I ran a recursive query and now there is no more space on device
I am getting this error:
2024-03-29 22:10:54.413 UTC [8] FATAL: could not write lock file "postmaster.pid": No space left on device
but i can't access the database with psql or anything to try and remove the uneccessary data. This is resulting in downtime for us.
15 Replies
Project ID:
N/Ahey @themkrage what is the current size of your volume?
50GB
It looks like the recusrive pg query i ran filled it. It's not like the database has 50GB's of content, just that one query appears to have filled the container
i don't see an option to increase the cap, and im not sure how to clear the "in-progress" query and data
i can't connect to the database either, so i can't run
VACUUM, which i think would fix thisOkay ill do my best to get a team member here asap, in the mean time please provide your project id
/98f66160-ce86-4b15-a73b-cab912256208
specifically,
https://railway.app/project/98f66160-ce86-4b15-a73b-cab912256208/service/30c393cd-fc3a-4ed3-a433-9017f7032947/variables
Railway
404 - Page not found
Railway is an infrastructure platform where you can provision infrastructure, develop with that infrastructure locally, and then deploy to the cloud.
Hello, working on expanding your cap now
One sec
thank you!
Okay, try now
Solution
Should be up to 100 GB
thank you! able to connect. now running VACUUM
thank you. Running vacuum brought our disk usage back to normal expected level
For future reference, is there no way to access a database that has hit the disk usage cap?
correct
postgres more or less locks itself up when it runs out of disk
is there some internal safe guard we could add to prevent this?
there is nothing that i am personally aware of