zenika/alpine-chrome

Chrome Headless docker images built upon alpine official image

ZeniLunch

Trop dur de choisir où manger le midi, enfin ça c'était avant cette application Android.

Lien vers la maquette Figma : https://www.figma.com/file/zTOajuYZ2MYCohjXHeSJuA/ZeniLunch?node-id=0%3A1&t=MHQir7ZKVIQ3yg5d-1

alpine-chrome's Issues

Push automatically to gcr.io

Very useful to use with the Google Cloud Platform.
Reuse the hook from here

Test the '--disable-dev-shm-usage' flag

Discussion from https://github.com/GoogleChrome/puppeteer/blob/master/docs/troubleshooting.md#tips

Follow the official docs to build the docker image

Docs from puppeeter

Add `web-gl` by default?

Context

The base alpine-chrome is 135 MB. The web-gl one is 138 MB.
Does it make sense to add it by default? 🤔

Discussion

Please help to decide 👍

Get a deno example using Chrome

Context

It could be fun to get a deno example using Chrome using the with-deno tag.
See the with-deno folder

Related to #89

Create a Documentation website

Context

What are you trying to do and how would you want to do it differently? Is it something you currently you cannot do? Is this related to an issue/problem?
Would like to clean the README and create a doc website for using this image.
There is too many sections in the README explaining the different usecases...

Perhaps a doc website using VuePress...

Publish image for arm

Hi, it would be awesome if you could publish images for ARM for running on Raspberry devices.

Use image for PDF generation

Context

One of our projects has created some debian image with node+chrome installed to be able to create PDFs.

But our standard for now of base images is alpine. Is you alpine-chrome image capable to do? I did a lot of reading about chrome and alpine, and came quickly to chromium installation. But the guy told me that chromium can't be used for the PDF generation, he needs the real chrome.

Regards
Walter

Permissions on mounted files

From an email

How to read the image generated after the screenshot?
The image is created into the mapped directory (to the volume) but the owner of the image is 1001 (root outside the container?):
-rw------- 1 1001 1001 2454 mar 8 16:48 screenshot.png

So I can't open the image until I've changed the permissions.

Any idea how to solve it?

Test a new `no-sandbox` way with Docker

From the lighthouse image:

https://github.com/ebidel/lighthouse-ci/blob/master/builder/Dockerfile#L35-L40

grab with ffmpeg

There is any option to grab screen to ffmpeg ?
I want to use this headless chrome to broadcast to youtube.
I now use https://github.com/kolyadin/live-video-broadcasting but this version crash all time

Getting symbol not found error

Building the image on archlinux and running

chromium-browser --version

Will result in:

Error relocating /usr/lib/chromium/chrome: hb_font_funcs_set_glyph_h_advances_func: symbol not found

Add new issue templates

Like this: https://github.com/nock/nock/tree/beta/.github/ISSUE_TEMPLATE

Failed to move to new namespace: PID namespaces supported, Network namespace supported, but failed: errno = Operation not permitted

Please tell me what I'm doing wrong?

docker run -it  zenika/alpine-chrome

Failed to move to new namespace: PID namespaces supported, Network namespace supported, but failed: errno = Operation not permitted
[0908/094739.839947:FATAL:zygote_host_impl_linux.cc(187)] Check failed: ReceiveFixedMessage(fds[0], kZygoteBootMessage, sizeof(kZygoteBootMessage), &boot_pid). 
Received signal 6
  r8: 00000000000000c1  r9: 00007fa92f74114c r10: 0000000000000008 r11: 0000000000000246
 r12: 00007ffe84cf5630 r13: 0000000000000000 r14: 00007ffe84cf53e0 r15: 00000000000000a0
  di: 0000000000000002  si: 00007ffe84cf4d60  bp: 00007ffe84cf4d60  bx: 0000000000000006
  dx: 0000000000000000  ax: 0000000000000000  cx: 00007fa92f6f75b7  sp: 00007ffe84cf4d48
  ip: 00007fa92f6f75b7 efl: 0000000000000246 cgf: 002b000000000033 erf: 0000000000000000
 trp: 0000000000000000 msk: 0000000000000000 cr2: 0000000000000000
[end of stack trace]
Calling _exit(1). Core file will not be generated.

When adding --no-sandbox - other errors appear

docker run -it  zenika/alpine-chrome --no-sandbox

[0908/101024.810903:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0908/101025.065766:ERROR:gpu_channel_manager.cc(398)] ContextResult::kFatalFailure: Failed to create shared context for virtualization.
[0908/101025.085427:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.

help me please!

