While I cannot help you debug configuration or application level issues, I can speak to the 85GB number.

That would be the memory in use by the virtual host machine, Railway currently uses a container based orchestrator, the side effects of that is the shared kernel, allowing you to view the memory in use by the parent host VM.

Yeah, i figured it was something related to that. I don't suspect it's an application or configuration issue tho, I'm fairly certain it's an actual railway UI/reporting issue. The estimated usage billing is even stating it'll be about 1/10th the cost of what the memory usage says it is.

Link to the service please.

the project ID?

A link to the Railway service.

pretty new to the platform, not sure what you mean by a link to the railway service

Just open the service and send the URL.

https://railway.com/project/a6915c70-5af9-4d71-ab08-1e212a2350ce/service/eeef17f1-f48a-4071-b334-8717c2856c77?environmentId=64ae1f94-1704-4239-b271-ef53b7b44108&id=bb24c315-fcbc-402a-b993-e921eaf4a873#deploy

that work?

Yep, the UI is correct, here is the raw data points for the memory usage graphed (6hr window).

Looks like the UI is even rounding down.

isn't it $10/GB for memory usage? i fixed a memory leak I had initially launching and now it shows estimated usage at only $6.50 more than the current usage so far

I figured it it was accurate the estimated usage would be a good amount higher

There are only 7 days until you are going to be billed, so you aren't going to see the $20 that I think you are expecting to see in the estimate.

If you used a flat 2GB for the entire billing cycle, sure, but you haven't used a flat 2GB for the billing cycle.

yeah I suppose that makes a bit more sense, everything else reduced pretty significantly after fixing the leak as well

i know you can't help debug, but is there anything normal about seeing a local nextjs project go from 200mb local to 2GB prod? only a few months into leaping from constant wordpress builds into nextjs land and learning a ton. Haven't seen any of my other projects have anyway near that kind of jump.

in the process of migrating ~40 sites to Railway btw, great platform

It looks like this was something you introduced on Aug 14th at 6:39 PM UTC, at that point memory really started to spike, climbing to 2GB+.

Just another day as a JavaScript developer, to be honest, and I'm not even joking.

🤣 yeah it's been fun so far. whole other world when you have to figure out how to debug issues only in production lol

appreciate the help

No problem. Sorry I couldn't do more for you, but we only have so much information on our end. After all, we are just running your code as is and reporting on its usage.

no worries, don't expect you to debug stuff. unrelated question, have any experience getting wp-cli to work? since it's a separate image the only thing I can think of is to detach the volume from the main wordpress image and attach it to the wp-cli image during use.

Never heard of "wp-cli" so I'm going to have to say no.

I can bounty this thread, and then perhaps a community member may be able to assist further?

not sure exactly what that entails but anyone wanting to use WP on railway is going to be interested in that question. it's the official way to run search/replace or cli operations on wordpress.

sorry for being so green 🤣 thanks man

I shall bounty this thread then.

We (Railway) will post this thread with a credit value. Whoever answers your question will be awarded the credits.

-> https://discord.com/channels/713503345364697088/1372169430091698289