[Bug]: The docker image does not launch since 0.9.1

[8176135]

crawl4ai version

0.9.1

Expected Behavior

Following the install guide:

docker run -d \
  -p 11235:11235 \
  --name crawl4ai \
  --shm-size=1g \
  unclecode/crawl4ai:latest

should work.

Current Behavior

However it repeatedly crashes, downgrading to 0.9.0's image works fine.

Issue seems to be:

playwright._impl._errors.Error: BrowserType.launch: Executable doesn't exist at /home/appuser/.cache/ms-playwright/chromium_headless_shell-1228/chrome-headless-shell-linux64/chrome-headless-shell

where it's looking for the headless-shell. However in your Dockerfile it's only copying the full chrome binary, so it doesn't exist, and crashes.

Is this reproducible?

Yes

Inputs Causing the Bug

Steps to Reproduce

Code snippets

OS

Linux

Python version

Whatever version is bundled inside the docker image

Browser

No response

Browser version

No response

Error logs & Screenshots (if applicable)

I am running podman instead of docker, but I believe the same issue would occur in docker as downgrading to 0.9.0 works fine.

[TheRealChickenlegs]

Came to report the same thing, since update container continually restarts in docker as well. Reverting to 0.9.0 resolves the issue.

This the log produced in docker:

crawl4ai  | entrypoint: no REDIS_PASSWORD provided; generated an ephemeral one.
crawl4ai  | 2026-07-10 03:15:51,853 INFO supervisord started with pid 1
crawl4ai  | 2026-07-10 03:15:52,858 INFO spawned: 'redis' with pid 9
crawl4ai  | 2026-07-10 03:15:52,861 INFO spawned: 'gunicorn' with pid 10
crawl4ai  | 9:C 10 Jul 2026 03:15:52.878 # WARNING Memory overcommit must be enabled! Without it, a background save or replication may fail under low memory condition. Being disabled, it can also cause failures without low memory condition, see https://github.com/jemalloc/jemalloc/issues/1328. To fix this issue add 'vm.overcommit_memory = 1' to /etc/sysctl.conf and then reboot or run the command 'sysctl vm.overcommit_memory=1' for this to take effect.
crawl4ai  | 9:C 10 Jul 2026 03:15:52.878 * oO0OoO0OoO0Oo Redis is starting oO0OoO0OoO0Oo
crawl4ai  | 9:C 10 Jul 2026 03:15:52.878 * Redis version=8.8.0, bits=64, commit=00000000, modified=1, pid=9, just started
crawl4ai  | 9:C 10 Jul 2026 03:15:52.878 * Configuration loaded
crawl4ai  | 9:M 10 Jul 2026 03:15:52.879 * Increased maximum number of open files to 10032 (it was originally set to 1024).
crawl4ai  | 9:M 10 Jul 2026 03:15:52.879 * monotonic clock: POSIX clock_gettime
crawl4ai  | 9:M 10 Jul 2026 03:15:52.880 * Running mode=standalone, port=6379.
crawl4ai  | 9:M 10 Jul 2026 03:15:52.881 * Server initialized
crawl4ai  | 9:M 10 Jul 2026 03:15:52.881 * Ready to accept connections tcp
crawl4ai  | [2026-07-10 03:15:53 +0000] [10] [INFO] Starting gunicorn 26.0.0
crawl4ai  | [2026-07-10 03:15:53 +0000] [10] [INFO] Listening at: http://[::]:11235 (10)
crawl4ai  | [2026-07-10 03:15:53 +0000] [10] [INFO] Using worker: uvicorn.workers.UvicornWorker
crawl4ai  | [2026-07-10 03:15:53 +0000] [16] [INFO] Booting worker with pid: 16
crawl4ai  | [2026-07-10 03:15:53 +0000] [10] [INFO] Control socket listening at /home/appuser/.gunicorn/gunicorn.ctl
crawl4ai  | WARNING:crawl4ai.security:No SECRET_KEY set. Auto-generated an ephemeral key (changes on restart, invalidating issued tokens). Set SECRET_KEY for any real deployment.
crawl4ai  | 2026-07-10 03:15:53,959 INFO success: redis entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)
crawl4ai  | 2026-07-10 03:15:53,959 INFO success: gunicorn entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)
crawl4ai  | /app/job.py:49: UserWarning: Field name "schema" in "LlmJobPayload" shadows an attribute in parent "BaseModel"
crawl4ai  |   class LlmJobPayload(BaseModel):
crawl4ai  | WARNING:crawl4ai.security:trusted_hosts is ['*'] on a non-loopback bind (0.0.0.0): the Host guard is disabled. Set security.trusted_hosts to your real hostname(s).
crawl4ai  | /app/server.py:1006: FastAPIDeprecationWarning: `regex` has been deprecated, please use `pattern` instead
crawl4ai  |   context_type: str = Query("all", regex="^(code|doc|all)$"),
crawl4ai  | MCP server running on 0.0.0.0:11235
crawl4ai  | [2026-07-10 03:15:54 +0000] [16] [INFO] Started server process [16]
crawl4ai  | [2026-07-10 03:15:54 +0000] [16] [INFO] Waiting for application startup.
crawl4ai  | [2026-07-10 03:15:54 +0000] [16] [ERROR] Traceback (most recent call last):