Text in screenshot is fuzzy

I am attempting to use Chrome to take some screenshots.

Here is a screenshot taken on Alpine docker container run -it --rm -v $(pwd):/usr/src/app --security-opt seccomp=$(pwd)/chrome.json zenika/alpine-chrome --screenshot --no-sandbox --hide-scrollbars --window-size=1200,630 https://github.com/Zenika/alpine-chrome

and a screenshot taken on macOS chromium --headless --use-gl=swiftshader --disable-software-rasterizer --disable-dev-shm-usage --screenshot --no-sandbox --hide-scrollbars --window-size=1200,630 https://github.com/Zenika/alpine-chrome

You can see in the Alpine screenshot, that all of the text appears super pixelated.

Any ideas?

Note: The Alpine screenshot looks much fuzzier in macOS finder than it does in the GitHub upload.

Add a `squash` version of alpine-chrome:latest

Context

💡 This is only an idea.
To get a smaller image, we could use the experimental features docker image build --squash.
But it's not working on the Docker Hub. See docker/hub-feedback#955

Alternatives

I can do it on my laptop and upload it.
Here is the difference: only ~5MB

docker image inspect zenika/alpine-chrome --format='{{.Size}}'
393543827
docker image inspect zenika/alpine-chrome:latest-squash --format='{{.Size}}'
388574396

Conclusion

Not very useful at the moment. 😅

Add a lighthouse version and example

Check https://github.com/MatthiasWinkelmann/lighthouse-chromium-alpine-docker

Cannot install stable alpine packages in :latest

Describe the bug
We are currently unable to build our cucumber test images since 3227d27 overwrites the alpine 3.10 repositories from the build, leaving only edge behind. While this may be necessary (not sure if it is) for the installation of chromium, I think it would be good practice to restore the original configuration afterwards.

There's also some feedback on the commit, requesting to open a ticket. I couldn't find one, so here it is.

To Reproduce
Steps to reproduce the behavior:

1. zenika/alpine-chrome@sha256:e436385b09839d8f6cd69f3a13a5decd74ea7e2e64a832129c27dc2ac9e64e71
2. Build custom image based on FROM zenika/alpine-chrome:latest
3. Attempt to install packages, e.g. RUN apk add bash

Step 3/10 : RUN apk add bash
 ---> Running in 5b44d631b561
fetch http://nl.alpinelinux.org/alpine/edge/community/x86_64/APKINDEX.tar.gz
fetch http://nl.alpinelinux.org/alpine/edge/main/x86_64/APKINDEX.tar.gz
ERROR: unsatisfiable constraints:
  readline-8.0.1-r0:
    masked in: @edge
    satisfies: bash-5.0.11-r0[so:libreadline.so.8]
  bash-5.0.11-r0:
    masked in: @edge
    satisfies: world[bash]
The command '/bin/sh -c apk add bash' returned a non-zero code: 2

Fails due to the unavailability of apk "normal" repositories; Only bash@edge would be supported.

What is the expected behavior?
We'd like to be able to install packages from stable alpine again, without being forced to use (potentially unstable) edge versions.

What is the actual behavior?
See error above

Possible solution
Ideas:

• refrain from using edge altogether
• branch out a separate line of edge-images for those who need it
• restore stable alpine repositories after edge is no longer necessary

Versions

Software	Version(s)
Chromium	Chromium 77.0.3865.75
Image	zenika/alpine-chrome@sha256:e436385b09839d8f6cd69f3a13a5decd74ea7e2e64a832129c27dc2ac9e64e71
Docker Engine	18.018.09

Additional Context
Since this also affects :77, this leaves us completely unable to use the image altogether. Alpine has removed older versions of the chromedriver from their 3.10 repositories, leaving version 77 as only option. That, in turns, only supports chromium 77 and above, which means the :76 tag will not work for us, either.

Add and explain the `no-sandbox` value.

inspiration from https://hub.docker.com/r/justinribeiro/chrome-headless/

Update badges about size of the image 🙌

Context

See https://badgen.net/docker to use working badges
https://badgen.net/docker/size/zenika/alpine-chrome/latest

Alternatives
https://shields.io/
http://img.shields.io/docker/image-size/zenika/alpine-chrome/81

add with selenium/playwright

Hi guys,

Hope you are doing well !

