reset} ${red_fg_strong}[ERROR]${reset} $2"
;;
*)
echo -e "${blue_bg}[$current_time]${reset} ${blue_fg_strong}[DEBUG]${reset} $2"
;;
esac
}
set -e
# Default to appuser (UID 10001), so old installations won't break
export PUID=${PUID:-10001}
export PGID=${PGID:-10001}
# Create symbolic links if necessary
if [ ! -d "/opt/templates" ]; then
ln -s /config/templates /opt/templates
log_message "INFO" "Created symbolic link: /config/templates -> /opt/templates"
fi
if [ ! -d "/opt/html" ]; then
ln -s /config/html /opt/html
log_message "INFO" "Created symbolic link: /config/html -> /opt/html"
fi
if [ ! -f "/opt/error-pages.yml" ]; then
ln -s /config/error-pages.yml /opt/error-pages.yml
log_message "INFO" "Created symbolic link: /config/error-pages.yml -> /opt/error-pages.yml"
fi
export ERRORPAGES_BUILDTIME=$(date +%s)
# Set privileges for /opt and /config but only if pid 1 user is root and we are dropping privileges.
# If container is run as an unprivileged user, it means owner already handled ownership setup on their own.
# Running chown in that case (as non-root) will cause error
if [ "$(id -u)" = "0" ] && [ "${PUID}" != "0" ]; then
chown -R ${PUID}:${PGID} /opt
chown -R ${PUID}:${PGID} /config
chown -R 10001:10001 /config/templates
chown -R 10001:10001 /config/html
log_message "INFO" "Changed ownership of /opt and /config to ${PUID}:${PGID}"
log_message "INFO" "Changed ownership of /config/templates and /config/html to 10001:10001"
fi
# Drop privileges (when asked to) if root, otherwise run as current user
if [ "$(id -u)" = "0" ] && [ "${PUID}" != "0" ]; then
log_message "INFO" "Starting error-pages as user ${PUID}:${PGID}"
su-exec ${PUID}:${PGID} /bin/error-pages "$@"
else
log_message "WARN" "No privileges to drop. Running error-pages as current user."
exec /bin/error-pages "$@"
fiMake sure to add docker-entrypoint.sh to the .dockerignore file
## Ignore everything
*
## Except the following files and directories
!/docker-entrypoint.sh
!/cmd
!/internal
!/templates
!/error-pages.yml
!/go.*
By following this approach, users can mount volumes with the docker compose file:
services:
error-pages:
image: error-pages:latest
container_name: error-pages
environment:
PGID: 1000
PUID: 1000
TEMPLATE_NAME: connection # set the error pages template
volumes:
- ./error-pages/data/templates:/config/templates
- ./error-pages/data/html:/config/html
- ./error-pages/data/config/error-pages.yml:/config/error-pages.yml:rw
ports:
- "8176:8080"
restart: unless-stoppedTrying to disable logs
Have an env var to disable logs
No response
Issue derived from discussion #129
After finding out that that serving fonts through Google Fonts is not GDPR compliant (bunny.net; official ruling), is it maybe an idea to update this in the templates by switching to another font service, like for example Bunny Fonts? It would be as easy as just replacing fonts.googleapis.com by fonts.bunny.net.
Replace fonts.googleapis.com by fonts.bunny.net in link tags.
No response
Trying out v3 in docker with tarampampam/error-pages:3.1.0
I have always been running a custom template and with the new version I need to run the cli flag --add-template and it makes the Docker-composefile a bit messy which was not needed prior.
Adding a companion env var ADD_TEMPLATE as for the other config flags, would allow me to not override the default command in docker.
No response
Great work on this project, it's certainly beautified my end users experience when they get an error however...
As transmission uses rpc to serve its pages to protect against CSRF it expects a valid session-id header in the request, otherwise it returns a 409 page.
Without error-pages defined on the entrypoint in Traefik, the request hits the rpc endpoint and on the initial try it obtains a session-id and then retries resulting in a successful response and the page loads.
With error-pages defined, all consequent hits to the rpc endpoint result in a 409 as this error page strips the X-Transmission-Session-id header and therefore gets rejected.
