Projects that are created with vercel will become deleted unfortunately if you disconnect or delete vercel
Usually you have to give it 24-48 hours for everything to clear. If you still haven't seen things come back then reach out to support@supabase.com
There are apps out there that are doing similar things btw, https://supabase.com/partners/integrations/dreambase this is one example that comes to mind but i'm sure there are others.
For Supabase MCP doesn't it gate behind feature groups instead of individual tools and then also allows restricting to read only? If there are changes to thins like the schema and what not i don't think i would have an impact to say the database tools and how they are used? Please correct me if i'm misunderstanding your point btw
Could you expand a bit more on the invalid logins and not being able to get in? How did you eventually get in? When you did get in, did you see your organisation/project but no database data or did you not even see your organisation/project at all. Could it be that an Ai model that you connected deleted some of the data?
This project looks like it is an internal tool, i'm not sure how open it is for external contributors
For the connection string i think you can make it yourself if you just follow the patterns laid out in the top of the dashboard connect tab provided you have the project id and the password. I'm assuming you have the project id to hand after you create the branch and then running `supabase branches get <project-id>` seems to give you the password? Does that work for you?
https://supabase.com/docs/guides/deployment/branching#:~:text=They%20are%20automatically%20paused%20after,staging%2C%20QA%2C%20or%20development see the section about preview branches for pausing. In terms of having multiple branches it gives you a cost when you create a new branch <img width="553" height="96" alt="image" src="https://github.com/user-attachments/assets/0cf06192-f901-4863-952b-2de087a6bf96" /> and it is per hour so you can calculate from that. I believe the max is 50 and the exact per hour cost will depend upon your instance size.
This is a very hacky way to do it but if you create a function, the end point and slug are created so in your case it would be `send-email` then if you were to go into the edge function -> settings -> name you can modify the name without changins the slug/endpoint so i think you can change the name to `Sends transactional emails after user registration` but invoke it as `send-email`. Though this might make things a bit ugly if descriptions are very long
https://github.com/supabase/supabase/blob/self-hosted/minimal-config/docker/docker-compose.min.yml this is a link to a docker composed file that is more minimal but you should be able to edit your one to remove logflare
Do the postgres/storage logs highlight anything? There was a similar issue in Supabase discord thread https://discord.com/channels/839993398554656828/1499260918528213012/1499426986957869117 which a user managed to resolve otherwise you might need to reach out to support in order to get a resolution.
Yeah i think that is part of the standard functionality. You should be able to do persistent branches if you would prefer that on merging it does not remove the merged branch.
Judging by this https://supabase.com/docs/guides/platform/migrating-within-supabase/dashboard-restore it does mention ```The backup file will be gzipped with a .gz extension. You will need to unzip the file to look like this: backup_name.backup``` and from there i think you run the backup. Did you try and target the gz file before unzipping or are you saying that backup_name.backup did not help much?
Are these users using auth tokens signed by the old JWT perhaps or have you not revoked the Legacy key yet as an option? Sometimes when auth get's into this state you can try resetting the project to "force reboot" or reach out to support to see if its an infra level issue
Sorry what is the issue you are seeing, i can't make out the error from above. Did you intend to attach a screenshot? Also if you are migrating away from lovable cloud, it is not so easy to do that.
If this was working before then you should leave an issue on cli but i'm wondering about the the use of `new URL` instead of just doing `Deno.readTextFile`
wouldn't your custom SMTP have logs?
Based on what you are asking for could you not use an auth trigger https://supabase.com/docs/guides/database/postgres/triggers to populate any information that you need when an auth user is updated e.g. `auth.users.is_confirmed_at`
Could this be a vpn or network related issue? Try using a different network like a phone hotspot to see if it gives the same error
What is service a in this example? Are we still working within the same app context or are you using different schemas for different frontend projects?
This seems like a state bug issue that users occasionally run into and is not that reproducible. You could try using a different browser/device/vpn or you could try signing in with github/google. This would link your accounts
Interesting request, what is the nature of the application that is being built that doesn't require a redirect to somewhere? Are you building a web app, an app, newsletter?
I believe when i user clicks to confirm an email then it goes to a supabase endpoint which confirms the user and the need for user redirecting arises because it can't just leave them on that endpoint which i don't think has a UI. What exactly is your application doing with auth to not need redirects?
You don't need to call `getUser`, I was thinking that you would capture the email that the user first added in and pass it along when you redirect to `/verify email` like passing some props through, if you want maybe you can also store it in a cookie or local storage but i'm not sure that is neccesary
<img width="763" height="644" alt="image" src="https://github.com/user-attachments/assets/09d154bc-e124-4568-8ae1-c16880dd2b47" /> when you specify a column type you can specify random UUID and then it comes up as a default when you enter a row <img width="675" height="175" alt="image" src="https://github.com/user-attachments/assets/e9cfbd0d-41ff-4c73-a54d-fefcb01af600" /> so you don't have to add anything. How would what you propose differ?