It would be awesome to have alpine chrome with selenium.

Is it something possible to add ?

Cheers,
X

How to enable webgl support?

Currently, there is no webgl support.( Softversion)

🤔 Remove tini

Describe the bug

From the tini lib

If you are using Docker 1.13 or greater, Tini is included in Docker itself. This includes all versions of Docker CE. To enable Tini, just pass the --init flag to docker run.

« Lost UI shared context » error

On OSX, I have this errors:

$ docker container run -it --rm --cap-add=SYS_ADMIN zenika/alpine-chrome                                                 stephane at MacBook-Pro-de-KLEIN.local (-)(master)
[0617/130836.688570:WARNING:dns_config_service_posix.cc(341)] Failed to read DnsConfig.
[0617/130836.693739:ERROR:gpu_process_transport_factory.cc(1009)] Lost UI shared context.
[0617/130836.698564:ERROR:instance.cc(49)] Unable to locate service manifest for metrics
[0617/130836.698614:ERROR:service_manager.cc(890)] Failed to resolve service name: metrics
[0617/130836.789673:ERROR:gl_implementation.cc(292)] Failed to load /usr/lib/chromium/swiftshader/libGLESv2.so: Error loading shared library /usr/lib/chromium/swiftshader/libGLESv2.so: No such file or directory
[0617/130836.794153:ERROR:instance.cc(49)] Unable to locate service manifest for metrics
[0617/130836.794245:ERROR:service_manager.cc(890)] Failed to resolve service name: metrics
[0617/130836.795285:ERROR:viz_main_impl.cc(196)] Exiting GPU process due to errors during initialization
[0617/130836.837893:ERROR:gl_implementation.cc(292)] Failed to load /usr/lib/chromium/swiftshader/libGLESv2.so: Error loading shared library /usr/lib/chromium/swiftshader/libGLESv2.so: No such file or directory
[0617/130836.840525:ERROR:viz_main_impl.cc(196)] Exiting GPU process due to errors during initialization
[0617/130836.840815:ERROR:instance.cc(49)] Unable to locate service manifest for metrics
[0617/130836.840888:ERROR:service_manager.cc(890)] Failed to resolve service name: metrics
[0617/130836.908523:ERROR:gl_implementation.cc(292)] Failed to load /usr/lib/chromium/swiftshader/libGLESv2.so: Error loading shared library /usr/lib/chromium/swiftshader/libGLESv2.so: No such file or directory
[0617/130836.914651:ERROR:instance.cc(49)] Unable to locate service manifest for metrics
[0617/130836.914733:ERROR:service_manager.cc(890)] Failed to resolve service name: metrics
[0617/130836.917125:ERROR:viz_main_impl.cc(196)] Exiting GPU process due to errors during initialization
[0617/130836.966942:ERROR:gl_implementation.cc(292)] Failed to load /usr/lib/chromium/swiftshader/libGLESv2.so: Error loading shared library /usr/lib/chromium/swiftshader/libGLESv2.so: No such file or directory
[0617/130836.968933:ERROR:viz_main_impl.cc(196)] Exiting GPU process due to errors during initialization
[0617/130836.969294:ERROR:instance.cc(49)] Unable to locate service manifest for metrics
[0617/130836.969336:ERROR:service_manager.cc(890)] Failed to resolve service name: metrics
[0617/130836.970184:ERROR:service_manager_context.cc(219)] Attempting to run unsupported native service: /usr/lib/chromium/content_gpu.service

Existing docker images based on alpine-chrome:with-puppeteer are now broken

Since this commit my build pipeline is broken.

My docker image did this in the final step:
CMD ["node", "run"]

And it seems that having tini set as the entrypoint breaks this behaviour.

Of course I can fix this in my Dockerfile but this won't fix old revisions of images that are based off of alpine-chrome:with-puppeteer. Was it a conscious decision to break existing behaviour?

Have a look on "dumb-init" to kill the process

Like on this image: https://github.com/ebidel/try-puppeteer

Add a version to "with-node" and "with-puppeteer"

Context

It could be useful to use a semver and not using the latest tag.
Example: 76, 76-with-node, 76-with-puppeteer

Alternatives

None.

Has the feature been requested before?

Nope.

`apk update` fails

I'm extending the Chrome v72 image. This is my Dockerfile:

$ cat Dockerfile
FROM zenika/alpine-chrome:72
USER root
RUN apk update

When I build it, the apk command fails.

