Small improvements

[markjgardner]

Added a few best practices and minor improvements to your package.json example.

[simonua]

Thank you, @markjgardner! I'll look this over to understand it better and incorporate.

[simonua]

@markjgardner, what is the process of updating runtest.js in the volume?

[markjgardner]

The point of putting it in a volume is that it is decoupled from the container. The assumption is that your test will change (much) more frequently than your chrome or selenium versions. So you can just publish new tests to the volume and restart the container vs having to build and publish a new image.

How you update the tests is up to you and dependent on how you are going to run this container. If it is for local development, then just using a bind mount per example.sh will let you work locally on the tests and just do a docker run ... whenever you want to validate. If, however, you want to run this container in k8s, then you would likely have a CD pipeline that pushed your tests to a blob container or file share which was subsequently mounted to the pods.

Make sense?

[simonua]

Yes, definitely. I'll pursue this further. Thank you!

[markjgardner]

Added a few best practices and minor improvements to your package.json example.

[simonua]

Thank you, @markjgardner! I'll look this over to understand it better and incorporate.

[simonua]

@markjgardner, what is the process of updating runtest.js in the volume?

[markjgardner]

The point of putting it in a volume is that it is decoupled from the container. The assumption is that your test will change (much) more frequently than your chrome or selenium versions. So you can just publish new tests to the volume and restart the container vs having to build and publish a new image.

How you update the tests is up to you and dependent on how you are going to run this container. If it is for local development, then just using a bind mount per example.sh will let you work locally on the tests and just do a docker run ... whenever you want to validate. If, however, you want to run this container in k8s, then you would likely have a CD pipeline that pushed your tests to a blob container or file share which was subsequently mounted to the pods.

Make sense?

[simonua]

Yes, definitely. I'll pursue this further. Thank you!

[markjgardner]

Added a few best practices and minor improvements to your package.json example.

[simonua]

Thank you, @markjgardner! I'll look this over to understand it better and incorporate.

[simonua]

@markjgardner, what is the process of updating runtest.js in the volume?

[markjgardner]

The point of putting it in a volume is that it is decoupled from the container. The assumption is that your test will change (much) more frequently than your chrome or selenium versions. So you can just publish new tests to the volume and restart the container vs having to build and publish a new image.

How you update the tests is up to you and dependent on how you are going to run this container. If it is for local development, then just using a bind mount per example.sh will let you work locally on the tests and just do a docker run ... whenever you want to validate. If, however, you want to run this container in k8s, then you would likely have a CD pipeline that pushed your tests to a blob container or file share which was subsequently mounted to the pods.

Make sense?

[simonua]

Yes, definitely. I'll pursue this further. Thank you!

[markjgardner]

Added a few best practices and minor improvements to your package.json example.

[simonua]

Thank you, @markjgardner! I'll look this over to understand it better and incorporate.

[simonua]

@markjgardner, what is the process of updating runtest.js in the volume?

[markjgardner]

The point of putting it in a volume is that it is decoupled from the container. The assumption is that your test will change (much) more frequently than your chrome or selenium versions. So you can just publish new tests to the volume and restart the container vs having to build and publish a new image.

How you update the tests is up to you and dependent on how you are going to run this container. If it is for local development, then just using a bind mount per example.sh will let you work locally on the tests and just do a docker run ... whenever you want to validate. If, however, you want to run this container in k8s, then you would likely have a CD pipeline that pushed your tests to a blob container or file share which was subsequently mounted to the pods.

Make sense?

[simonua]

Yes, definitely. I'll pursue this further. Thank you!

[markjgardner]

Added a few best practices and minor improvements to your package.json example.

[simonua]

Thank you, @markjgardner! I'll look this over to understand it better and incorporate.

[simonua]

@markjgardner, what is the process of updating runtest.js in the volume?

[markjgardner]

The point of putting it in a volume is that it is decoupled from the container. The assumption is that your test will change (much) more frequently than your chrome or selenium versions. So you can just publish new tests to the volume and restart the container vs having to build and publish a new image.

How you update the tests is up to you and dependent on how you are going to run this container. If it is for local development, then just using a bind mount per example.sh will let you work locally on the tests and just do a docker run ... whenever you want to validate. If, however, you want to run this container in k8s, then you would likely have a CD pipeline that pushed your tests to a blob container or file share which was subsequently mounted to the pods.

Make sense?

[simonua]

Yes, definitely. I'll pursue this further. Thank you!

[markjgardner]

Added a few best practices and minor improvements to your package.json example.

[simonua]

Thank you, @markjgardner! I'll look this over to understand it better and incorporate.

[simonua]

@markjgardner, what is the process of updating runtest.js in the volume?

[markjgardner]

The point of putting it in a volume is that it is decoupled from the container. The assumption is that your test will change (much) more frequently than your chrome or selenium versions. So you can just publish new tests to the volume and restart the container vs having to build and publish a new image.

How you update the tests is up to you and dependent on how you are going to run this container. If it is for local development, then just using a bind mount per example.sh will let you work locally on the tests and just do a docker run ... whenever you want to validate. If, however, you want to run this container in k8s, then you would likely have a CD pipeline that pushed your tests to a blob container or file share which was subsequently mounted to the pods.

Make sense?

[simonua]

Yes, definitely. I'll pursue this further. Thank you!

[markjgardner]

Added a few best practices and minor improvements to your package.json example.

[simonua]

Thank you, @markjgardner! I'll look this over to understand it better and incorporate.

[simonua]

@markjgardner, what is the process of updating runtest.js in the volume?

[markjgardner]

The point of putting it in a volume is that it is decoupled from the container. The assumption is that your test will change (much) more frequently than your chrome or selenium versions. So you can just publish new tests to the volume and restart the container vs having to build and publish a new image.

How you update the tests is up to you and dependent on how you are going to run this container. If it is for local development, then just using a bind mount per example.sh will let you work locally on the tests and just do a docker run ... whenever you want to validate. If, however, you want to run this container in k8s, then you would likely have a CD pipeline that pushed your tests to a blob container or file share which was subsequently mounted to the pods.

Make sense?

[simonua]

Yes, definitely. I'll pursue this further. Thank you!

[markjgardner]

Added a few best practices and minor improvements to your package.json example.

[simonua]

Thank you, @markjgardner! I'll look this over to understand it better and incorporate.

[simonua]

@markjgardner, what is the process of updating runtest.js in the volume?

[markjgardner]

The point of putting it in a volume is that it is decoupled from the container. The assumption is that your test will change (much) more frequently than your chrome or selenium versions. So you can just publish new tests to the volume and restart the container vs having to build and publish a new image.

How you update the tests is up to you and dependent on how you are going to run this container. If it is for local development, then just using a bind mount per example.sh will let you work locally on the tests and just do a docker run ... whenever you want to validate. If, however, you want to run this container in k8s, then you would likely have a CD pipeline that pushed your tests to a blob container or file share which was subsequently mounted to the pods.

Make sense?

[simonua]

Yes, definitely. I'll pursue this further. Thank you!