Page unload

[peprah727]

Unable to open my website page which ha a api linke

[theluckystrike]

This error page comes from Next.js, not from Supabase. It is the default page Next.js shows in production when a page throws an error on the server. Next.js hides the real message from the browser on purpose. The number at the bottom (ERROR 500438034) is the error's digest, and Next.js writes the same digest to the server log next to the real error.

To find the real error on Vercel:

1. Open the page again so the error happens again. On the Hobby plan, Vercel keeps runtime logs for only 1 hour, so the log from when you took the photo is probably gone.
2. In the Vercel dashboard, open your project and select Logs in the sidebar.
3. Search for the new digest number, or filter by status code 500.
4. Open the entry. It shows the full error message and stack trace.

When a Next.js page that reads from Supabase fails like this, the log line usually points to one of these causes:

• supabaseUrl is required. or Invalid API key. The Supabase environment variables are missing or wrong in Vercel. Set them under Project Settings → Environment Variables for the Production environment, then redeploy. Vercel applies changed variables only to new deployments.
• TypeError: fetch failed. The server could not reach Supabase. Check that your Supabase project is not paused (free projects can be paused when they are inactive) and that the URL in Vercel matches the one in your Supabase project settings.
• Cannot read properties of null. Your code used data from a query that returned an error or no rows, for example because an RLS policy hides the rows from that user. Check error before you use data.

Your home page at pink-fm-website.vercel.app loads for me right now, so the error may happen only on one page, for example a page that needs a signed-in user. Open that page, then look at the log for that request.

If you paste the error line from the log here (without any keys), I can help you narrow it down.

Docs: https://nextjs.org/docs/app/api-reference/file-conventions/error#errordigest

[peprah727]

That problem is solved but what about the fact that the when a news or article which includes videos and images is posted on the pink fm website but the videos and images is disappeared from the website homepage or preview of that article

…

On Sun, 25 Oct 2026, 11:53 Michael Lip, ***@***.***> wrote: This error page comes from Next.js, not from Supabase. It is the default page Next.js shows in production when a page throws an error on the server. Next.js hides the real message from the browser on purpose. The number at the bottom (ERROR 500438034) is the error's digest, and Next.js writes the same digest to the server log next to the real error. To find the real error on Vercel: 1. Open the page again so the error happens again. On the Hobby plan, Vercel keeps runtime logs for only 1 hour, so the log from when you took the photo is probably gone. 2. In the Vercel dashboard, open your project and select *Logs* in the sidebar. 3. Search for the new digest number, or filter by status code 500. 4. Open the entry. It shows the full error message and stack trace. When a Next.js page that reads from Supabase fails like this, the log line usually points to one of these causes: - supabaseUrl is required. or Invalid API key. The Supabase environment variables are missing or wrong in Vercel. Set them under Project Settings → Environment Variables for the Production environment, then redeploy. Vercel applies changed variables only to new deployments. - TypeError: fetch failed. The server could not reach Supabase. Check that your Supabase project is not paused (free projects can be paused when they are inactive) and that the URL in Vercel matches the one in your Supabase project settings. - Cannot read properties of null. Your code used data from a query that returned an error or no rows, for example because an RLS policy hides the rows from that user. Check error before you use data. Your home page at pink-fm-website.vercel.app loads for me right now, so the error may happen only on one page, for example a page that needs a signed-in user. Open that page, then look at the log for that request. If you paste the error line from the log here (without any keys), I can help you narrow it down. Docs: https://nextjs.org/docs/app/api-reference/file-conventions/error#errordigest — Reply to this email directly, view it on GitHub <#52271?email_source=notifications&email_token=BRNHK5GYZ5R5WJJFTPP2QPT5VXSTHA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOJRG43TIMJXUZZGKYLTN5XKMYLVORUG64VFMV3GK3TUVRTG633UMVZF6Y3MNFRWW#discussioncomment-19177417>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BRNHK5GZ5O5HFQUMQRLK62D5VXSTHAVCNFSNUABIKJSXA33TNF2G64TZHMZDCNBVHA3TCOJTHNCGS43DOVZXG2LPNY5TCMJRGI2TCOBXUF3AE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[theluckystrike]

