Cypress using Lambda Containers

[tomardern]

Hi,

AWS has recently announced the ability to run Docker containers in Lambda. This is very exciting, as opens the possibility of running Cypress on lambda in a less hacky, easier way.

I've been playing with this over the last day or so, and have got pretty close. I've started with one of the pre-built Cypress docker images, added the AWS NodeJs Runtime Client, however, when trying to test it, I get the following logs:

Verifying Cypress can run /cypress-cache/6.0.0/Cypress [started][14:59:50]  Verifying Cypress can run /cypress-cache/6.0.0/Cypress [failed][14:59:50] → Cypress failed to start.This is usually caused by a missing library or dependency.The error below should indicate which dependency is missing.https://on.cypress.io/required-dependenciesIf you are using Docker, we provide containers with all required dependencies installed.----------prctl(PR_SET_NO_NEW_PRIVS) failedprctl(PR_SET_NO_NEW_PRIVS) failed[70:1202/145938.893744:WARNING:discardable_shared_memory_manager.cc(194)] Less than 64MB of free space in temporary directory for shared memory files: 0
cypress:ts Running without ts-node hook in environment "production"
cypress:server:appdata path: /tmp/Cypress/cy/production/browsers
cypress:server:util:node_options NODE_OPTIONS check failed, forking { NODE_OPTIONS: '--max-http-header-size=1048576 ', ORIGINAL_NODE_OPTIONS: '' }prctl(PR_SET_NO_NEW_PRIVS) failedprctl(PR_SET_NO_NEW_PRIVS) failed[84:1202/145946.776462:WARNING:discardable_shared_memory_manager.cc(194)] Less than 64MB of free space in temporary directory for shared memory files: 0
cypress:ts Running without ts-node hook in environment "production"[70:1202/145947.315893:ERROR:bus.cc(393)] Failed to connect to the bus: Failed to connect to socket /var/run/dbus/system_bus_socket: No such file or directory[70:1202/145947.475609:ERROR:platform_shared_memory_region_posix.cc(250)] Creating shared memory in /dev/shm/.org.chromium.Chromium.rsaAO5 failed: No such file or directory (2)[70:1202/145947.475664:ERROR:platform_shared_memory_region_posix.cc(253)] Unable to access(W_OK|X_OK) /dev/shm: No such file or directory (2)[70:1202/145947.475672:FATAL:platform_shared_memory_region_posix.cc(255)] This is frequently caused by incorrect permissions on /dev/shm.  Try 'sudo chmod 1777 /dev/shm' to fix.
cypress:server:appdata path: /tmp/Cypress/cy/production/browsers
cypress:server:util:node_options NODE_OPTIONS check passed, not forking { NODE_OPTIONS: '--max-http-header-size=1048576 ' }
cypress:server:util:node_options restoring NODE_OPTIONS { NODE_OPTIONS: '--max-http-header-size=1048576 ', ORIGINAL_NODE_OPTIONS: '' }
cypress:server:cypress starting cypress with argv [ '/cypress-cache/6.0.0/Cypress/Cypress', '--no-sandbox', '--smoke-test', '--ping=607' ]
cypress:server:args argv array: [ '/cypress-cache/6.0.0/Cypress/Cypress', '--no-sandbox', '--smoke-test', '--ping=607' ]
cypress:server:args parsed argv options { options: { _: [ '/cypress-cache/6.0.0/Cypress/Cypress' ], sandbox: false, 'smoke-test': true, smokeTest: true, ping: 607 } }
cypress:server:args argv parsed: { _: [ '/cypress-cache/6.0.0/Cypress/Cypress' ], sandbox: false, smokeTest: true, ping: 607, invokedFromCli: false, cwd: '/cypress-cache/6.0.0/Cypress/resources/app/packages/server' }
cypress:server:util:proxy found proxy environment variables { npm_config_proxy: '', npm_config_https_proxy: '', npm_config_noproxy: '' }
cypress:server:args options { _: [ '/cypress-cache/6.0.0/Cypress/Cypress' ], sandbox: false, smokeTest: true, ping: 607, invokedFromCli: false, cwd: '/cypress-cache/6.0.0/Cypress/resources/app/packages/server', config: {} }
cypress:server:args argv options: { _: [ '/cypress-cache/6.0.0/Cypress/Cypress' ], sandbox: false, smokeTest: true, ping: 607, invokedFromCli: false, cwd: '/cypress-cache/6.0.0/Cypress/resources/app/packages/server', config: {}, pong: 607 }
cypress:server:cypress from argv [ '/cypress-cache/6.0.0/Cypress/Cypress', '--no-sandbox', '--smoke-test', '--ping=607' ] got options { _: [ '/cypress-cache/6.0.0/Cypress/Cypress' ], sandbox: false, smokeTest: true, ping: 607, invokedFromCli: false, cwd: '/cypress-cache/6.0.0/Cypress/resources/app/packages/server', config: {}, pong: 607 }[84:1202/145949.715445:ERROR:bus.cc(393)] Failed to connect to the bus: Failed to connect to socket /var/run/dbus/system_bus_socket: No such file or directory[84:1202/145949.792308:WARNING:gpu_process_host.cc(1267)] The GPU process has crashed 1 time(s)[84:1202/145949.800269:WARNING:gpu_process_host.cc(1267)] The GPU process has crashed 2 time(s)[84:1202/145949.801111:WARNING:gpu_process_host.cc(1267)] The GPU process has crashed 3 time(s)[84:1202/145949.801754:WARNING:gpu_process_host.cc(1267)] The GPU process has crashed 4 time(s)
cypress:server:appdata path: /tmp/Cypress/cy/production[84:1202/145950.033153:WARNING:gpu_process_host.cc(1267)] The GPU process has crashed 5 time(s)
cypress:server:util:process_profiler current & mean memory and CPU usage by process group:
┌─────────┬───────────┬──────────────┬────────────────────┬────────────┬────────────────┬──────────┬──────────────┬─────────────┐
│ (index) │   group   │ processCount │        pids        │ cpuPercent │ meanCpuPercent │ memRssMb │ meanMemRssMb │ maxMemRssMb │
├─────────┼───────────┼──────────────┼────────────────────┼────────────┼────────────────┼──────────┼──────────────┼─────────────┤
│    0    │ 'cypress' │      1       │        '84'        │    0.03    │      0.03      │   98.5   │     98.5     │    98.5     │
│    1    │  'other'  │      4       │ '88, 89, 239, 240' │     0      │       0        │   3.36   │     3.36     │    3.36     │
│    2    │  'TOTAL'  │      5       │        '-'         │    0.03    │      0.03      │  101.86  │    101.86    │   101.86    │
└─────────┴───────────┴──────────────┴────────────────────┴────────────┴────────────────┴──────────┴──────────────┴─────────────┘
[84:1202/145950.392900:WARNING:gpu_process_host.cc(1267)] The GPU process has crashed 6 time(s)[84:1202/145950.392923:FATAL:gpu_data_manager_impl_private.cc(445)] GPU process isn't usable. Goodbye.----------Platform: linux (Debian - 10.5)
Cypress Version: 6.0.0Cypress failed to start.This is usually caused by a missing library or dependency.
The error below should indicate which dependency is missing.
https://on.cypress.io/required-dependenciesIf you are using Docker, we provide containers with all required dependencies installed.
----------
prctl(PR_SET_NO_NEW_PRIVS) failedprctl(PR_SET_NO_NEW_PRIVS) failed[70:1202/145938.893744:WARNING:discardable_shared_memory_manager.cc(194)] Less than 64MB of free space in temporary directory for shared memory files: 0

