sambhav
Does railway observability support for log level=warning ?
I have a NodeJs / svelte kit project and I am using just the standard console logs.
All the
console.log appear in railway observability tab as info (blue). whereas errors appear as read (which is correct.
now for certain logs I want to mark them as warnings (preferably in orange color) as they are somewhat important to see but not critical.
If I use console.warn in railway it still shows as red and are marked as errors.
Is it possible to mark the logs as warnings in railway's observability tab and if so how ?5 replies
Not able to bring back a couple of CRON jobs to live status
During the outage today all of our services were down.
I redeployed now and they seem to be up now except for a couple of CRON jobs that still show in error state.
I do not see a way to redeploy those jobs. If I click on 'run now' then it shows an error of ' Cannot redeploy without a snapshot '
note - the business owner changed the region for all services as he thought the error is only for us-west. not sure if this has created a problem with the cron jobs services
How to I make the two services green state now ?
17 replies
Is there a limit to how many custom domains can be added ?
I have created a multi-tenant application for my client and based on the custom domain name entered in railway settings the application routes to different places.
I have a question on behalf my client that is there a limit imposed by Railway on how many custom domains may be added to a single application?
8 replies
How to check which region is the database in and how to request to change it (if needed)?
We started on the hobby plan and now we are Pro plan.
For GDPR compliance we need to check which region is the database in and how to move it to EU (if needed).
Client asked to just collect the info right now, no actual migration needed.
Can you please tell me how to check it?
8 replies
Added a template as CRON service and it shows "Your project has no deploys"
I added this template to take automatic DB backup - https://railway.app/template/UGKaB8
I added a CRON schedule for it. It shows that it is supposed to run every 15 mins but it also shows "Your project has no deploys" for this service and I waited for more than 1/2 hour and it doesn't seem to run that CRON job.
what am I missing ?
14 replies
service in only one environment?
Is it possible to have certain service(s) only in specific environment?
I am making a new cron job that is needed only for production. I have a dev environment where it is not really needed. is this something possible ?
10 replies
Automated Postgres backup onto a volume ?
Since now the project I was developing for a while is almost ready for production, we need to make sure that we take automatic DB backups everyday.
I found this article on blog which shows how to do it with an S3 bucket - https://blog.railway.app/p/automated-postgresql-backups
My question is though, Since my client does not already have an AWS account, and since now there are volumes support available from Railway, will the following solution work ?
Create a new service and mount a volume to it. Run that new service on a CRON. Within that service, take database dumps and store it onto the volume ?
Some things I haven't thought about - How to actually restore from backups? how to download the backup (if needed) to local machine. Keep only last N backups and delete the older ones
(All of the above are possible with S3 already).
5 replies
The new log explorer is great, Is it possible to show the service name in there ?
Currently in the log explorer the service name is an ID and I was wondering if you plan to show the service name in there same way as the service is named in the dashboard ?
7 replies
What is a good / recommended resource use for a web app? Plus question about scaling.
I am deploying a few applications for a client of mine. It is a multi-tenant app made in sveltekit.
Right now there are no actual external users, only a couple of people testing it and what I see is the memory for the frontend (sveltekit) app is about 100Mb mark (and similar for the API which is nodeJS/express).
I am on Hobby plan right now but as soon the app is complete we plan to move to PRO.
The client is not clear on how the scaling will work. The PRO plan mentions that it includes horizontal scaling, but do we have to edit any settings or buy more CPUs ? OR will the scaling just automatically work ?
13 replies
problem with pnpm lock file
The project is a few months old. Everything was running smoothly.
I added a new dependency in the project and now the build is failing. (works locally on my machine).
imp thing to note is that I updated my node version on local machine, not sure if this causes the problem.
The error I see on railway log is as following:
I have already tried running
pnpm install --no-frozen-lockfile but that doesn't solve anything.7 replies
"Can't reach database server" from local machine due to using a wifi which has a connection wall.
Everything works correctly in my home office. I am travelling this weekend I am using this Hotel's wifi which actually shows a page where we need to accept terms before using internet.
I am writing post using same wifi so I know everything else is working correctly, browsers, other apps etc.
Just that railway CLI and any sort of connection attempt to database is not working.
Any idea how do I get around this problem ?
even issuing
railway login results in a 404 error in terminal.
I am using Ubuntu Laptop.10 replies
Deployment is stuck in "deploying" state. Tried redeploying but it is still getting stuck
I have a simple sveltekit app. It has a few deployments in last 2 months and everytime it deployed within 2 mins.
When I added a few more commits today it goes to a "deploying" status and just remains there forever.
Is there any outage going on ?
In the deploy log the last few lines reads ::
5 replies
Is there a possibility to set an upper limit for billing per-project basis ?
Usage-based pricing are good, but I want to ensure that I don't incur charges which I cannot pay.
Is there a way to set an upper limit for the account on a per project basis (or at least at account level) to ensure that the billing is not going to be a surprise ?
thanks!
8 replies
Is it Possible to see logs and metrics of all services in one single place?
Hello All, I posted this in quick-qs earlier, but I think it is better suited here.
Background: I am setting up a new project which will have an API server, another server to deal with messaging email etc, probably redis and a couple of frontend services (nextjs). Last time I did something similar I built it on kubernetes but I want to do it on railway this time to save me all the cluster management headaches.
My question is: Is it possible to see the logs and health/matrix of all the services in the project in a single place ? In kubernetes I did it with prometheus, grafana and loki which would aggregate everything. how do I set this up so that I can look into one single place if I need to see if everything running fine and if I need to debug with the logs ?
I see there is a premade template for datadogagent, but I had trouble with datadog previously plus it consumes too much resources to run the datadog agent and it is not super easy to configure. An ideal solution would be something which automatically works with minimal configuration. We have some budget to pay for a service but not a lot at the moment.
Any pointers in right direction will be highly appreciated.
Thanks.
5 replies