Did you deploy a template or is this a custom service?

(refering to NocoDB)

Template

But this is very old for 1 year it was working perfect and from yesterday because it crashed i cant deploy it again

Ah I see, in that case you can go to the template you deployed and on the right you can select Get template help, this will request help from the template author (since this seems to be an issue with the template having broken or something)

ooh that doesn't look good

i dont know much (or anything for that matter) about NocoDB unfortunately so I can't help much here 🥲

is there railway support that can help with this?

yea someone who knows more about NocoDB might jump into this thread hopefully

in the mean time I have one idea

can you send a screenshot of the top of your settings of the nocodb service?

Yesa

I saw nocodb had update yesteraday and im trying to deploy with the latest version but still it crashes

oh so you edited the source image?

my theory was it was on the latest tag, and so upon redeploy it deployed a new version which might've not been compatible with the previous version causing the crash loop

ah so i should find the version it was working before?

It was on latest originally , yes

Should i deploy from backup?

oh neat, and yes, if i were in your shoes I'd find the version it was on before it started crashing

those only backs up the volume content, not the service settings or deployment environment

you'll want to see what date and time your last healthy deployment went out, use that to find what the latest version at the time was and then move back to that version

if you got it back on an the same version it used to be and it still crash loops then I'd recommend restoring a backup (cause the new version might've made the files unreadable for the old version, potentially)

yea, just find what version was latest 11 months ago

okay so im looking for the version 11 months ago

exactly

Not working, i will try deploying the backup

alright! hopefully that works

Here is something strange

Yesterday's backup is 3.34GB

was yesterday's backup made on the newer version?

Most likely. It's auto backup

the newer version must've made huge changes to how storage works, as for how it got to 3.4gb is beyond me

So i rolledback the nocodb version , mounted the old storage and it says "online" but still its not accessible.

any errors in logs?

no 0 errors

tried deploying with the new volume but gave error. Now trying to deploy again with the old volume and it gives error...

so upon deploy it had no errors, then the next deploy with that volume it got errors again?

yes

Now its online

but not accesible

that service is in a very unstable state wow

are you sure that is the correct version?

this is the old version i deployed 11 months ago

then I'm out of ideas

This is intresting

that is some odd ordering 🤨

probs just a UI issue

network is not started so thats why i cant connect.. anyway i will update if i find a solution

your serivce has healthcheck endpoints, I don't think it'd start if the network wasnt working

well, i guess considering your service is inaccessible, those healthchecks arent doing their job 😓

unless you removed the healthcheck endpoints

I fixed the issue .

Redeployed brand new nocodb. And just replaced the variables so its connected to the original database

Nice! Glad you got it working