honzasterba
Unable to connect to PSQL
I am having trouble perform DB backups (running outside Railway), for some DBs I am able to connect but for this one I am getting a timeout. Any clues what could be causing this?
https://railway.app/project/06f3919c-f0af-4531-8cf3-97d17baf0d80/plugin/0c85bfb1-2499-4463-87a4-8fd8ae8013f8/Connect
failed: connection to server at "containers-us-west-43.railway.app" (104.196.248.235), port 6010 failed: Connection timed out
19 replies
Trouble starting Typesense - volume not writable?
Hi,
this is painful to debug, I am also in cotact with the typesense comnunity here:
https://github.com/typesense/typesense/issues/1278
affected service
https://railway.app/project/09c93cce-d2c9-4843-80e8-d068e6719f1e/service/0db5ed7f-6ac5-4619-8521-907cd610b023?id=e321b389-43e9-4a07-a543-3890698b096d
I am trying to create template (and service for myself) from the typesense doceker image. It requires a volume, but even if I do not mount a volume and use /tmp for data storage the image fails to boot.
I have been asked to verify that the mounted path is writable (duh?!) of course it is, but how I am to verify that it REALLY IS?
In non-railway setting the typesense docker image works just fine.
39 replies
How to change the number of replicas via API?
1) the API introspection in postman does not seem to know about the possibility of updating numReplicas (parsed from dashboard requests
2) I am new to gql so I tried a req that would work:
mutation serviceInstanceUpdate {
serviceInstanceUpdate(
serviceId: "ID",
input: { healthcheckTimeout: 300 }
)
}
but doing the same for numReplicas fails:
mutation serviceInstanceUpdate {
serviceInstanceUpdate(
serviceId: "ID",
input: { numReplicas: 5 }
)
}
with "message": "Problem processing request"
39 replies
Postgres could not resize shared memory segment
This comes up when trying to run VACUUM on some of our tables, so far its not a big deal but could become in the future.
The fix is to increase shared memory size limit for the postgres container.
Wondering if anyone else has had the same issue, since response from Railway was "we do not allow increasing SHM size".
Project ID: 76e242e9-db08-4279-a677-6c0063fdb9ce
2 replies
Postgres could not resize shared memory segment
This comes up when trying to run VACUUM on some of our tables, so far its not a big deal but could become in the future.
The fix is to increase share memory size limit for the postgress container.
Wondering if anyone else has had the same issue, since response from Railway was "we do not allow increasing SHM size.
3 replies
Unable to run PG VACUUM: could not resize shared memory segment
we have been getting
PG::DiskFull: ERROR: could not resize shared memory segment "/PostgreSQL.1528362745" to 717476224 bytes: No space left on device
Which I found out could be related to shared memory limits in docker.
Anyone resolved a similar issue?
Project: 06f3919c-f0af-4531-8cf3-97d17baf0d80
DB: 0c85bfb1-2499-4463-87a4-8fd8ae8013f8
3 replies
Ruby app crashing because of missing (bundle ran)
Recently we added some google API gems to our Ruby app and we have seen front-end and back-end services crashing.
+ bundle exec sidekiq
/usr/local/rvm/gems/ruby-3.1.2/gems/bundler-2.3.4/lib/bundler/definition.rb:489:in `materialize': Could not find google-protobuf-3.21.9, grpc-1.50.0 in any of the sources (Bundler::GemNotFound)
While this gem was installed during build-phase. Seems like the native platform for build image and runtime image is different?
Had to hot-fix this by adding another bundle install to the run time.
https://railway.app/project/76e242e9-db08-4279-a677-6c0063fdb9ce/service/e5852326-cf78-4686-b1bf-bd7534119d29
2 replies
libxml2-dev required to deploy a Rails app with Nokogiri
When deploying our Rails app I struggled a lot with the initial deploy with some gems with native deps failing to install (Nokogiri)
Eventually I ended up adding this to my env
NIXPACKS_APT_PKGS=libxml2-dev
Is that expected? I would think Nokogiri being such a staple that libxml would be included in the Nixpacks.
1 replies
Intermittent redis connection issues
Getting about a dozen of these per day:
Error connecting to Redis on containers-us-west-41.railway.app:6553 (Redis::TimeoutError)
/usr/local/rvm/gems/ruby-3.1.2/gems/redis-4.8.0/lib/redis/client.rb:398:in `rescue in establish_connection'
Or:
Redis::TimeoutError
My project:
project/76e242e9-db08-4279-a677-6c0063fdb9ce
Is there something I should update to fix this?
31 replies