Glad the 500 is sorted. Media that shows up at first and then disappears usually comes down to the URL saved with the article. Open that stored image or video URL in a private browser window and see what happens. A URL with `/storage/v1/object/sign/` and a `token=` parameter is a signed URL, and signed URLs stop working once `expiresIn` runs out. That fits media that works right after posting and vanishes later. Make the bucket public and save the result of `getPublicUrl()`, or save only the file path and call `createSignedUrl()` each time the page renders. A `/storage/v1/object/public/` URL that returns 400 or 404 is different. The bucket isn't public, or the stored path doesn't match the uploaded file, so check the bucket's Public toggle in Storage and compare the path with the file listed there. When the URL opens fine but the homepage card is still empty, look at how the card builds its preview. Code that turns the article body into a text excerpt drops `<img>` and `<video>` tags, so the card needs its own cover image field. Share one stored URL with the token removed, plus the line that saves it, and I'll point to the exact cause. Thank you, Michael Lip Founder of https://zovo.one, 108 Chrome extensions with 150,000+ users Also building https://makeroll.com and https://ml0x.com, open-source guardrails for Claude Code agents Open source has to win, and small fixes like this one are my part of it GitHub Sponsors keeps these fixes coming, https://github.com/sponsors/theluckystrike

[peprah727]

Thanks, I'll check it out.

…

On Mon, 26 Oct 2026, 12:18 Michael Lip, ***@***.***> wrote: Glad the 500 is sorted. Media that shows up at first and then disappears usually comes down to the URL saved with the article. Open that stored image or video URL in a private browser window and see what happens. A URL with `/storage/v1/object/sign/` and a `token=` parameter is a signed URL, and signed URLs stop working once `expiresIn` runs out. That fits media that works right after posting and vanishes later. Make the bucket public and save the result of `getPublicUrl()`, or save only the file path and call `createSignedUrl()` each time the page renders. A `/storage/v1/object/public/` URL that returns 400 or 404 is different. The bucket isn't public, or the stored path doesn't match the uploaded file, so check the bucket's Public toggle in Storage and compare the path with the file listed there. When the URL opens fine but the homepage card is still empty, look at how the card builds its preview. Code that turns the article body into a text excerpt drops `<img>` and `<video>` tags, so the card needs its own cover image field. Share one stored URL with the token removed, plus the line that saves it, and I'll point to the exact cause. Thank you, Michael Lip Founder of https://zovo.one, 108 Chrome extensions with 150,000+ users Also building https://makeroll.com and https://ml0x.com, open-source guardrails for Claude Code agents Open source has to win, and small fixes like this one are my part of it GitHub Sponsors keeps these fixes coming, https://github.com/sponsors/theluckystrike — Reply to this email directly, view it on GitHub <#52271?email_source=notifications&email_token=BRNHK5FQBFSQT7YJ2VESCP35V46H3A5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOJRHE2TEMZUUZZGKYLTN5XKMYLVORUG64VFMV3GK3TUVRTG633UMVZF6Y3MNFRWW#discussioncomment-19195234>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BRNHK5GDZDGMPDE4EPKLDQ35V46H3AVCNFSNUABIKJSXA33TNF2G64TZHMZDCNBVHA3TCOJTHNCGS43DOVZXG2LPNY5TCMJRGI2TCOBXUF3AE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[peprah727]

Please does supabase has the ability to handle files and urls .ive creted a website but but any time i post or bublish article or news into my website homepage, files do not display . Video files included in the article dismisses, images files, image and video urls do not dispay its contents. Please help me out , what might be wrong is it settings in supabase or the serup in my code