$ docker build --no-cache --rm -f Dockerfile .
Sending build context to Docker daemon  3.072kB
Step 1/3 : FROM zenika/alpine-chrome:72
 ---> 361552ddc6fb
Step 2/3 : USER root
 ---> Running in b7440d6d4cdb
Removing intermediate container b7440d6d4cdb
 ---> d906a9c502cf
Step 3/3 : RUN apk update
 ---> Running in 77125b8983dd
fetch http://dl-cdn.alpinelinux.org/alpine/v3.9/main/x86_64/APKINDEX.tar.gz
ERROR: http://dl-cdn.alpinelinux.org/alpine/v3.9/main: Bad file descriptor
WARNING: Ignoring APKINDEX.b89edf6e.tar.gz: Bad file descriptor
fetch http://dl-cdn.alpinelinux.org/alpine/v3.9/community/x86_64/APKINDEX.tar.gz
ERROR: http://dl-cdn.alpinelinux.org/alpine/v3.9/community: Bad file descriptor
WARNING: Ignoring APKINDEX.737f7e01.tar.gz: Bad file descriptor
fetch http://nl.alpinelinux.org/alpine/edge/community/x86_64/APKINDEX.tar.gz
fetch http://nl.alpinelinux.org/alpine/edge/main/x86_64/APKINDEX.tar.gz
ERROR: http://nl.alpinelinux.org/alpine/edge/community: Bad file descriptor
WARNING: Ignoring APKINDEX.49e1404d.tar.gz: Bad file descriptor
ERROR: http://nl.alpinelinux.org/alpine/edge/main: Bad file descriptor
WARNING: Ignoring APKINDEX.1ebd6aac.tar.gz: Bad file descriptor
4 errors; 115 distinct packages available
The command '/bin/sh -c apk update' returned a non-zero code: 4

Chrome fails at startup with Jenkins agent

Describe the bug
I have a Jenkins with Kubernetes, and I would like to run a headless Chrome for unittesting (using Karma). I extended a Jenkins Agent image with alpine-chrome:with-node and use that in my agent pods. However karma is not able to start Chrome:

[0920/093525.387820:ERROR:zygote_host_impl_linux.cc(259)] Failed to adjust OOM score of renderer with pid 202: Permission denied (13)
[0920/093525.474421:ERROR:gpu_channel_manager.cc(398)] ContextResult::kFatalFailure: Failed to create shared context for virtualization.

To Reproduce
Use this image:

FROM jenkins/jnlp-slave:alpine as jnlp

FROM alpine-chrome:with-node

USER root
RUN apk -U add openjdk8-jre

COPY --from=jnlp /usr/local/bin/jenkins-slave /usr/local/bin/jenkins-agent
COPY --from=jnlp /usr/share/jenkins/slave.jar /usr/share/jenkins/slave.jar

ENTRYPOINT ["/usr/local/bin/jenkins-agent"]

What is the expected behavior?
Karma able to start Chrome.

Logs

20 09 2019 09:35:24.438:INFO [launcher]: Starting browser Chrome

20 09 2019 09:35:32.813:ERROR [launcher]: Cannot start Chrome
[0920/093525.113146:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093525.387820:ERROR:zygote_host_impl_linux.cc(259)] Failed to adjust OOM score of renderer with pid 202: Permission denied (13)
[0920/093525.474421:ERROR:gpu_channel_manager.cc(398)] ContextResult::kFatalFailure: Failed to create shared context for virtualization.
[0920/093525.505667:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093532.787035:ERROR:browser_process_sub_thread.cc(203)] Waited 1 ms for network service

20 09 2019 09:35:32.813:ERROR [launcher]: Chrome stdout: 
20 09 2019 09:35:32.813:ERROR [launcher]: Chrome stderr: [0920/093525.113146:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093525.387820:ERROR:zygote_host_impl_linux.cc(259)] Failed to adjust OOM score of renderer with pid 202: Permission denied (13)
[0920/093525.474421:ERROR:gpu_channel_manager.cc(398)] ContextResult::kFatalFailure: Failed to create shared context for virtualization.
[0920/093525.505667:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093532.787035:ERROR:browser_process_sub_thread.cc(203)] Waited 1 ms for network service

20 09 2019 09:35:32.818:INFO [launcher]: Trying to start Chrome again (1/2).