Is there anyway of setting "Pass-through request/response headers" or an abilitiy to set a list of headers that are allowed to be kept?
Thanks in advance.
Logs:
Without error-pages:
172.16.200.15 - - [20/Feb/2022:23:21:13 +0000] "POST /transmission/rpc HTTP/1.1" 409 593 "https://transmission.mydomain.com/transmission/web/" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.102 Safari/537.36" "10.32.100.189"
172.16.200.3 - - [20/Feb/2022:23:21:15 +0000] "POST /transmission/rpc HTTP/1.1" 200 424 "-" "Python-urllib/3.9" "-"
172.16.200.3 - - [20/Feb/2022:23:21:15 +0000] "POST /transmission/rpc HTTP/1.1" 200 23925 "-" "Python-urllib/3.9" "-"
172.16.200.3 - - [20/Feb/2022:23:21:15 +0000] "POST /transmission/rpc HTTP/1.1" 200 1563 "-" "Python-urllib/3.9" "-"
With error-pages:
172.16.200.15 - - [20/Feb/2022:23:21:18 +0000] "POST /transmission/rpc HTTP/1.1" 409 593 "https://transmission.mydomain.com/transmission/web/" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.102 Safari/537.36" "10.32.100.189"
172.16.200.15 - - [20/Feb/2022:23:21:23 +0000] "POST /transmission/rpc HTTP/1.1" 409 593 "https://transmission.mydomain.com/transmission/web/" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.102 Safari/537.36" "10.32.100.189"
172.16.200.15 - - [20/Feb/2022:23:21:28 +0000] "POST /transmission/rpc HTTP/1.1" 409 593 "https://transmission.mydomain.com/transmission/web/" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.102 Safari/537.36" "10.32.100.189"
172.16.200.15 - - [20/Feb/2022:23:21:33 +0000] "POST /transmission/rpc HTTP/1.1" 409 593 "https://transmission.mydomain.com/transmission/web/" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.102 Safari/537.36" "10.32.100.189"
172.16.200.15 - - [20/Feb/2022:23:21:38 +0000] "POST /transmission/rpc HTTP/1.1" 409 593 "https://transmission.mydomain.com/transmission/web/" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.102 Safari/537.36" "10.32.100.189"
No response
No response
No response
Hi 👋🏼 It would be pretty neat to have your project support ingress-nginx, currently I've been using 181192/custom-error-pages which has some of your themes (there's credit to you in the README). However it would be awesome if your project was a solution for both traefik and ingress-nginx. Thanks!
Edit: This is a very basic implentation of the default-backend as provided by the ingress-nginx project.
https://github.com/kubernetes/ingress-nginx/tree/main/images/custom-error-pages/rootfs
Hello everyone!
I've been working hard over the past few weeks on a major version of this project, and my work is nearly complete. The most important parts are done and ready to be tested. Please note that the following changes are not backward-compatible, which is why this is a major update:
error-pages.yml) is no longer needed. Everything is now configured using CLI flags and environment variables--add-template /path/to/your/tpl.htmv2, some have changed. Please update your charts accordingly. For the latest information, check here. The README file is still in progress, but the CLI documentation is available/{code}, /{code}.html, and /{code}.htm. Additionally, you can use the flag --send-same-http-code to retain the same HTTP status code as the page, if needed--add-http-code="…" in the format %code%=%message%/%description%. Moreover, you can use "placeholder" codes like 4**, which means if the requested code is not described but is between 400 and 499, the provided message with its description will be returned. This approach is similar to the one used in the OpenAPI specification, where instead of describing each code individually, you can describe a patterncurl an error page, you will receive more readable content without tons of HTML tags. This change should not cause any issues, but it mayfasthttp HTTP server or switch to the stdlib HTTP server, and whether to keep the templates caching mechanism used in v2. Currently, I have switched to the stdlib HTTP server (due to better support and built-in HTTP/2 support, though with lower performance - 40k RPS has dropped to 11k). I've also removed the caching mechanism, which should reduce memory consumption but may increase CPU load slightly. I haven't finalized these decisions yet, so please share your thoughtsl7-dark and l7-light have been merged into a single l7 theme, and the matrix theme has been removed (because I created it and it wasn't very appealing)/metrics has been not implemented yet, If someone found it useful and has used it - just tell me, and I'll re-implement itI'm reaching out to see if anyone would like to assist with early testing of the app. I can provide early builds (Docker image tags) via messages in this thread. It would be fantastic if you could run them locally or on your staging cluster and provide feedback on whether everything is working smoothly or if you encounter any issues.