I think the main problem is that Electron (or Chrome) is trying to use /dev/shm, however, it doesn't seem to exist when running via Lambda. (as per this comment and this discussion

When running ls -l /dev, I don't see /dev/shm:

crw-rw-rw- 1 root root 1, 7 Dec  2 13:53 full
crw-rw-rw- 1 root root 1, 3 Dec  2 13:53 null
crw-rw-rw- 1 root root 1, 8 Dec  2 13:53 random
lrwxrwxrwx 1 root root   15 Dec  2 13:53 stderr -> /proc/self/fd/2
lrwxrwxrwx 1 root root   15 Dec  2 13:53 stdin -> /proc/self/fd/0
lrwxrwxrwx 1 root root   15 Dec  2 13:53 stdout -> /proc/self/fd/1
crw-rw-rw- 1 root root 1, 9 Dec  2 13:53 urandom
crw-rw-rw- 1 root root 1, 5 Dec  2 13:53 zero

And when calling df -h (to get the mounted drives), I get the following:

/mnt/root-rw/opt/amazon/asc/worker/tasks/rtfs/inline-manifest  1.5G   31M  1.4G   3% /
/dev/vdb                                                       1.5G   31M  1.4G   3% /dev
/dev/vdd                                                   526M   41M  474M   8% /tmp
/dev/root                                                  7.8G  6.7G  1.1G  87% /etc/passwd

Another error is the following, which could be fixed using this method - I'll need to investigate this tomorrow.

