Free project stuck in recovery after disk full — is Pause/Resume safe? #50084
Unanswered
rgzpkb26vn-code
asked this question in
Questions
Replies: 1 comment
|
Pause/Restore is safe as far as data loss goes. |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
My free project is down and I cannot reach the database at all.
Project: iljpbwmdkfkskbbsomqu (eu-west-1, Free plan, Nano compute)
Postgres is running but never finishes recovery:
FATAL: 57P03: the database system is not accepting connections
DETAIL: Hot standby mode is disabled.
the schema cache")
Cause is known: I bulk-loaded a public reference dataset into ONE table
(public.parcelles_personnes_morales) and filled the volume. WAL replay
cannot complete because there is no free space left to write to.
MY MAIN QUESTION, for anyone who has been through this:
Is it safe to Pause and then Resume the project in this state? Does a
Resume restore from a logical backup (which would come back compacted,
without the WAL backlog and dead tuples), or does it restore the same
full volume as-is? If it is the second, pausing gains me nothing and
risks the data, so I would rather not try it blindly.
FOR THE SUPABASE TEAM, if someone sees this — either of these fixes it:
TRUNCATE that table within minutes and the space comes straight
back — my own data is under 100 MB.
TRUNCATE public.parcelles_personnes_morales;
I have sent two support emails with no reply yet. I am not asking for
free resources, just for one write so the instance can start again.
All reactions