I think you have 3 flows here: Flow 1 User signs up with email/password -> redirect to /verify-email -> ask user to put OTP in sent in email -> call verifyOtp (the email is already in the context and you can fetch it from the returned user object) Flow 2 User signs up with email/password -> leaves site -> returns < 1 hour later -> tries to signIn -> Met with email not confirmed error -> redirct to /verify-email -> user fetches token from email -> verifies email and signs in Flow 3 User signs up with email/password -> leaves site -> returns > 1 hour later -> tries to signIn -> Met with email not confirmed error -> redirect to /verify-email -> user fetches token from email -> tries to verify Otp -> met with expiry error -> let user resend Otp to email -> let them add otp to input field -> signs user in In all of these flows we know the email in question because the user has supplied it themselves, all we need to do capture the correct error at the time of request which i think you just need to but `try catch` blocks around signUp/signIn and make match on the message to determine what are the next steps. I think the above would work but please let me know if i've made a mistake somewhere or i'm wrong about something
I have a few ideas: 1. The first is, why don't you not ask for the password right away if the otp if important. You could try calling `const { data, error } = await supabase.auth.signInWithOtp({ email: 'example@email.com', options: { emailRedirectTo: 'https://example.com/verify-email } })` then you have to change `{{ .ConfirmationURL }}` to be `{{ .Token }} ` which will give a token in the email (update the email template to reflect the change and not the default here is 8 digit code which you can keep or replace). After that if you do verify the code on that page with something like `const { data, error } = await supabase.auth.verifyOtp({ email, token, type: 'email'})` you will end up with a session and then you can try and update their password once they have a session with `const { data, error } = await supabase.auth.updateUser({ password: 'new password' })`. If the user comes back later then if they have not confirmed their account yet you can just call signInWithOtp again after a button called "resend code" is clicked. 2. Another idea and i haven't check this myself but worth trying, what happens if you swap out `{{ .ConfirmationURL }}` to be `{{ .Token }} ` in the email template for the normal username and password flow. Does that provide you with a token that you can then call supabase.auth.verifyOtp on? If so then you sign up with email/password -> redirect to verify email -> ask user to enter token -> verify otp. Provided that the user has signed up but not confirmed their email, the email_confirmed_at field on the user object will be null so that can keep track of where you are in the flow. Supabase won't return a session until the email is confirmed. Let me know if that makes sense and also let me know what you decide
I think that you can't go back to nano once you upgrade to pro judging by this here <img width="321" height="219" alt="image" src="https://github.com/user-attachments/assets/3e182f4a-41da-431e-83e6-dc95e6e08097" />
I think for such an issue it is only support that can resolve it, they aim to get round to all issues even if on free which hopefully they will do soon enough.
I think reach out to support and explain what happened, if you still have access to the email you originally signed up for, they may be able to locate the project(s)https://supabase.com/dashboard/support/new
https://github.com/orgs/supabase/discussions/35895 this looks like a similar github discussion about this topic. It means a few solutions so perhaps one might help?
I haven't seen anything down the pipeline mentioned. There is a discussion here https://github.com/orgs/supabase/discussions/37765 if you would like to also contribute. Would probably make the issue more on the radar.
Is this Claude Code that is running in this sandboxed docker harness any different from claude code the cli? Since the claude code in the cli https://supabase.com/docs/guides/getting-started/mcp allows you to scope the supabase mcp server to have read only permissions and this is the newer way to configure mcp servers instead of access tokens. However, maybe this current setup requires the PAT?
If you haven't done it already i recommend posting this in the #jobs channel inside of discord as there tends to be a bit more activity there
These are a bit old so not sure how helpful they would still be https://supabase.com/blog/using-supabase-replit and https://blog.replit.com/making-real-time-chat-app-with-supabase-on-replit
Any chance you are affected by the current issues with ISPs in india https://status.supabase.com/
I had a look on my side and it found the same issue. Perhaps this is a bug, in which case you can add it as an issue in the cli repo https://github.com/supabase/cli/issues
I'm not sure what claude is referring to here but the defaut for Supabase is IPV6 https://supabase.com/docs/guides/platform/ipv4-address. I'm not sure if Railway is one of the ones that needs IPV4 but if you head to the connection tab in the dashboard home page you can use different connection strings for your Database including the session pooler for IPv4 <img width="313" height="40" alt="image" src="https://github.com/user-attachments/assets/aca67739-9fc2-4b3d-a976-61807cf8a8ea" />
Would cloning the Supabase Github repo locally help?
What is your setup currently. If you are using the main branch to deploy to staging and prod, then if you have Supabase connected to Github wouldn't you update your prod database whenever you push a change to the main branch before you even get to the staging deployment? I'm wondering if the issue is less branching but the fact that any pushes to main is not necessarily a push to your prod database?
Might recommend adding thoughts to this branching discussion here https://github.com/orgs/supabase/discussions/18937
looks cool, there is a showcase channel on discord also if you want to share it there
For what it is worth this feature does already exist i think but only for team/enterprise https://supabase.com/docs/guides/platform/access-control
It is saying that `site_content` could not be found. I assume you just need to create the table and it's columns.
Reach out to support i reckon is the best plan of action, they'll have more visibility into internal changes like that
Unpausing a project wouldn't cause any errors. What was the reason it was paused? Not used for 7 days or did you pause it?