Supabase
Replies: 6 comments 3 replies
|
UPDATE: Here is the support thread. If I am to be treated as a free account, I hereby request refund for all subscription fees paid for past 6 months. At this rate of deflection, I bet I can rebuild the project faster than Supabase will even respond. And double the bet that I can rebuild the project before Supa can repair the fatal error, since it would appear Supa is unable to eve fix this known fatal bug Re: ESCALTION 2 after 18 hours of non response on my PAID plan: CRITICAL OUTAGE Supabase ticket SU-26309. Based on this experience, and my self-searching on the persistent known bug , I bet I can rebuild my project faster than Supabase can provide a resolution under its paid SLA, and that seems to be the best-practice approach for builders right now. -A response to my polite request for refund on past 6 months. -- Thanks for reaching out to us. We have received your support request and your ticket id is SU-263951 Free plan will receive no guaranteed support response - We do our best to respond quickly to outage-related issues, but please note that tickets from paid plans are prioritized. Upgrade here. For non-outage related issues, Free plan users are encouraged to ask their questions on our community Discord Server https://discord.supabase.com/ or on our GitHub Discussions board https://github.com/supabase/supabase/discussions . The Supabase Team. |
|
There are some regions that are definitely going to have issues right now... https://status.supabase.com/incidents/5hq69pvqdzmm Did you send your requests with the troubleshooting tab in your instance or direct email. If email then likely they don't know the category of your org from what I've heard. The support/ticketing system uses something from the dashboard support request to tell. Pro does not have an SLA with a time frame that I know of. Not counting these current project hangs for the past week or two they usually clear pro issues in a day, but that is not guaranteed. Right now there is a bottle neck in infra with hardware resources. Make sure your ticket is thru the dashboard and has in the title "Project Stuck Restoring" would be my recommendation. Supabase does not regularly monitor these github discussions. I have a contact that can probably make sure you are to the infra queue. Note to anyone reading this I personally would not do any upgrades/restarts/pauses unless absolutely necessary until this settles down. |
|
Thanks Gary. I am told it is in the infra queue and yes, I created it from dashboard in the sense that the ticket creation was a button next to the error. I'll put in another ticket from dash as you suggest.
If you were a betting man, what would you put the over/under that infra can resolve/restore at all, given enough time? is there a 50%+ chance that this can be resolved inside the footprint of my live project?
I am embarking on the journey to rebuild my entire live site (Vercel-Supa-Stripe-Railway integrated stack). I bet I can rebuild before Supa even responds, but I can't afford to wait only to learn they can't even fix it. Thanks for the help it is an encouragement on a dark day for my app.
…________________________________
From: GaryAustin1 ***@***.***>
Sent: Tuesday, October 7, 2025 8:45 AM
To: supabase/supabase ***@***.***>
Cc: Neo ***@***.***>; Author ***@***.***>
Subject: Re: [supabase/supabase] Critical Outage Affecting SmartMarkets.ai – Supabase Ticket SU-26309 (Discussion #39315)
There are some regions that are definitely going to have issues right now... https://status.supabase.com/incidents/5hq69pvqdzmm
Did you send your requests with the troubleshooting tab in your instance or direct email. If email then likely they don't know the category of your org from what I've heard. The support/ticketing system uses something from the dashboard support request to tell.
Pro does not have an SLA with a time frame that I know of. Not counting these current project hangs for the past week or two they usually clear pro issues in a day, but that is not guaranteed. Right now there is a bottle neck in infra with hardware resources.
Make sure your ticket is thru the dashboard and has in the title "Project Stuck Restoring" would be my recommendation.
Supabase does not regularly monitor these github discussions.
I have a contact that can probably make sure you are to the infra queue.
Note to anyone reading this I personally would not do any upgrades/restarts/pauses unless absolutely necessary until this settles down.
—
Reply to this email directly, view it on GitHub<#39315 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BPTRKOONID3ZKLHCC3UL67L3WO7YFAVCNFSM6AAAAACIPRBOT6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTINRRGU3DCNQ>.
