# Styles are messed up **URL:** \ **Category:** Issues & Bugs\ **Tags:** aws, stylesheets, session-replay\ **Created:** [January 14, 2025, 11:23am UTC](https://forum.openreplay.com/t/styles-are-messed-up/469 "2025-01-14T11:23:16Z")\ **Posts on this page:** 8\ **Page:** 1
**Author:** ![nithinofx](https://avatars.discourse-cdn.com/v4/letter/n/9e8a1a/32.png) [@nithinofx](https://forum.openreplay.com/u/nithinofx)\ **Post date:** [January 14, 2025, 11:23am UTC](https://forum.openreplay.com/t/styles-are-messed-up/469/1 "2025-01-14T11:23:16Z")
We have deployed OpenReplay on our ec2 instance. Our application is using Angular. When we do session recordings we see some styles are missing in the session recording. Most of these styles are in JS files after angular build and not exactly in CSS files (CSS files are loaded and general styles in the app is working but some specific parts are not rendered). I also noted that some JS files show 4xx-5xx in the session recording. I have tried to access the same URLs to verify they are working and I made sure these JS files are actually accessible. ![js4xx](https://europe1.discourse-cdn.com/flex006/uploads/openreplay/original/1X/1ceac502c11f4960668da903099fc3c504dbb46f.png) Openreplay version: v1.19.0 Angular version: v14 ---
**Author:** ![mehdi](https://dub1.discourse-cdn.com/flex006/user_avatar/forum.openreplay.com/mehdi/32/50_2.png) [@mehdi](https://forum.openreplay.com/u/mehdi)\ **Post date:** [January 14, 2025, 3:07pm UTC](https://forum.openreplay.com/t/styles-are-messed-up/469/2 "2025-01-14T15:07:48Z")
Resource timings is an unreliable API that doesn’t provide any failure data or flags so we consider something with HTTP code \>=400 when duration of the connection is (in current/prev versions) + transferred data is also 0 (in new versions, which means that request had 0ms connection and transmitted 0 bytes of data). As for styles, open `browser inspector -> network` then check if session attempts to load styles (check inside iframe collection inside the list as well, since replay is played in artificial iframe). If something is missing here, then check console for errors (you can paste ` __OPENREPLAY_DEV_TOOLS__.logStuff()` then reload the page to enable debug info), if everything is there but failing then its caching issue on backend services. ---
**Author:** ![mathieudek](https://avatars.discourse-cdn.com/v4/letter/m/7c8e57/32.png) [@mathieudek](https://forum.openreplay.com/u/mathieudek)\ **Post date:** [January 28, 2025, 10:15am UTC](https://forum.openreplay.com/t/styles-are-messed-up/469/3 "2025-01-28T10:15:38Z")
Hello, I’m encountering a similar issue after deploying OpenReplay. Many sessions have their styles “messed up” during replay. Since users are able to use the service without issues, I assume the styles are loading correctly in real time. However, when I check the Network console during replay, I notice HTTP errors with codes \>= 400. Interestingly, the Network console also indicates that some resources are “Served from cache”: ![Capture d’écran 2025-01-28 à 11.11.27](https://europe1.discourse-cdn.com/flex006/uploads/openreplay/original/1X/e9e46bfe7397105489e4ce74b3dee1d45af76bc9.png) Could it be that OpenReplay is mistakenly interpreting cached content as an error? Any insights or guidance on how to resolve this would be greatly appreciated! Openreplay version : v1.21.0 ---
**Author:** ![mehdi](https://dub1.discourse-cdn.com/flex006/user_avatar/forum.openreplay.com/mehdi/32/50_2.png) [@mehdi](https://forum.openreplay.com/u/mehdi)\ **Post date:** [January 28, 2025, 12:04pm UTC](https://forum.openreplay.com/t/styles-are-messed-up/469/4 "2025-01-28T12:04:11Z")
It could be yes. What version of tracker are you using? Also, are you stylesheets behind a proxy or a CDN? Your OpenReplay instances will basically try to cache those stylesheets, which may be restricted by your CDN. I’d look into the `assets` service logs by running `openreplay -l assets`. You should see the errors there. ---
**Author:** ![mathieudek](https://avatars.discourse-cdn.com/v4/letter/m/7c8e57/32.png) [@mathieudek](https://forum.openreplay.com/u/mathieudek)\ **Post date:** [January 28, 2025, 3:45pm UTC](https://forum.openreplay.com/t/styles-are-messed-up/469/5 "2025-01-28T15:45:40Z")
I’m using Cloudflare as a CDN, so maybe it is blocking OpenReplay instances. I’m trying to bypass it by using resourceBaseHref option. How can I check that this option is working ? it there a way to see all logs from assets service ? ---
**Author:** ![mehdi](https://dub1.discourse-cdn.com/flex006/user_avatar/forum.openreplay.com/mehdi/32/50_2.png) [@mehdi](https://forum.openreplay.com/u/mehdi)\ **Post date:** [January 30, 2025, 2:58pm UTC](https://forum.openreplay.com/t/styles-are-messed-up/469/6 "2025-01-30T14:58:12Z")
Can you try running `openreplay -l "assets --no-follow"`? The best way to check if the option is working is stylesheets you see in the replay. ---
**Author:** ![nithinofx](https://avatars.discourse-cdn.com/v4/letter/n/9e8a1a/32.png) [@nithinofx](https://forum.openreplay.com/u/nithinofx)\ **Post date:** [February 26, 2025, 11:38am UTC](https://forum.openreplay.com/t/styles-are-messed-up/469/7 "2025-02-26T11:38:12Z")
[How to increase the message size limit? - Community Guides - OpenReplay Forum](https://forum.openreplay.com/t/how-to-increase-the-message-size-limit/65) Answer to this http: ingress: annotations: [nginx.ingress.kubernetes.io/proxy-body-size:](http://nginx.ingress.kubernetes.io/proxy-body-size:) 10m env: BEACON\_SIZE\_LIMIT: “10000000” seems to have fixed our original issue. Based on this, what could have been the original issue? ---
**Author:** ![mehdi](https://dub1.discourse-cdn.com/flex006/user_avatar/forum.openreplay.com/mehdi/32/50_2.png) [@mehdi](https://forum.openreplay.com/u/mehdi)\ **Post date:** [February 26, 2025, 5:48pm UTC](https://forum.openreplay.com/t/styles-are-messed-up/469/8 "2025-02-26T17:48:21Z")
The recording were missing bits here and there because of the beaconSize (the maximum buffer size on the tracker level). Most likely that explains why the replays were messed up.