…

On Sun, 25 Oct 2026, 11:53 Michael Lip, ***@***.***> wrote: This error page comes from Next.js, not from Supabase. It is the default page Next.js shows in production when a page throws an error on the server. Next.js hides the real message from the browser on purpose. The number at the bottom (ERROR 500438034) is the error's digest, and Next.js writes the same digest to the server log next to the real error. To find the real error on Vercel: 1. Open the page again so the error happens again. On the Hobby plan, Vercel keeps runtime logs for only 1 hour, so the log from when you took the photo is probably gone. 2. In the Vercel dashboard, open your project and select *Logs* in the sidebar. 3. Search for the new digest number, or filter by status code 500. 4. Open the entry. It shows the full error message and stack trace. When a Next.js page that reads from Supabase fails like this, the log line usually points to one of these causes: - supabaseUrl is required. or Invalid API key. The Supabase environment variables are missing or wrong in Vercel. Set them under Project Settings → Environment Variables for the Production environment, then redeploy. Vercel applies changed variables only to new deployments. - TypeError: fetch failed. The server could not reach Supabase. Check that your Supabase project is not paused (free projects can be paused when they are inactive) and that the URL in Vercel matches the one in your Supabase project settings. - Cannot read properties of null. Your code used data from a query that returned an error or no rows, for example because an RLS policy hides the rows from that user. Check error before you use data. Your home page at pink-fm-website.vercel.app loads for me right now, so the error may happen only on one page, for example a page that needs a signed-in user. Open that page, then look at the log for that request. If you paste the error line from the log here (without any keys), I can help you narrow it down. Docs: https://nextjs.org/docs/app/api-reference/file-conventions/error#errordigest — Reply to this email directly, view it on GitHub <#52271?email_source=notifications&email_token=BRNHK5GYZ5R5WJJFTPP2QPT5VXSTHA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOJRG43TIMJXUZZGKYLTN5XKMYLVORUG64VFMV3GK3TUVRTG633UMVZF6Y3MNFRWW#discussioncomment-19177417>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BRNHK5GZ5O5HFQUMQRLK62D5VXSTHAVCNFSNUABIKJSXA33TNF2G64TZHMZDCNBVHA3TCOJTHNCGS43DOVZXG2LPNY5TCMJRGI2TCOBXUF3AE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[theluckystrike]

Yes. Supabase Storage can serve images, videos and other files. A URL stored in a database row is just text; the page must also fetch that field and render the media.

Please test one affected item:

1. Open the actual image or video URL from the article record in a private browser window. If it fails, share the HTTP status and whether its bucket is public or private. Remove tokens and query strings before posting the URL.
2. If the file opens there, check the homepage card code: does its query include the media field, and does the component render it? A plain-text article excerpt will not display embedded images or video.

For a public bucket, getPublicUrl() constructs the file URL; it does not make a private bucket public. For private media, keep the bucket private and use an authorised request or generate a fresh signed URL when needed. Signed URLs expire, so a saved temporary URL can work initially and fail later.

Please paste one redacted stored URL, the line that saves it, and the homepage code that renders it. Those details will show whether the failure is access, the saved path, or the page code.

Docs: Serving assets and getPublicUrl.

[theluckystrike]

I checked your linked homepage. For the featured NPP article, the rendered image URL is https://www.citinewsroom.com/wp-content/uploads/2026/10/NPP-Gen-Sec-Justin-Kodua-Frimpong.png. That image is visibly broken here, and my checks did not return a usable image. This URL points outside Supabase, so changing a Supabase bucket policy cannot fix this particular external image.

Update that article to a working hosted image URL, then test that exact src.

This is one example. For any remaining Supabase-hosted media, please share one failing URL with tokens removed, its Network status, and the component that renders it.

[peprah727]

Thanks, I will!

