Hey guys, so I upgraded Calcom recently as well as moved to the new Dockerfile. I need to build the Calcom image, have some changes I need, the build fails at the yarn install step with exit code 129. I'm not able to figure out what is causing this error. Vercel build for the upgraded calcom is also failing at the yarn install step. I posted an issue in the github repo about this as well - https://github.com/calcom/docker/issues/233. There isn't any information on what is causing the yarn install to fail in either case, so it would be great if anyone could point me in the right direction.

yarn install

works completely fine in the calcom repo when run locally.. docker container also builds fine locally. Only failing on an EC2 and Vercel

wondering if there are there any resource constraints on the cluster that could be affecting this. do you have ram/cpu/disk info?

Sure, it's an Amazon Linux instance

t2.large

it has a 16GB EBS volume attached, 8 GB ram. It built the previous version of the calcom docker image successfully.

Sorry for the delays, travelling today after wrapping up a conference

I'm not specifying any start up command for calcom, using the Dockerfile in the calcom docker repo as is.

COPY calcom/package.json calcom/yarn.lock calcom/.yarnrc.yml calcom/playwright.config.ts calcom/turbo.json calcom/git-init.sh calcom/git-setup.sh ./
COPY calcom/.yarn ./.yarn
COPY calcom/apps/web ./apps/web
COPY calcom/packages ./packages

RUN yarn config set httpTimeout 1200000 && \ 
    npx turbo prune --scope=@calcom/web --docker && \
    yarn install && \
    yarn db-deploy && \
    yarn --cwd packages/prisma seed-app-store    <----- This is where the build fails

RUN yarn turbo run build --filter=@calcom/web

I'm running

docker-compose up --build

with docker-compose.yaml

services:
  calcom:
    image: calendso/calendso:latest
    build:
      context: .
      dockerfile: Dockerfile
      args:
        NEXT_PUBLIC_WEBAPP_URL: ${NEXT_PUBLIC_WEBAPP_URL}
        NEXT_PUBLIC_LICENSE_CONSENT: ${NEXT_PUBLIC_LICENSE_CONSENT}
        CALCOM_TELEMETRY_DISABLED: 1
        NEXTAUTH_SECRET: ${NEXTAUTH_SECRET}
        CALENDSO_ENCRYPTION_KEY: ${CALENDSO_ENCRYPTION_KEY}
        DATABASE_URL: ${DATABASE_URL}
    restart: always
    networks:
      - ba-calcom
    ports:
      - 3000:3000
    env_file: .env
    environment:
      NEXT_PUBLIC_WEBAPP_URL: ${NEXT_PUBLIC_WEBAPP_URL}
      DATABASE_URL: ${DATABASE_URL}
      VIRTUAL_HOST: my_domain   // This is for my https-portal service
      VIRTUAL_PORT: 3000

What does your disk usage look like?

Apologies for the delayed response, I'm still at work so popping back in as soon as I can. My disk usage at the time of the crash. Please let me know if you're looking for some other information. Nothing else is running in the EC2. I ran

docker system prune

before this.

Filesystem     1K-blocks    Used Available Use% Mounted on
devtmpfs         4059516       0   4059516   0% /dev
tmpfs            4068308       0   4068308   0% /dev/shm
tmpfs            4068308     444   4067864   1% /run
tmpfs            4068308       0   4068308   0% /sys/fs/cgroup
/dev/xvda1      16764908 7707752   9057156  46% /
tmpfs              10240       0     10240   0% /tmp
tmpfs             813664       0    813664   0% /run/user/1000

Can you modify the RUN command near the crash to break the &&'d commands into individual RUN commands?

Thank you! I did this and the error turned out to not be

yarn install

.

ERROR [builder 11/13] RUN yarn --cwd packages/prisma seed-app-store

I'll try running this locally to see if I can replicate and maybe see the error output

We're not getting it from the nightly or release builds, so I wonder if it's related to configuration or a certain app config that could have broken

I fixed this by adding a

.env.appStore

and moving the app store related creds there. This step now runs successfully, but I now have an error at

yarn turbo run build --filter=@calcom/web

#0 420.5 @calcom/web:build: > Build error occurred
#0 420.5 @calcom/web:build: Error: Collecting page data for [object Object] is still timing out after 2 attempts. See more info here <https://nextjs.org/docs/messages/page-data-collection-timeout>
#0 420.5 @calcom/web:build:     at onRestart (/calcom/node_modules/next/dist/build/index.js:601:35)
#0 420.5 @calcom/web:build:     at Worker.isPageStatic (/calcom/node_modules/next/dist/lib/worker.js:48:40)
#0 420.5 @calcom/web:build:     at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
#0 420.5 @calcom/web:build:     at async Span.traceAsyncFn (/calcom/node_modules/next/dist/trace/trace.js:79:20)
#0 420.5 @calcom/web:build:     at async /calcom/node_modules/next/dist/build/index.js:725:52
#0 420.5 @calcom/web:build:     at async Span.traceAsyncFn (/calcom/node_modules/next/dist/trace/trace.js:79:20)
#0 420.5 @calcom/web:build:     at async Promise.all (index 49)
#0 420.5 @calcom/web:build:     at async /calcom/node_modules/next/dist/build/index.js:664:17
#0 420.5 @calcom/web:build:     at async Span.traceAsyncFn (/calcom/node_modules/next/dist/trace/trace.js:79:20)
#0 420.5 @calcom/web:build:     at async /calcom/node_modules/next/dist/build/index.js:625:170

I’m getting this error too. Tried setting staticPageGenerationTimeout to 1000 but still get the same error

@Ayush - Did you manage to resolve this? I tested the next build step locally and it seems to work fine, except for TRPC errors to the wrong domain

<< query #1 viewer.features.list {
  input: undefined,
  result: TRPCClientError: request to <http://https//mynextpublic_url/api/trpc/viewer.features.list?batch=1&input=%7B%220%22%3A%7B%22json%22%3Anull%2C%22meta%22%3A%7B%22values%22%3A%5B%22undefined%22%5D%7D%7D%7D> failed, reason: getaddrinfo ENOTFOUND https

Yeah, the weird thing is the timing out after 2 attempts error doesn’t happen on my laptop (m1 Mac). But I get it on my desktop and server (both are Debian).

Yes, same here, the build goes through successfully on my Mac as well, failing in the Amazon Linux EC2 and the Vercel deployment too, all with very vague errors..