20 09 2019 09:35:33.093:ERROR [launcher]: Cannot start Chrome
[0920/093532.949748:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093532.961242:ERROR:zygote_host_impl_linux.cc(259)] Failed to adjust OOM score of renderer with pid 248: Permission denied (13)
[0920/093533.004353:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093533.005844:ERROR:gpu_channel_manager.cc(398)] ContextResult::kFatalFailure: Failed to create shared context for virtualization.
[0920/093533.076144:ERROR:browser_process_sub_thread.cc(203)] Waited 3 ms for network service

20 09 2019 09:35:33.093:ERROR [launcher]: Chrome stdout: 
20 09 2019 09:35:33.093:ERROR [launcher]: Chrome stderr: [0920/093532.949748:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093532.961242:ERROR:zygote_host_impl_linux.cc(259)] Failed to adjust OOM score of renderer with pid 248: Permission denied (13)
[0920/093533.004353:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093533.005844:ERROR:gpu_channel_manager.cc(398)] ContextResult::kFatalFailure: Failed to create shared context for virtualization.
[0920/093533.076144:ERROR:browser_process_sub_thread.cc(203)] Waited 3 ms for network service

20 09 2019 09:35:33.096:INFO [launcher]: Trying to start Chrome again (2/2).

20 09 2019 09:35:33.373:ERROR [launcher]: Cannot start Chrome
[0920/093533.231075:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093533.244953:ERROR:zygote_host_impl_linux.cc(259)] Failed to adjust OOM score of renderer with pid 290: Permission denied (13)
[0920/093533.289019:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093533.289656:ERROR:gpu_channel_manager.cc(398)] ContextResult::kFatalFailure: Failed to create shared context for virtualization.
[0920/093533.359067:ERROR:browser_process_sub_thread.cc(203)] Waited 3 ms for network service

20 09 2019 09:35:33.373:ERROR [launcher]: Chrome stdout: 
20 09 2019 09:35:33.373:ERROR [launcher]: Chrome stderr: [0920/093533.231075:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093533.244953:ERROR:zygote_host_impl_linux.cc(259)] Failed to adjust OOM score of renderer with pid 290: Permission denied (13)
[0920/093533.289019:WARNING:dns_config_service_posix.cc(340)] Failed to read DnsConfig.
[0920/093533.289656:ERROR:gpu_channel_manager.cc(398)] ContextResult::kFatalFailure: Failed to create shared context for virtualization.
[0920/093533.359067:ERROR:browser_process_sub_thread.cc(203)] Waited 3 ms for network service

20 09 2019 09:35:33.376:ERROR [launcher]: Chrome failed 2 times (cannot start). Giving up.

Versions

Software	Version(s)	How-to
Chromium	76.0.3809.132	docker run -it --rm --entrypoint "" zenika/alpine-chrome chromium-browser --version
Image	a449df36	docker image inspect --format='{{index .RepoDigests 0}}' zenika/alpine-chrome
Docker Engine	19.03.2	docker version

Additional context
I start Chrome with the following flags:
--headless, --no-sandbox, --disable-gpu, --disable-software-rasterizer, --disable-dev-shm-usage, --disable-web-security

Create an example with web-vitals

Context

see https://web.dev/vitals/

Add the number of stars on DockerHub

Context

Use https://img.shields.io/docker/stars/zenika/alpine-chrome.svg?style=flat-square to get it.

npm not found

Hi,

npm has not been found since the last update.

BR
Sebastian

Error on container launch: Failed to move to new namespace

I've tried to use it in my docker image based on zenika/alpine-chrome.
see dockerfile here: https://github.com/SuddenGunter/pandaren/tree/master/cmd

It didn't work for me, so I just try tried to do docker run zenika/alpine-chrome - same error.

Error message (happens in both cases)

Failed to move to new namespace: PID namespaces supported, Network namespace supported, but failed: errno = Operation not permitted
[0625/214704.494573:FATAL:zygote_host_impl_linux.cc(187)] Check failed: ReceiveFixedMessage(fds[0], kZygoteBootMessage, sizeof(kZygoteBootMessage), &boot_pid).
Received signal 6
  r8: 00000000000000c1  r9: 00007fb2bce9314c r10: 0000000000000008 r11: 0000000000000246
 r12: 00007ffec57e1860 r13: 00000000000000a0 r14: 000055c1b5d4ba64 r15: 00007ffec57e1880
  di: 0000000000000002  si: 00007ffec57e1730  bp: 00007ffec57e1730  bx: 0000000000000006
  dx: 0000000000000000  ax: 0000000000000000  cx: 00007fb2bce49fa0  sp: 00007ffec57e1718
  ip: 00007fb2bce49fa0 efl: 0000000000000246 cgf: 002b000000000033 erf: 0000000000000000
 trp: 0000000000000000 msk: 0000000000000000 cr2: 0000000000000000
[end of stack trace]
Calling _exit(1). Core file will not be generated.

