atendai/evolution-api Docker image: Create container fails instantly, zero logs (all tags)
gustavozetum
HOBBYOP

a month ago

Project: AgendaConsulta (9ac61145-e1b4-4765-b38f-e23783f61494), environment production, service agenda-evolution-api (a9dd57fd-48a1-4d78-b5ae-ab82f3ecc96a), region sfo, Hobby plan.

Summary: every deployment attempt of the public Docker Hub image atendai/evolution-api fails at "Deploy > Create container" within 0-5 seconds, with zero logs produced (no build logs, no deploy logs, nothing via railway logs CLI either). Error shown is only the generic "Failed to create deployment." Railway's own Diagnose tool either fails ("Diagnosis failed for this deployment") or returns nothing useful.

What I tried, all in the same service/region:

  1. atendai/evolution-api:v2.2.3 (Docker image source), on Trial plan -> failed instantly, zero logs
  2. Same, after upgrading account to Hobby plan specifically to try to fix this -> failed identically
  3. atendai/evolution-api:latest -> failed identically
  4. atendai/evolution-api@sha256:28ed9cb0e60fc14e4ddf9644dd78a8e71383aea6f20b436c12c3cc5fdcd78efd (pinned amd64 digest, bypassing the multi-arch manifest index) -> failed identically
  5. nginx:alpine in the same service -> deployed successfully, went Online
  6. node:20 (~1GB, much larger than evolution-api's ~280MB) in the same service -> deployed/completed successfully
  7. A minimal Dockerfile (FROM atendai/evolution-api:v2.2.3) deployed via railway up (build path instead of image-source path) -> build also failed, no build logs at all This rules out billing/trial limits, image architecture, image size, deploy method (image vs Dockerfile build), and account/region capacity, since unrelated images deploy fine in the exact same service/region right before and after. Reference deployment IDs (all failed identically): a0612eb5-028e-4ff3-8cd1-af3553fe5241, 5b7b6c51-2eee-49ec-9972-5ae9be7a797e, cb56b20c-2e8d-4ec0-ab2a-8ee5aec6b490, c7027d80-20f1-4c99-a990-80a792757bbd, b9caa74a-dfb8-4e7f-9992-56ad10289877. I noticed two other threads posted in the last few hours also reporting deploys stuck/failing at CREATE_CONTAINER, so this might be a broader platform issue rather than something specific to this image. What I need: an explanation of what's failing in container creation for this specific image, or a workaround to get atendai/evolution-api:v2.2.3 running as a service in this project.
$10 Bounty

1 Replies

Railway
BOT

a month ago

This thread has been opened as a bounty so the community can help solve it.

Status changed to Open Railway 28 days ago


a month ago

The image is moved to evoapicloud/evolution-api, change your image to evoapicloud/evolution-api:v2.2.3


Welcome!

Sign in to your Railway account to join the conversation.

Loading...