Please leave a comment to let me know if you're willing to help. Your assistance would be greatly appreciated!
/cc @deffcolony @moschlar @r2DoesInc @fuog @kfirfer @CodeAnthem @NicosKaralis @GoliathLabs @onedr0p @modem7 @ToshY
I have this setup as a default error handler on my k8s cluster, and it works great overall. My apis deployed on the cluster have a bit of an issue though.
When returning a 404 or something that this project is setup to respond for, my error content is ignored.
This turns my
{
"error": "location_invalid",
"message": "Given location either does not exist or is invalid."
}
Into
{
"error": true,
"code": "404",
"message": "Not Found",
"description": "The server can not find the requested page",
}
I propose modifying the error page to pull from the response body if it is provided
Something like
props.Message = string(ctx.Request.Body.Peek(Message))
I dont know go, so not sure if thats how it would be implemented.
This could be configured behind a bool for ALLOW_ERROR_MESSAGE_PASSTHROUGH which would default to false to maintain consistency with prior behavior.
No response
When sending regular not found URL with GET method, im getting 404 as usual
when sending the same thing with methods other than GET, its returns getting 405
Using ingress-nginx controller
And the helm chart configuration & log below:
No response
controller:
config:
custom-http-errors: "400,401,403,404,405,407,408,409,410,411,412,413,416,418,429,500,502,503,504,505"
defaultBackend:
enabled: true
image:
repository: tarampampam/error-pages
tag: "2.20.0"
extraEnvs:
- name: TEMPLATE_NAME # Optional: change the default theme
value: "app-down"
- name: SHOW_DETAILS # Optional: enables the output of additional information on error pages
value: "true"
- name: LOG_FORMAT
value: "json"
### Relevant log output
```shell
This is the log from default backend pod:
[ingress-nginx-defaultbackend-869c98f496-d644d ingress-nginx-default-backend] 2023-02-19T10:26:44.092636417Z {"level":"info","ts":1676802404.092549,"msg":"HTTP request processed","useragent":"PostmanRuntime/7.28.4","method":"POST","url":"/","referer":"","status_code":405,"content_type":"text/plain; charset=utf-8","connection_close":false,"duration":0.000008231,"request_headers":["Host: paa-theruntime.local.platform.test","Content-Length: 243","Content-Type: application/json","User-Agent: PostmanRuntime/7.28.4","Cookie: INGRESSCOOKIE=2ec12f7f26a5b8a4fd603d8b8e5fbd0c|d8628a0b5df4f2b5b6b4858b98a105ce","X-Code: 404","X-Format: */*","X-Original-Uri: /api/runtime/token/v3","X-Request-Id: f4d77e12250191364f9b947e3182d74f","X-Forwarded-For: 192.168.65.4","Accept: */*","Postman-Token: 9d83ccbe-2284-4699-b9ae-42f9eabda61d","Accept-Encoding: gzip, deflate, br","X-Forwarded-Proto: http","X-Envoy-Internal: true","X-B3-Traceid: f5ded2c870cdfe6564cb192bbd092bb9","X-B3-Spanid: 64cb192bbd092bb9","X-B3-Sampled: 1"],"response_headers":["Content-Type: text/plain; charset=utf-8","Allow: GET, OPTIONS"]}
### Anything else?
_No response_
I´m running nextcloud with a nextcloud push server container. Nextcloud itself is running with priority 1 and the sidecar container with priority 2. However, when adding error-pages with priority 10, nextcloud is down. When removing the priorities from nextcloud, they came back. I don´ t know how to fix this problem.
Thanks in advance!
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.