Environment:
macOS 10.14.5
docker v2.0.0.3 (31259), engine: 18.09.2

chromium-browser segfaults :(

I can't get either of the latest zenika/alpine-chrome or zenika/alpine-chrome:with-node images to successfully launch a browser. If I expand one of the readme examples, this is what happens:

> docker run -it --rm --entrypoint sh zenika/alpine-chrome
> /usr/src/app $ chromium-browser --headless --disable-gpu --disable-software-rasterizer --disable-dev-shm-usage --no-sandbox --dump-dom https://www.chromestatus.com/
Segmentation fault

The only way I've got that example to work is by using the image tagged zenika/alpine-chrome:13.

Do you know what's going on?

propose: add tini to reape (possible) zombie process

When building a pptraas.com-like service, I had to write this in Dockerfile (and they add to layers 😞):

USER root
RUN apk add tini
USER chrome
ENTRYPOINT ['tini', '--']

I am guessing the need to reape zombie process is common, can we add tini (or dumb-init or alike) to base image?

Add a new way to get the swiftshader libs

Context

Thanks to @pxlrbt we have the so libs compiled and available for swiftshader (see #50)

Alternatives

Can you achieve the same result doing that?

It could be interesting to get or generate the libs without rebuilding Chrome (1,5 day), getting the generated APK, unpacking it and retrieving the so files.

RUN apk add --no-cache mesa-egl mesa-gles \
	&& rm -rf /var/lib/apt/lists/* \
    /var/cache/apk/* \
    /usr/share/man \
    /tmp/* \
    && mkdir /usr/lib/chromium/swiftshader/ \
    && cp /usr/lib/libGLESv2.so.2 /usr/lib/chromium/swiftshader/libGLESv2.so \
    && cp /usr/lib/libEGL.so.1 /usr/lib/chromium/swiftshader/libEGL.so
#    && ln -s /usr/lib/libGLESv2.so.2 /usr/lib/chromium/swiftshader/libGLESv2.so \
#    && ln -s /usr/lib/libEGL.so.1 /usr/lib/chromium/swiftshader/libEGL.so
# COPY swiftshader/libEGL.so /usr/lib/chromium/swiftshader/libEGL.so
# COPY swiftshader/libGLESv2.so /usr/lib/chromium/swiftshader/libGLESv2.so
RUN chmod 755 /usr/lib/chromium/swiftshader/libEGL.so
RUN chmod 755 /usr/lib/chromium/swiftshader/libGLESv2.so

It is not working... 😢

** Other alternative **

See @lygstate info
#50 (comment)

What do you think?

Bug image with-puppeter

Hi, you have a bug in the docker image with-puppeter, if you don't add a path "/usr/src/app" in the command "npm install", you will recived a error message in the execution.

module.js:549
throw err;
^

Error: Cannot find module 'puppeteer'
at Function.Module._resolveFilename (module.js:547:15)
at Function.Module._load (module.js:474:25)
at Module.require (module.js:596:17)
at require (internal/module.js:11:18)
at Object. (/usr/src/app/pdf.js:1:81)
at Module._compile (module.js:652:30)
at Object.Module._extensions..js (module.js:663:10)
at Module.load (module.js:565:32)
at tryModuleLoad (module.js:505:12)
at Function.Module._load (module.js:497:3)

Then, if you modify the command for "RUN npm install /usr/src/app" you will resolved the problem. Best regards.

Unable to run it in Jenkins

Error response from daemon: Container is not running

Chromium package in v3.10 has upgraded to 77.x.x.x

We are using puppeteer 1.12.0 in our project. For this version, we need chromium 73.x.x.x which was present in v3.10 branch. Recently version has been upgraded to 77.x.x.x, which broke our application. can you pls mention the branch which has chromium 73.x.x.x package?

unable to run with-puppeteer example

thanks for the work; the image looks like exactly what I need.
However I'm having trouble executing the with-puppeteer example:

docker run -it --rm -v $(pwd)/src:/usr/src/app/src --cap-add=SYS_ADMIN zenika/alpine-chrome:with-puppeteer node src/pdf.js
docker: Error response from daemon: OCI runtime create failed: container_linux.go:344: starting container process caused "exec: "tini": executable file not found in $PATH": unknown.