/var/run/dbus/system_bus_socket: No such file or directory

I've attempted to run using the following environment variables:

ELECTRON_EXTRA_LAUNCH_ARGS = “--disable-dev-shm-usage”;
DEBUG = "cypress:*"
XDG_CONFIG_HOME = “/tmp”; // https://github.com/electron/electron/blob/master/docs/api/app.md#appgetpathname

Is anyone else attempting/working with/playing with getting cypress to work on Lambda Containers?

[ryparker]

I'm trying to get this to work as well. It works just fine locally on my M1 mac but Lambda keeps throwing errors:

Dockerfile

# Dockerfile
# Install dependencies only when needed
FROM --platform=linux/arm64 cypress/included:10.3.1-typescript AS deps

WORKDIR /app
COPY package.json yarn.lock ./
RUN yarn install --frozen-lockfile

# Rebuild the source code only when needed
FROM --platform=linux/arm64 cypress/included:10.3.1-typescript AS builder
COPY --from=deps /root/.cache/Cypress /root/.cache/Cypress
WORKDIR /app
COPY --from=deps /app/node_modules ./node_modules
COPY . .

ENV CI=1

ENTRYPOINT ["yarn", "test:headless"]

Lambda error:

$ cypress run --headless
--
warning Cannot find a suitable global folder. Tried these: "/usr/local, /home/sbx_user1051/.yarn"
[STARTED] Task without title.
[FAILED] Cypress verification timed out.
[FAILED]
[FAILED] This command failed with the following output:
[FAILED]
[FAILED] /root/.cache/Cypress/10.9.0/Cypress/Cypress --no-sandbox --smoke-test --ping=712
[FAILED]
[FAILED] ----------
[FAILED]
[FAILED] prctl(PR_SET_NO_NEW_PRIVS) failed
[FAILED] prctl(PR_SET_NO_NEW_PRIVS) failed
[FAILED] [32:1003/061928.407802:WARNING:discardable_shared_memory_manager.cc(197)] Less than 64MB of free space in temporary directory for shared memory files: 0
[FAILED] [32:1003/061942.161140:ERROR:bus.cc(398)] Failed to connect to the bus: Failed to connect to socket /run/dbus/system_bus_socket: No such file or directory
[FAILED] [32:1003/061942.161189:ERROR:bus.cc(398)] Failed to connect to the bus: Failed to connect to socket /run/dbus/system_bus_socket: No such file or directory
[FAILED] [32:1003/061942.841092:ERROR:bus.cc(398)] Failed to connect to the bus: Address does not contain a colon
[FAILED] [32:1003/061942.841121:ERROR:bus.cc(398)] Failed to connect to the bus: Address does not contain a colon
[FAILED] [32:1003/061943.364936:WARNING:bluez_dbus_manager.cc(248)] Floss manager not present, cannot set Floss enable/disable.
[FAILED] [32:1003/061943.365038:ERROR:gpu_process_host.cc(968)] GPU process launch failed: error_code=1002
[FAILED] [32:1003/061943.365049:WARNING:gpu_process_host.cc(1274)] The GPU process has crashed 1 time(s)
[FAILED] [32:1003/061943.431097:ERROR:gpu_process_host.cc(968)] GPU process launch failed: error_code=1002
[FAILED] [32:1003/061943.431124:WARNING:gpu_process_host.cc(1274)] The GPU process has crashed 2 time(s)
[FAILED] [32:1003/061943.431918:ERROR:gpu_process_host.cc(968)] GPU process launch failed: error_code=1002
[FAILED] [32:1003/061943.431937:WARNING:gpu_process_host.cc(1274)] The GPU process has crashed 3 time(s)
[FAILED] [32:1003/061943.433490:ERROR:gpu_process_host.cc(968)] GPU process launch failed: error_code=1002
[FAILED] [32:1003/061943.433510:WARNING:gpu_process_host.cc(1274)] The GPU process has crashed 4 time(s)
[FAILED] [32:1003/061943.434286:ERROR:gpu_process_host.cc(968)] GPU process launch failed: error_code=1002
[FAILED] [32:1003/061943.434300:WARNING:gpu_process_host.cc(1274)] The GPU process has crashed 5 time(s)
[FAILED] [32:1003/061943.444696:ERROR:gpu_process_host.cc(968)] GPU process launch failed: error_code=1002
[FAILED] [32:1003/061943.444711:WARNING:gpu_process_host.cc(1274)] The GPU process has crashed 6 time(s)
[FAILED] [32:1003/061943.444757:FATAL:gpu_data_manager_impl_private.cc(450)] GPU process isn't usable. Goodbye.
[FAILED]
[FAILED] ----------

[ryparker]

Looks like this is being tracked in this issue: #1232