Postgres crashes after config
I ran the following command:
ALTER SYSTEM SET shared_preload_libraries = 'timescaledb,pg_stat_statements';
and my deploy is crashing:
PostgreSQL Database directory appears to contain a database; Skipping initialization
2024-04-29 12:31:11.626 UTC [8] FATAL: could not access file "pg_stat_statements,timescaledb": No such file or directory
2024-04-29 12:31:11.626 UTC [8] LOG: database system is shut down
how can i fix?48 Replies
Project ID:
6f1d26dd-c104-43d6-8f3d-2f6bc6f9042c6f1d26dd-c104-43d6-8f3d-2f6bc6f9042c
i cant set the previous value:
shared_preload_libraries = 'timescaledb';
did you have a backup?
@Brody, actually i don't have a backup
is there important data on the database?
@Brody yes, its company's production database
i found this user with same error: https://discord.com/channels/713503345364697088/1119792911819751566
it's seems your engineers change the config file
yeah but the only difference is that they are on the Pro plan
oh no 😦
I'll do my best to help you remove that config yourself, but I make no promises
going forward please upgrade to Pro as well, you should not be running a company's production database on a Hobby plan
i'll appreciate that
if i migrate for pro right now, helps?
while I could get the team involved if you upgraded to pro right now, there's no guarantee that they would be able to fix it faster than we could
what is the source image for your database?
i understand, then i'll wait and migrate for pro after this
it's railway's postgres base template
railway has a few, please tell me your database's source
im trying to work out a plan of action, so id need to know that
i found this on project: https://github.com/railwayapp-templates/timescale-postgis-ssl/pkgs/container/timescale-postgis-ssl
GitHub
Build software better, together
GitHub is where people build software. More than 100 million people use GitHub to discover, fork, and contribute to over 420 million projects.
what version
pg 15
thats right?
i take it this is a migrated database?
yes! from mandatory migration that occurred at jan/2024, right?
how much of the volume is used?
are you referring to disk space and memory?
just disk space
less than 2 GB
if im able to work out a plan of action would you be down to let me do it?
yes, of course
know that i dont take any responsibility for data loss
ok, no problems!
i was able to recover a database with the same error you are facing, if you want to add me to the project i will get started
yes, i will
you have a backup service?
yes, from railways s3 backup template
but only bakcups once a day
from last backup to now, there was new data added
gotcha
btw your backup service is using the database's public url, meaning you are paying extra egress when a database backup runs
oh, that's interesting. an alternative would a private url?
i can change that if you'd like?
ill appreciate it!
Solution
database is back up
omg, worked. and was no data loss!!
backup service updated too
thanks a lot, Brody!!!
no problem!
non admins (me) cant delete stuff, so you can delete these two things
i'll migrate to pro now. i'll need to redeploy some service or additional configs??
after youve upgraded you will be asked what projects you want to move over to the pro workspace
ok, i'll do it
thank you one more time and have a nice day!
no prob, and you as well!