I'm running in on docker for OSX, in case that matters.

I also tried building a modified image (basically commenting out the ENTRYPOINT with tini) and then I get this error instead:

[0408/191634.027504:ERROR:headless_shell.cc(559)] Open multiple tabs is only supported when remote debugging is enabled.

Any clues on how can I troubleshoot or fix this?

Docker-compose

Hi There. I am trying to set it up to use with docker-compose. However when I try to activate it the container keeps restarting and the terminal gives me:

[0511/154746.370034:WARNING:dns_config_service_posix.cc(335)] Failed to read DnsConfig.
docker-compose is as follows:

version: '3'
services:
testservice
build: ./installer
container_name: testservice
restart: always
ports:
- 9222:9222

Remove the `with-webgl` tag

Describe the bug
Since the base image got the support of webgl, the with-webgl tag doesn't mean anything now.

• Update README.md
• Remove the webgl folder
• Update the build on the Docker Hub

\"tini\": executable file not found in $PATH

After adding ENTRYPOINT ["tini", "--"] it breaks the Puppeteer Docker image.

Reference commit:
7b4f2b3#diff-c29f529b3e98482e07a5768ff85c139dR9

Reproduce by:

docker run -it --rm -v $(pwd)/src:/usr/src/app/src --cap-add=SYS_ADMIN zenika/alpine-chrome:with-puppeteer node src/pdf.js

docker: Error response from daemon: OCI runtime create failed: container_linux.go:344: starting container process caused "exec: \"tini\": executable file not found in $PATH": unknown.
ERRO[0001] error waiting for container: context canceled

chrome 77 on stable alpine

Hi
I'm having stability issues with chromium77 on alpine edge, (selenium freezes), I like to use alpine stable (v3.10). I also like to use chromium76, the only chromium I found in stable was chrome73.
I currently use ubuntu 18.04 but it is 1.5GB large, while the alpine edge is only 384MB.

Cannot manage to reproduce to use `with-playwright`

Hello everybody !

I'm facing the following issue:

Describe the bug
I'm not able to reproduce the getting started instructions for Playwright

To Reproduce
Steps to reproduce the behavior:

1. Version of the image :

$ docker image inspect --format='{{index .RepoDigests 0}}' zenika/alpine-chrome:with-playwright
zenika/alpine-chrome@sha256:c3a832060aa529bd0ca87bc60ff669015bcc5c8d5138c25b729aee5c2ae8cad6

2. Execute :

docker container run -it --rm -v $(pwd):/usr/src/app --cap-add=SYS_ADMIN zenika/alpine-chrome:with-playwright npm start

3. See error

> [email protected] start /usr/src/app
> node src/useragent

internal/modules/cjs/loader.js:969
  throw err;
  ^

Error: Cannot find module 'playwright-chromium'
Require stack:
- /usr/src/app/src/useragent.js
    at Function.Module._resolveFilename (internal/modules/cjs/loader.js:966:15)
    at Function.Module._load (internal/modules/cjs/loader.js:842:27)
    at Module.require (internal/modules/cjs/loader.js:1026:19)
    at require (internal/modules/cjs/helpers.js:72:18)
    at Object.<anonymous> (/usr/src/app/src/useragent.js:1:22)
    at Module._compile (internal/modules/cjs/loader.js:1138:30)
    at Object.Module._extensions..js (internal/modules/cjs/loader.js:1158:10)
    at Module.load (internal/modules/cjs/loader.js:986:32)
    at Function.Module._load (internal/modules/cjs/loader.js:879:14)
    at Function.executeUserEntryPoint [as runMain] (internal/modules/run_main.js:71:12) {
  code: 'MODULE_NOT_FOUND',
  requireStack: [ '/usr/src/app/src/useragent.js' ]
}
npm ERR! code ELIFECYCLE
npm ERR! errno 1
npm ERR! [email protected] start: `node src/useragent`
npm ERR! Exit status 1
npm ERR!
npm ERR! Failed at the [email protected] start script.
npm ERR! This is probably not a problem with npm. There is likely additional logging output above.
npm WARN Local package.json exists, but node_modules missing, did you mean to install?

npm ERR! A complete log of this run can be found in:
npm ERR!     /home/chrome/.npm/_logs/2020-06-17T10_09_15_847Z-debug.log

What is the expected behavior?
As per the documentation:

A example-chromium.png file will be created in your with-playwright folder.