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/starlette/routing.py", line 638, in lifespan
crawl4ai  |     async with self.lifespan_context(app) as maybe_state:
crawl4ai  |                ^^^^^^^^^^^^^^^^^^^^^^^^^^
crawl4ai  |   File "/usr/local/lib/python3.12/contextlib.py", line 210, in __aenter__
crawl4ai  |     return await anext(self.gen)
crawl4ai  |            ^^^^^^^^^^^^^^^^^^^^^
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/fastapi/routing.py", line 216, in merged_lifespan
crawl4ai  |     async with original_context(app) as maybe_original_state:
crawl4ai  |                ^^^^^^^^^^^^^^^^^^^^^
crawl4ai  |   File "/usr/local/lib/python3.12/contextlib.py", line 210, in __aenter__
crawl4ai  |     return await anext(self.gen)
crawl4ai  |            ^^^^^^^^^^^^^^^^^^^^^
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/fastapi/routing.py", line 216, in merged_lifespan
crawl4ai  |     async with original_context(app) as maybe_original_state:
crawl4ai  |                ^^^^^^^^^^^^^^^^^^^^^
crawl4ai  |   File "/usr/local/lib/python3.12/contextlib.py", line 210, in __aenter__
crawl4ai  |     return await anext(self.gen)
crawl4ai  |            ^^^^^^^^^^^^^^^^^^^^^
crawl4ai  |   File "/app/server.py", line 199, in lifespan
crawl4ai  |     await init_permanent(BrowserConfig(
crawl4ai  |   File "/app/crawler_pool.py", line 141, in init_permanent
crawl4ai  |     await PERMANENT.start()
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/crawl4ai/async_webcrawler.py", line 183, in start
crawl4ai  |     await self.crawler_strategy.__aenter__()
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/crawl4ai/async_crawler_strategy.py", line 121, in __aenter__
crawl4ai  |     await self.start()
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/crawl4ai/async_crawler_strategy.py", line 131, in start
crawl4ai  |     await self.browser_manager.start()
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/crawl4ai/browser_manager.py", line 938, in start
crawl4ai  |     self.browser = await self.playwright.chromium.launch(**browser_args)
crawl4ai  |                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/playwright/async_api/_generated.py", line 16546, in launch
crawl4ai  |     await self._impl_obj.launch(
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/playwright/_impl/_browser_type.py", line 97, in launch
crawl4ai  |     await self._channel.send(
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/playwright/_impl/_connection.py", line 69, in send
crawl4ai  |     return await self._connection.wrap_api_call(
crawl4ai  |            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
crawl4ai  |   File "/usr/local/lib/python3.12/site-packages/playwright/_impl/_connection.py", line 563, in wrap_api_call
crawl4ai  |     raise rewrite_error(error, f"{parsed_st['apiName']}: {error}") from None
crawl4ai  | playwright._impl._errors.Error: BrowserType.launch: Executable doesn't exist at /home/appuser/.cache/ms-playwright/chromium_headless_shell-1228/chrome-headless-shell-linux64/chrome-headless-shell
crawl4ai  | ╔════════════════════════════════════════════════════════════╗
crawl4ai  | ║ Looks like Playwright was just installed or updated.       ║
crawl4ai  | ║ Please run the following command to download new browsers: ║
crawl4ai  | ║                                                            ║
crawl4ai  | ║     playwright install                                     ║
crawl4ai  | ║                                                            ║
crawl4ai  | ║ <3 Playwright Team                                         ║
crawl4ai  | ╚════════════════════════════════════════════════════════════╝
crawl4ai  | 
crawl4ai  | [2026-07-10 03:15:54 +0000] [16] [ERROR] Application startup failed. Exiting.
crawl4ai  | [2026-07-10 03:15:54 +0000] [16] [INFO] Worker exiting (pid: 16)
crawl4ai  | [2026-07-10 03:15:54 +0000] [10] [ERROR] Worker (pid:16) exited with code 3.
crawl4ai  | 2026-07-10 03:15:54,629 INFO reaped unknown pid 38 (exit status 0)
crawl4ai  | [2026-07-10 03:15:54 +0000] [10] [ERROR] Shutting down: Master
crawl4ai  | [2026-07-10 03:15:54 +0000] [10] [ERROR] Reason: Worker failed to boot.

[krim404]

if i recall correctly, its because the docker bundles the full version of chrome-shell and not the headless anymore.

[ntohidi]

Thanks for the report, @8176135. The fix is on develop — the Dockerfile now copies chromium_headless_shell-* alongside chromium-* into the runtime cache. Will be in the next release very soon.