# MIME type issue

**URL:** <https://community.traefik.io/t/mime-type-issue/25943>\
**Category:** Traefik v3 (latest)\
**Tags:** docker\
**Created:** [January 11, 2025, 6:59am UTC](https://community.traefik.io/t/mime-type-issue/25943 "2025-01-11T06:59:14Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![nrgbistro](https://sea2.discourse-cdn.com/flex020/user_avatar/community.traefik.io/nrgbistro/32/9820_2.png) [@nrgbistro](https://community.traefik.io/u/nrgbistro)\
**Post date:** [January 11, 2025, 6:59am UTC](https://community.traefik.io/t/mime-type-issue/25943/1 "2025-01-11T06:59:14Z")

</div>

I'm new to Traefik (coming from NPM) and I'm facing an issue that I can't find elsewhere on this forum or the internet. I'm able to access a few of my services using the wildcard certificate I set up, absolutely no issues. However, I recently set up a Radarr instance to be routed via Traefik and I get this logged to my browser console after login:

```auto
Refused to apply style from 'https://radarr.domain.com/Content/Fonts/fonts.css?h=ZYP7+3tB+/eme7MGOlDOLg' because its MIME type ('text/plain') is not a supported stylesheet MIME type, and strict MIME checking is enabled.
radarr.domain.com/:1 Refused to apply style from 'https://radarr.domain.com/Content/styles.css?h=VAwjWra8YBWH8JiyE7JFDA' because its MIME type ('text/plain') is not a supported stylesheet MIME type, and strict MIME checking is enabled.
radarr.domain.com/:15 Refused to apply style from 'https://radarr.domain.com/Content/styles.css?h=VAwjWra8YBWH8JiyE7JFDA' because its MIME type ('text/plain') is not a supported stylesheet MIME type, and strict MIME checking is enabled.

```

For some reason the MIME type is being changed from `text/css` to `text/plain` for these 3 files. When accessing it directly via [http://ip](http://ip):port, I do not see this issue at all. The reason this is a serious problem is because it is cutting off most of the application, see below:

 ![image](https://us1.discourse-cdn.com/flex020/uploads/containo/original/2X/3/31cde712f128c9c07cf8ed8efff025152d8b79e7.png)  
 ![image](https://us1.discourse-cdn.com/flex020/uploads/containo/original/2X/8/8c90b88859af0b0e0786b3b8365ff0fccb0b00e9.png)

This makes the application completely unusable and I don't understand why this would be happening. I'm not running any middlewares or plugins, this is a very simple configuration.

static config:

```auto
global:
  sendAnonymousUsage: false
log:
  level: DEBUG
api:
  dashboard: true
  insecure: true
entryPoints:
  https:
    address: :443
    http:
      tls:
        certResolver: cloudflare
        domains:
          - main: "domain.com"
            sans:
              - "*.domain.com"
certificatesResolvers:
  cloudflare:
    acme:
      email: "email@gmail.com"
      storage: /var/traefik/certs/cloudflare-acme.json
      caServer: "https://acme-v02.api.letsencrypt.org/directory"
      keyType: EC384
      dnsChallenge:
        provider: cloudflare
        resolvers:
          - "1.1.1.1:53"
          - "8.8.8.8:53"
providers:
  docker:
    exposedByDefault: false

```

Radarr compose:

```auto
services:
  radarr:
    image: linuxserver/radarr:latest
    container_name: radarr
    restart: unless-stopped
    ports:
      - 7878:7878
    environment:
      - PUID=1000
      - PGID=1001
      - UMASK=022
    volumes:
      - /mnt/appdata/radarr:/config
      - /mnt/downloads:/downloads
      - /mnt/media/movies:/movies
    labels:
      - traefik.enable=true
      - traefik.http.routers.radarr.rule=Host(`radarr.domain.com`)
      - traefik.http.services.radarr.loadbalancer.server.port=7878

```

Again, this works exactly as expected locally, so I'm pretty confident that Traefik is to blame, I just can't figure out how. I don't see anything in the Traefik debug logs that suggest any issues.

EDIT: I just tried this again with a completely fresh installation and still see the exact same problem.

EDIT2: I eventually see this error log, but it takes a long time to show up after I try loading the page so it's hard to tell if it's definitely related

```auto
DBG github.com/traefik/traefik/v3/pkg/middlewares/recovery/recovery.go:45 > Request has been aborted [172.70.115.216:44036 - /api/hosts/75f8bb84-9211-483a-a195-ec9603aa0d63/containers/e97a5229dfcb/logs/stream?stdout=1&stderr=1&filter=radarr&levels=info&levels=debug&levels=warn&levels=error&levels=fatal&levels=trace&levels=unknown]: net/http: abort Handler middlewareName=traefik-internal-recovery middlewareType=Recovery

```

---

<div class="post-metadata">

**Author:** ![bluepuma77](https://avatars.discourse-cdn.com/v4/letter/b/a9adbd/32.png) [@bluepuma77](https://community.traefik.io/u/bluepuma77)\
**Post date:** [January 11, 2025, 7:51am UTC](https://community.traefik.io/t/mime-type-issue/25943/2 "2025-01-11T07:51:20Z")

</div>

Traefik usually doesn’t change any headers like mime-type by itself.

The mine-type is usually supplied by the web server, that’s included in the target service image.

Use `curl -v` to request `font.css`, it will show the headers. Then run the service with own port published and check again if headers change.

If both have the wrong nime-type, then the image creator needs to update the settings.

At the end it’s the browser which seems a bit picky and doesn’t like to apply the file.