…

On Fri, 30 Oct 2026, 01:02 Michael Lip, ***@***.***> wrote: Yes, Supabase Storage supports images and videos. I can't tell whether the saved link or homepage code is causing this. Please share one stored media URL with tokens and other secrets removed, the line that saves it to the article, and whether the original link opens in a private browser window. Those details would help narrow it down. The Storage guide is https://supabase.com/docs/guides/storage Thank you, Michael Lip Founder of https://zovo.one, 108 Chrome extensions with 150,000+ users Also building https://makeroll.com and https://ml0x.com, open-source guardrails for Claude Code agents Open source has to win, and small fixes like this one are my part of it GitHub Sponsors keeps these fixes coming, https://github.com/sponsors/theluckystrike On Fri, Oct 30, 2026 at 3:10 AM George ***@***.***> wrote: > Please does supabase has the ability to handle files and urls .ive creted > a > website but but any time i post or bublish article or news into my website > homepage, files do not display . Video files included in the article > dismisses, images files, image and video urls do not dispay its contents. > Please help me out , what might be wrong is it settings in supabase or the > serup in my code > > On Sun, 25 Oct 2026, 11:53 Michael Lip, ***@***.***> wrote: > > > This error page comes from Next.js, not from Supabase. It is the default > > page Next.js shows in production when a page throws an error on the > server. > > Next.js hides the real message from the browser on purpose. The number > at > > the bottom (ERROR 500438034) is the error's digest, and Next.js writes > > the same digest to the server log next to the real error. > > > > To find the real error on Vercel: > > > > 1. Open the page again so the error happens again. On the Hobby plan, > > Vercel keeps runtime logs for only 1 hour, so the log from when you took > > the photo is probably gone. > > 2. In the Vercel dashboard, open your project and select *Logs* in the > > sidebar. > > 3. Search for the new digest number, or filter by status code 500. > > 4. Open the entry. It shows the full error message and stack trace. > > > > When a Next.js page that reads from Supabase fails like this, the log > line > > usually points to one of these causes: > > > > - supabaseUrl is required. or Invalid API key. The Supabase > > environment variables are missing or wrong in Vercel. Set them under > > Project Settings → Environment Variables for the Production environment, > > then redeploy. Vercel applies changed variables only to new deployments. > > - TypeError: fetch failed. The server could not reach Supabase. Check > > that your Supabase project is not paused (free projects can be paused > when > > they are inactive) and that the URL in Vercel matches the one in your > > Supabase project settings. > > - Cannot read properties of null. Your code used data from a query > > that returned an error or no rows, for example because an RLS policy > hides > > the rows from that user. Check error before you use data. > > > > Your home page at pink-fm-website.vercel.app loads for me right now, so > > the error may happen only on one page, for example a page that needs a > > signed-in user. Open that page, then look at the log for that request. > > > > If you paste the error line from the log here (without any keys), I can > > help you narrow it down. > > > > Docs: > > > https://nextjs.org/docs/app/api-reference/file-conventions/error#errordigest > > > > — > > Reply to this email directly, view it on GitHub > > < > #52271?email_source=notifications&email_token=BRNHK5GYZ5R5WJJFTPP2QPT5VXSTHA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOJRG43TIMJXUZZGKYLTN5XKMYLVORUG64VFMV3GK3TUVRTG633UMVZF6Y3MNFRWW#discussioncomment-19177417>, > > > or unsubscribe > > < > https://github.com/notifications/unsubscribe-auth/BRNHK5GZ5O5HFQUMQRLK62D5VXSTHAVCNFSNUABIKJSXA33TNF2G64TZHMZDCNBVHA3TCOJTHNCGS43DOVZXG2LPNY5TCMJRGI2TCOBXUF3AE> > > > . > > You are receiving this because you authored the thread.Message ID: > > ***@***.***> > > > > — > Reply to this email directly, view it on GitHub > < #52271?email_source=notifications&email_token=AMFLKPEREPL4NQNLUFFBFK35WOP2NA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOJSG43TMMBQUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSWGM33PORSXEX3DNRUWG2Y#discussioncomment-19277600>, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/AMFLKPCLVVYJQ4Z2YJBIXSL5WOP2NAVCNFSNUABIKJSXA33TNF2G64TZHMZDCNBVHA3TCOJTHNCGS43DOVZXG2LPNY5TCMJRGI2TCOBXUF3AE> > . > Triage notifications, keep track of coding agent tasks and review pull > requests on the go with GitHub Mobile for iOS > < https://github.com/notifications/mobile/ios/AMFLKPGULHV7WFXKCPJH4N35WOP2NA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOJSG43TMMBQUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSVGM33PORSXEX3JN5ZQ> > and Android > < https://github.com/notifications/mobile/android/AMFLKPEAKBJEDPYPLG43MRT5WOP2NA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOJSG43TMMBQUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSXGM33PORSXEX3BNZSHE33JMQ>. > Download it today! > You are receiving this because you commented.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub <#52271?email_source=notifications&email_token=BRNHK5BC43ZODO7Z3RYGQRD5WPSCPA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOJSHAZDEMRTUZZGKYLTN5XKMYLVORUG64VFMV3GK3TUVRTG633UMVZF6Y3MNFRWW#discussioncomment-19282223>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BRNHK5HIAHGHIT233NZQTXD5WPSCPAVCNFSNUABIKJSXA33TNF2G64TZHMZDCNBVHA3TCOJTHNCGS43DOVZXG2LPNY5TCMJRGI2TCOBXUF3AE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[yos88da]