You are receiving this because you authored the thread.Message ID: ***@***.***>
|
|
They all eventually get working again from what I've seen here and Discord. The issue is the infra team has to get to and touch each project and each case is slightly different. And most Pro users in the queue are in the same boat as you with a live project down. So eventually it will be fixed is my observation. The question is when. If you are in the infra queue and not in an impacted region it will be a lot sooner that someone who is still not triaged out of the support queue (especially if marked as free) and keyword search does not catch that it is this issue. |
|
I'm US. I just got this note from support, which sounds encouraging: ((Hi,
I have taken a look at your project and can see that the restore failed with the following error:
* pg_restore: error: could not execute query: error: relation \"qbr_nested_function_source_11\" does not exist
This is due to an unqualified reference to the shown function. You would need to prefix this with the schema like so; "public.qbr_nested_function_source_11"
To be able to fix this I have supplied a copy of your latest backup below. With this you will be able to correct the issue and test locally. Once happy you can restore back to your Supabase project using a client such as psql. To aide you in this I have set your project to being accessible again. Some details on how to restore from the cli can be seen here: https://supabase.com/docs/guides/platform/migrating-within-supabase/backup-restore#restore-backup-using-cli
Backup Link<https://supa-backups-prod-us-east-1.s3.us-east-1.amazonaws.com/ugtepudnmpvgormpcuje/logical/db_cluster-02-10-2025%4007-53-40.backup.gz?response-content-disposition=inline&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEA8aCXVzLWVhc3QtMSJIMEYCIQCE6AscJ8BNynJbwPwu6FdDY6vE6DlMz6dUZ%2FM28PZY0AIhAKyX2ydKLWJkY0Ofj0Ze5a0gKkFxMbVag6yM89nv5frxKtYECKf%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQAxoMNjQ2MTgyMDY0MDQ4IgwhvWaVHCoituH%2F1WcqqgQhMJ3XGXEie4TCdpBehaSME%2F8te9pWsNXZWR6eeaMiPsobTjaQx27QFwnH0kBZUVFOQ76Dzo7wjbHCg%2B2BeLBs2X66LOGnIXfnYV7fO6OclG3kUtdkRKp%2Fk2yjqsoZR1JcASCx1ghs%2B8h75abkCFeik54BmZH%2F7CMx9dTTHD83FVYoOiLghDs28E3%2BCMmTOGNGkzxcpp30BgZJROi29ivIVn0p5H74IUFq6YTaya7x9b%2BGTH14YitM2RZftrbir%2FHOfDs%2FxbOQq3CoiBgn1G4X3dbVTKgtWnQWyygWqtcMyQbB1AIJn8KiJESw7%2BhfxiOtr3JNZEnokZPgSBrt2cMIk%2FO%2BQeb0v0ucczLY5Km%2BTYR2vZ%2FdLhREebO0kQnXf91ntcQxTFKK84KJ7R2ZPrAlpUBJIL9GybEOcNuCX4ya6KIYMh22J%2BtWwXIosCykrAwuney%2BR%2F3CW0QBMWIoxcUAag0VFETpNy31l3EPEz4DJCihaSaoyzDQty9zXqjUUzWWptzPS4oeEOED6mfQm4OqPA9EYvYnSozMtc1VCuy70s0ORrHJF4%2BMx0NExW6BmMKPhKQZY7FKtOp%2Bn3JVz1ZyqKbxECfMPuAuCCVqT3I4L3Q%2BNf5Fd0Jt6UjzqgHed%2F69mijViB%2BJiuvLPbEgqoKQ8HvXCxei%2BnYXLnloAFGGDqkhrSLfl7fRTFpGLl%2FKkiqjzzZa9H%2FYdItPop830vXU7kbvH%2FgGqd5yZjD9mJPHBjrEAm0eraGeTlp1blJgsRjhFzwMhQ5eTxygMMu%2BXFmJKBJPtbrYp0h1Fjf1CjcPb%2FvZ0sRcOJkooCrFYpqIVuQd5qM%2BK8TbWImr5ZDYSYVpj8qnkdYaIlOPb6hgp5wuqRtIIlRsElWfi2%2Fg%2Bwbbb%2BqeFXLN5%2Bcuv5ys2ZZ3Wp3a52iLW9q%2FTzAL%2B8a9SUpKvS8%2FPUBoB5a10dt0pfQBFrW7pWbLjIXeiz9ine6daiX868rQ%2BeYr4upX%2FrwSk3d%2BuGYv9cqWJr%2F9HkxaSNpduO1GkEZkb%2BsdY0lfelg2t8FwPUBI67%2BPNQYXwnAevA3t6WFzsoM3VS2iIfLAPji5jMfnUZY3CuXNiALsqHidHypdOLkwg9JhzfDTNlG1IybjorY%2FtuKrgew4OEmnogxa0Ln0xqjdLiM7vI09PTQWn6LwiXaaH4ks3g%3D%3D&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=ASIAZM43TZ6YP3LHLAA4%2F20251007%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20251007T140928Z&X-Amz-Expires=43200&X-Amz-SignedHeaders=host&X-Amz-Signature=d83a64fbd2f4ee83502be176a38e3dee094732d8dda036b463f1ef741c84db1b> (expires 12 hours from the time of writing). ))
Gary - Any tips on following these directions? The approach that support reco does not involve setting up a new project right? and after restore_from_cli, is the idea that the "stuck restoring will then autoresolve in footprint of current project?" That would be a good resolution!