---

<div class="post-metadata">

**Author:** ![nrgbistro](https://sea2.discourse-cdn.com/flex020/user_avatar/community.traefik.io/nrgbistro/32/9820_2.png) [@nrgbistro](https://community.traefik.io/u/nrgbistro)\
**Post date:** [January 11, 2025, 6:51pm UTC](https://community.traefik.io/t/mime-type-issue/25943/3 "2025-01-11T18:51:26Z")

</div>

That's what chatGPT and other sources I found online said, that Traefik generally doesn't change the mime-type. However, my current results seem to contradict that. Here are my results from using `curl -v`

connecting directly via ip:

```auto
* Trying 192.168.0.11:7878...
* Connected to 192.168.0.11 (192.168.0.11) port 7878
> GET /Content/Fonts/fonts.css?h=ZYP7+3tB+/eme7MGOlDOLg HTTP/1.1
> Host: 192.168.0.11:7878
> User-Agent: curl/8.7.1
> Accept: */*
> 
* Request completely sent off
< HTTP/1.1 200 OK
< Content-Length: 1049
< Content-Type: text/css
< Date: Sat, 11 Jan 2025 18:46:31 GMT
< Server: Kestrel
< Cache-Control: max-age=31536000, public
< Last-Modified: Sun, 05 Jan 2025 11:24:58 GMT
...

```

connecting via [radarr.domain.com](http://radarr.domain.com):

```auto
... <TLS info> ...
> GET /Content/Fonts/fonts.css?h=ZYP7+3tB+/eme7MGOlDOLg HTTP/2
> Host: radarr.domain.com
> User-Agent: curl/8.7.1
> Accept: */*
> 
* Request completely sent off
< HTTP/2 404 
< date: Sat, 11 Jan 2025 18:45:19 GMT
< content-type: text/plain; charset=utf-8
< cache-control: max-age=31536000
< vary: Accept-Encoding
< x-content-type-options: nosniff
< cf-cache-status: HIT
< age: 48787
< report-to: {"endpoints":[{"url":"https:\/\/a.nel.cloudflare.com\/report\/v4?s=SUQkdfG1jLLujWy2bifAolX8TU8MtWv9s38G4sFkfg9DEaWmUU%2FnhKM%2Bk7%2Feh%2FHn6tQe7DwvOeOTD1CHQOEAey7iJq2qCCE69Qua6NFvf1St3f1xVu3LVgNJfL2gEaQI3SFNBp7g"}],"group":"cf-nel","max_age":604800}
< nel: {"success_fraction":0,"report_to":"cf-nel","max_age":604800}
< strict-transport-security: max-age=7776000; includeSubDomains; preload
< server: cloudflare
< cf-ray: 9007102cff347cb2-EWR
< alt-svc: h3=":443"; ma=86400
< server-timing: cfL4;desc="?proto=TCP&rtt=11664&min_rtt=9923&rtt_var=3747&sent=5&recv=10&lost=0&retrans=0&sent_bytes=2891&recv_bytes=615&delivery_rate=291847&cwnd=249&unsent_bytes=0&cid=1dc8b2dcd4118b94&ts=34&x=0"
< 
404 page not found

```

I will note that this installation used to be behind an Authentik instance, and I plan to change it back to that, but I would really like to understand why this is happening to get a better feel for how Traefik works.

---

<div class="post-metadata">

**Author:** ![bluepuma77](https://avatars.discourse-cdn.com/v4/letter/b/a9adbd/32.png) [@bluepuma77](https://community.traefik.io/u/bluepuma77)\
**Post date:** [January 12, 2025, 7:36am UTC](https://community.traefik.io/t/mime-type-issue/25943/4 "2025-01-12T07:36:36Z")

</div>

Ok, but this is really a different issue: it’s not about from mime-type, but about the page not being found (status 404), so the path seems not to be working.

---

<div class="post-metadata">

**Author:** ![nrgbistro](https://sea2.discourse-cdn.com/flex020/user_avatar/community.traefik.io/nrgbistro/32/9820_2.png) [@nrgbistro](https://community.traefik.io/u/nrgbistro)\
**Post date:** [January 13, 2025, 2:46pm UTC](https://community.traefik.io/t/mime-type-issue/25943/5 "2025-01-13T14:46:50Z")

</div>

Do you know what might cause those specific paths to not be found when connecting via Traefik? Is there any other debug info I can share here to help figure out why this might be happening?

I've successfully put the service behind a domain-wide forward auth, but I still get this MIME type issue after logging into Authentik.

---

<div class="post-metadata">

**Author:** ![bluepuma77](https://avatars.discourse-cdn.com/v4/letter/b/a9adbd/32.png) [@bluepuma77](https://community.traefik.io/u/bluepuma77)\
**Post date:** [January 13, 2025, 4:11pm UTC](https://community.traefik.io/t/mime-type-issue/25943/6 "2025-01-13T16:11:57Z")

</div>

Enable and check Traefik debug log ([doc](https://doc.traefik.io/traefik/observability/logs/#level)) and Traefik access log in JSON format ([doc](https://doc.traefik.io/traefik/observability/access-logs/#format)).

---

<div class="post-metadata">

**Author:** ![nrgbistro](https://sea2.discourse-cdn.com/flex020/user_avatar/community.traefik.io/nrgbistro/32/9820_2.png) [@nrgbistro](https://community.traefik.io/u/nrgbistro)\
**Post date:** [January 15, 2025, 3:21pm UTC](https://community.traefik.io/t/mime-type-issue/25943/7 "2025-01-15T15:21:29Z")

</div>

I'm not exactly sure what happened, I only changed Traefik to start logging to a JSON file in the static config and now radarr seems to work. Thanks for your help!