@peprah727 I checked the media on your homepage from outside, and the Supabase side is working:

• All 8 Storage URLs on the homepage, images and the .mp4 video, load with HTTP 200 from your public news-media bucket.
• The broken picture is the NPP article image, and it is not in Supabase. It is linked straight from citinewsroom.com. Their server returns the image (200) when it is opened directly, but returns 403 when the request says it comes from your site. That is hotlink protection: the other site blocks its images from being shown on other websites, so the link works when you open it yourself and stays blank on your page.

What to change in your code:

1. When an article uses a picture from another website, download it, upload it to your news-media bucket and save the getPublicUrl() result on the article, the same way your other articles already work. Only use images you have the right to publish.
2. Stopgap for external images you already saved: add referrerPolicy="no-referrer" to that <img> tag, so the browser does not send your site as the referrer.

If a video goes missing from one article, open that article's saved video URL in a private window and note the status: a 400 or 404 from a public Storage URL happens when the saved path doesn't match an uploaded file.

Still stuck? Our AI agent can fix it in your code and test your live app again. The first check is free. (I'm building VibeFix) https://vibe-fixer.com/?src=gh-52271

[peprah727]

Yes everything was resolved 3 days ago Thank you so much

…

On Sun, 1 Nov 2026, 07:00 yos88da, ***@***.***> wrote: @peprah727 <https://github.com/peprah727> I checked the media on your homepage from outside, and the Supabase side is working: - All 8 Storage URLs on the homepage, images and the .mp4 video, load with HTTP 200 from your public news-media bucket. - The broken picture is the NPP article image, and it is not in Supabase. It is linked straight from citinewsroom.com. Their server returns the image (200) when it is opened directly, but returns 403 when the request says it comes from your site. That is hotlink protection: the other site blocks its images from being shown on other websites, so the link works when you open it yourself and stays blank on your page. What to change in your code: 1. When an article uses a picture from another website, download it, upload it to your news-media bucket and save the getPublicUrl() result on the article, the same way your other articles already work. Only use images you have the right to publish. 2. Stopgap for external images you already saved: add referrerPolicy="no-referrer" to that <img> tag, so the browser does not send your site as the referrer. If a video goes missing from one article, open that article's saved video URL in a private window and note the status: a 400 or 404 from a public Storage URL happens when the saved path doesn't match an uploaded file. Still stuck? Our AI agent can fix it in your code and test your live app again. The first check is free. (I'm building VibeFix) https://vibe-fixer.com/?src=gh-52271 — Reply to this email directly, view it on GitHub <#52271?email_source=notifications&email_token=BRNHK5AWR6SBCWFSTWAFCRD5W3NQNA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOJTGI4DKMRZUZZGKYLTN5XKO3LFNZ2GS33OUVSXMZLOOSWGM33PORSXEX3DNRUWG2Y#discussioncomment-19328529>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BRNHK5AMFQ74FCFG2M7WG335W3NQNAVCNFSNUABIKJSXA33TNF2G64TZHMZDCNBVHA3TCOJTHNCGS43DOVZXG2LPNY5TCMJRGI2TCOBXUF3AE> . You are receiving this because you were mentioned.Message ID: ***@***.***>