What is the actual behavior?

The above output - no png file created.

Possible solution
(Not found yet)

Logs
Log file 2020-06-17T10_09_15_847Z-debug.log

Versions

Software	Version(s)	How-to
Chromium	Chromium 83.0.4103.61	docker run -it --rm --entrypoint "" zenika/alpine-chrome chromium-browser --version
Image	zenika/alpine-chrome@sha256:c3a832060aa529bd0ca87bc60ff669015bcc5c8d5138c25b729aee5c2ae8cad6	docker image inspect --format='{{index .RepoDigests 0}}' zenika/alpine-chrome
Docker Engine Client	19.03.11	docker version
Docker Engine Server	19.03.5	docker version
Docker Machine	0.16.2, build bd45ab1	docker-machine version
Darwin Kernel Version	19.3.0
MacOS version	Catalina 10.15.3 (19D76)

Additional context
I also tried to replace require("playwright-chromium"); with require("playwright"); nothing better.
I found the following issue but it's not the same version so I'm not sure it's relevant.

Update the README with the provided features

Context

I wrote a post on dev.to. It could be nice to reuse the content to complete the README.md

Add Autotest to test the alpine-chrome image behavior

Have a look to https://docs.docker.com/docker-hub/builds/automated-testing/

chromium-browser not starting

Describe the bug
First I want to thank you for this image, which is really helpful for our build system.
However, since today it seems like chromium fails to start with, displaying the following error:

Error relocating /usr/lib/chromium/chrome: _ZNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEC1Ev: symbol not found

It looks like this happens since the last image was pushed (5 hours ago)

To Reproduce
Steps to reproduce the behavior:

1. Execute docker container run -it --rm --entrypoint "" zenika/alpine-chrome chromium-browser --version
2. See error

Error relocating /usr/lib/chromium/chrome: _ZNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEC1Ev: symbol not found
Error relocating /usr/lib/chromium/chrome: _ZNSt7__cxx1119basic_ostringstreamIcSt11char_traitsIcESaIcEEC1Ev: symbol not found
Error relocating /usr/lib/chromium/chrome: _ZNSt19_Sp_make_shared_tag5_S_eqERKSt9type_info: symbol not found

I was able to reproduce the issue on multiple machines (mac and linux)

What is the expected behavior?
Chromium should be able to start and display its version

What is the actual behavior?
Chromium crashes

Possible solutions
I don't know yet but I'll try to dig in see if I find something.
If anybody knows how to fix it right away that would be greatly appreciated.

Versions

Software	Version(s)	How-to
Chromium	?	docker run -it --rm --entrypoint "" zenika/alpine-chrome chromium-browser --version
Image	zenika/alpine-chrome@sha256:794b317131be727d5b7489e42f28d62f400d1aa7ffade0948f0fd7942e9d27c9	docker image inspect --format='{{index .RepoDigests 0}}' zenika/alpine-chrome
Docker Engine	19.03.2	docker version

Dockerfile cleanups do not minimize image size

Hey,

first of all - thank you for this repo. You helped me in solving quite an annoying issue.
Thank you!

Secondly, I've found in Dockerfile quite intriguing lines:

alpine-chrome/Dockerfile

Lines 22 to 31 in 2d0b861

	# Minimize size
	RUN apk del --purge --force linux-headers binutils-gold gnupg zlib-dev libc-utils
	RUN rm -rf /var/lib/apt/lists/* \
	/var/cache/apk/* \
	/usr/share/man \
	/tmp/* \
	/usr/lib/node_modules/npm/man \
	/usr/lib/node_modules/npm/doc \
	/usr/lib/node_modules/npm/html \
	/usr/lib/node_modules/npm/scripts

Just wanted to mention that this will not minimize image size.

Each command creates a layer. Each layer adds a size of appended changes. Running cleanup commands in a separate layer will append 0-byte changes. It won't affect overall image size

There're few ways to fix this:

• build image with --squash option
• append cleanup tasks to previous RUN
• create a separate installation script, copy it to image and run. This will create only two layers of changes (COPY of script and RUN which will install&cleanup)

Cheers! :)

How can I remove page number when I use --print-to-pdf?

How can I remove page number when I use --print-to-pdf?