Docker-selenium: session deleted because of page crash from unknown error: cannot determine loading status from tab crashed

Created on 6 Oct 2017  路  7Comments  路  Source: SeleniumHQ/docker-selenium

Chrome browser crashes on docker-selenium instance with error message : session deleted because of page crash from unknown error: cannot determine loading status from tab crashed.

Meta -

Image(s):
selenium/standalone-chrome

Docker-Selenium Image Version(s): selenium/standalone-chrome:3.4.0 & 3.3.0

Browser : Chrome

Version : 59.0

Expected Behavior -

Browser should not crash in middle of the execution.

Actual Behavior -

Browser is getting crashed in middle of the execution and session got closed.

Anyone has any idea regarding this? Help would be appreciated.

Most helpful comment

@vamshiangala
I am not sure how it works in that specific environment, but I would double check and make sure that the running container /dev/shm size is at least 2GB (the default size is 64MB). Is there a way for you to do that? Go inside the container and check it, I am sure the EC2 instance has it but I am not sure if the container is actually using it.

All 7 comments

@vamshiangala
Can you share the docker command you use to start the containers?

@diemol
Below is the command used to start the docker containers.

$ docker run -d -p 4444:4444 selenium/standalone-chrome

This is a know thing.
Looks like you are not sharing the /dev/shm volume. Check: https://github.com/SeleniumHQ/docker-selenium#running-the-images

@diemol
We have pulled docker image for selenium standalone chrome in a instance and converted into AMI and added to a cluster. While we are creating Task Definition for this we are defining shared memory (/dev/shm).

As we are predefining shared memory in task definition, we are not giving /dev/shm (2GB) as a parameter while starting docker instance. We can also see the shared memory (/dev/shm) in the file system when we log into the instance and check.

But still chrome is crashing while execution.

@vamshiangala
I am not sure how it works in that specific environment, but I would double check and make sure that the running container /dev/shm size is at least 2GB (the default size is 64MB). Is there a way for you to do that? Go inside the container and check it, I am sure the EC2 instance has it but I am not sure if the container is actually using it.

This worked for us, thank you for the /dev/shm tip @diemol

Closing this issue due to inactivity from original OP, feel free to reopen in case more info is provided.

Was this page helpful?
0 / 5 - 0 ratings

Related issues

vikramvi picture vikramvi  路  6Comments

lluisteamcmp picture lluisteamcmp  路  6Comments

westlakem picture westlakem  路  5Comments

peterstory picture peterstory  路  4Comments

matthewsamari picture matthewsamari  路  4Comments