Best Regards,
Paul C
Supabase Support
…________________________________
From: GaryAustin1 ***@***.***>
Sent: Tuesday, October 7, 2025 9:12 AM
To: supabase/supabase ***@***.***>
Cc: Neo ***@***.***>; Author ***@***.***>
Subject: Re: [supabase/supabase] Critical Outage Affecting SmartMarkets.ai – Supabase Ticket SU-26309 (Discussion #39315)
They all eventually get working again from what I've seen here and Discord. The issue is the infra team has to get to and touch each project and each case is slightly different. And most Pro users in the queue are in the same boat as you with a live project down.
If you are in one of the resource limited regions (AWS/Supabase having to get more resource to handle the explosion in new instances) then that could be slower.
At this time, these workflows are disabled in ap-south-1, eu-central-1, eu-central-2, eu-north-1, eu-west-2 while we work to provision additional capacity.
That list has expanded since yesterday...
So eventually it will be fixed is my observation. The question is when. If you are in the infra queue and not in an impacted region it will be a lot sooner that someone who is still not triaged out of the support queue (especially if marked as free) and keyword search does not catch that it is this issue.
—
Reply to this email directly, view it on GitHub<#39315 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BPTRKOOSIIIBX2GGGS4J2UT3WPC57AVCNFSM6AAAAACIPRBOT6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTINRRGU4TSOA>.
You are receiving this because you authored the thread.Message ID: ***@***.***>
|
|
Sorry, I don't have Pro myself, but this is the first time I've seen someone get a backup from them for a stuck restore. That is progress as there was no way to even get a backup with the dashboard when stuck... You would follow the backup/restore guide they linked. Your fix is to find that function call I assume in other functions or trigger and add public. in front of it. If you go the Supabase Discord there are several mods and users there that might be able to help. |
|
You would eventually restore to your existing instance.... |
|
Thx for the tips Gary, I get it.
…________________________________
From: GaryAustin1 ***@***.***>
Sent: Tuesday, October 7, 2025 9:29 AM
To: supabase/supabase ***@***.***>
Cc: Neo ***@***.***>; Author ***@***.***>
Subject: Re: [supabase/supabase] Critical Outage Affecting SmartMarkets.ai – Supabase Ticket SU-26309 (Discussion #39315)
Sorry, I don't have Pro myself, but this is the first time I've seen someone get a backup from them for a stuck restore. That is progress as there was no way to even get a backup with the dashboard when stuck...
You would follow the backup/restore guide they linked.
Your fix is to find that function call I assume in other functions or trigger and add public. in front of it.
If you go the Supabase Discord there are several mods and users there that might be able to help.
—
Reply to this email directly, view it on GitHub<#39315 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BPTRKOJUROBRLJBHR2BMEWD3WPE5FAVCNFSM6AAAAACIPRBOT6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTINRRGYYTSMQ>.
You are receiving this because you authored the thread.Message ID: ***@***.***>
|
|
Great news, the issue resolved magically without my intervention. Thx Gary for the help and peace of mind. I imagine root cause was the EU backlog, not public* schema prefix, so thanks for surfacing that perspective ...that alone might have helped wait a bit for natural resoluton. Unclear root cause or resolution, but the bug gods are shining down on me today so I will take it as check in win column. Have great day!
…________________________________
From: GaryAustin1 ***@***.***>
Sent: Tuesday, October 7, 2025 9:32 AM
To: supabase/supabase ***@***.***>
Cc: Neo ***@***.***>; Author ***@***.***>
Subject: Re: [supabase/supabase] Critical Outage Affecting SmartMarkets.ai – Supabase Ticket SU-26309 (Discussion #39315)
You would eventually restore to your existing instance....
So you want to make sure you test on local your change you make.
No idea what the complexity of the use of that function is.
—
Reply to this email directly, view it on GitHub<#39315 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BPTRKOM2YZTO63NKXCNW7RT3WPFIZAVCNFSM6AAAAACIPRBOT6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTINRRGYZDENI>.
You are receiving this because you authored the thread.Message ID: ***@***.***>
|
Uh oh!
There was an error while loading. Please reload this page.
Hello Supabase Team,
My app at https://smartmarkets.ai/ (hosted on Vercel, backed by Supabase) is experiencing a severe outage—over 12 hours offline. This is blocking mission-critical analytics for state Rural Health Transformation grant plans (deadline Nov 5).
Active ticket: SU-26309.
(Here is error message: Something went wrong while restoring your project; Your project's data is intact, but your project is inaccessible due to a restoration failure. )
I urgently need:
-Root cause explanation
-Timeline for resolution
-Assurance of active triage
The impact is significant, with direct consequences for state grant deadlines. Please update urgently.
Thank you,
Luke Mizell
luke@smartmarkets.ai
979-484-6949 (text/cell)
All reactions