garyaustin

There are no discount plans. I don't see how realtime would be cached egress. That is almost surely storage. Check you usage charts and click thru on egress it should show you what is using it.

The site is English only please.

You just have to see if a user mod or user comes along. I would provide though more info on your frame work and session handling (cookies, create client). That is where the problems have occurred before where the user server code does not cleaning start a new createclient with a new session and picks up old session. These were not strictly signup but ssr pulling a session from the last user to make a call versus the user now making a call. Coding errors and not a bug in SSR though. I would be very surprised if Supabase auth is returning a matching UUID to an existing user unless you do have the same email being sent of another user and this is supposed to be a separate SSO user with the same email. That should not be the case, but would seem more in the realm of a mixup than two different emails. Here is one example: https://github.com/supabase/supabase-js/issues/1682 Is see several possible hits on the web for `supabase ssr user session from another user` on the web.

You probably need to give more details of how you do auth and framework. Easy on server side code to reuse a client with a previous users session if not done right. I won't be the one to help though as I don't do SSR or serverside auth. If this happened with browser based auth that would be odd but I also don't use OAuth. Certainly not a known bug and not heard of user session mix up on just browser auth.

I really don't know much about Vercel. If like Supabase there is no escalation besides support in dashboard and email. Add Vercel to your title maybe a Vercel user will come along with a suggestion. But only support of one of the two companies (likely Vercel as 2fa normally means you are out for good if you lose your factors or at least have to prove ownership without email being the way).

I assume you sent that to both Supabase support and Vercel support? support@supabase.com... don't know Vercel's. Supabase does not have much flexibility if I understand on manipulating accounts "owned" by Vercel/Lovable/etc. No one at this community forum can help resolve it.

?

This is usually a sign of Disk i/o burst being used up (or exceeding cpu/memory. When Disk i/o burst is used up then restart still faces the same issue. It slowly replenishes over 24 hours. Sometimes upgrading to a larger CPU size temporarily allows getting over this. BUT there are other cases when overload needs Supabase help.

The employee has not marked it as seen. I won't really get any updates other than that. And no one else is complaining about jwt types and storage. There are 2 other threads here with users having some sort of issue with auth.uid() and storage, but not related to jwt type. So right now you are still in the Support normal flow which we have no insight to.

Very unlikely a Supabase thing deleting a single row in a postgres database. There are no logs for deletes in SQL (dashboard and database ports). Only REST API would show up.

You would need to contact them via your new email and point to your old email. support@supabase.com. This is a community helping forum, only support can solve this.

You will need to contact support. They might need other info from you as you can't confirm the email.

Support is the way. Why would your project be deleted in 5 days?

https://status.supabase.com/incidents/308hm84ntd47

This is a community forum and not monitored regularly by staff. Supabase has said though there is no way to restore a deleted project. This is why they have you type in project name or org code before doing the delete.

Free support can be days especially on weekends and if you did not get a 2nd email a few hours later prioritizing based on automation. How long ago did you submit ticket?

OK FYI a couple of devs poked around and between the reddit thread and this info here earlier not clear they found anything. Or at least they did not post anything back to me. Not a formal path. But your ticket was listed there. Everyone seems to be slightly different on the issue being edge functions not working, just edge functions showing healthy and they may or may not even use edge functions, us-west for 2, unknown for reddit (which I don't have an account), some DB issues, etc. Not sure what they will do with it. No other reports here or github so this is some really small thing or just random issues being lumped together....

So your DB is having issues being accessed from your app versus edge function issues reported by the OP? Just your dashboard only shows edge as unhealthy and everything else as healthy?

Just wanting to add detail to an escalation on if just UI showing unhealthy or actually using the function is impacted.

Not sure what you mean. No trouble running but erroring? You mean when they fail you can't get to them to run?

<@693193631527469207> Are the edge functions erroring or just showing unhealthy in the dashboard?

<@693193631527469207> Can you drop your ticket # here.

Then definitely should be making support requests. No idea what the common issue is if any, but even with that list, it is very small so not a mass issue and something specific to part of a region or service provider in common. There is not reported outage from Supabase on their status page

Seems like some security on your net or stale DNS. No reports besides the one above.

https://discord.com/channels/839993398554656828/1445840597305790628/1446618138014781491 this one from above. It links info from Carol.

The link should have tools.

You would need to contact support. No one here can see your project.

I've never seen anyone go straight from Lovable Cloud to selfhosted post about it here and not searched the web. There is no built in process from Lovable or Supabase to migrate even to hosted Supabase. There are some 3rd party tools people have done and likely external guides. All will basically do backups of the DB, reconfig of the instance, move the edge functions and copy storage files. The difference would be you would be learning how to do all that in a self-hosted setup versus the hosted one. Steps would be similar, but self-hosted has different config methods and edge functions are not really edge functions without you working on deploying your own (they run on your server not the edge). https://discord.com/channels/839993398554656828/1445840597305790628/1446618138014781491

Where are you located?

There is no other access to storage or temp setting to resolve. I think people have even tried transfer the project to a new org but not sure about another org in a 2nd account. Pro for a month is the typical way or wait until end of period if that what the message says.

Yeah, calling signInAnonymously() will create a new user.

That is the intended use of anonymous users. They only last until you signout. That is by design.

<@917834098205134898> Unfortunately Supabase has said they cannot restore deleted projects as the warnings say when you delete. This is why deleting has you type in project/org info to confirm. This is also a community forum and not monitored by staff regularly.

You have to wait on support. ibrahim added your ticket # to the informal queue we have but only one employee checks it when he is arround and it is the weekend.

Not sure why you are using the refresh token as a check. That is used to get a new JWT and can only be used once. It does not expire until used.

The password reset template with default wording seems to get caught up in spam filters enough that it comes up here as the one email that has issues while others work. But with out errors or logs or trying it to see, I'm just throwing out one possibility.

There is an open status on Brazil and one provider... but it has been that way for a long time... DNS... https://status.supabase.com/incidents/byfbl0lj2cc3

No one has reported any issues here with us-west2 or their instances that would lead to thing there is an outage. Sort of unclear what is unhealthy for you. Did you try a restart? Did you try a VPN to rule out something local to you? Dashboard issue or customers not on your same network having same issue?

If you drop or dm just your instance url/id I'll do a quick check if API resolves with error or DNS error from Texas just to rule out something on your end. And there are no other issues I've seen along these lines here or github.

Did you try a project restart? Try VPN? Sounds like you tried other networks which was next thing.

This is a community forum. No SB staff will likely see your post here.

<@1512894524211593308> There is no evidence the op has this issue with upgrade... unhealthy can happen for many things and usually is overload of DB. The 2nd poster has another thread, but more importantly a github issue going. You should at a minimum add to the github issue.

Your current ticket will get flagged as pro.

I can't check the ticket. I just add it to a thread in a private section here that an employee checks when he can. He will flag the ticket as appropriate. All I will know is when he has seen the queue and done something.

Usually Pro support tickets get much faster response than free. But everything is slower on the weekend. If infra team has to look at (versus support sees that you did something wrong) then Pro would be ahead of all free projects in the infra queue.

Unless another user has ideas the support ticket is the way.

This is a community forum all I can do is use an informal queue with your ticket # an employee checks to flag critical issues that likely need Supabase to resolve.

You can drop your ticket # here and we have an informal path to get someone to notice it, but it is the weekend.

Try the restart when you can. If that does not work drop your ticket # here and we have an informal path to try and get it found (but it is the weekend).

Are you free? Did you get a 2nd automated email about being moved? Did you try a project restart?