config: relax specification of Config.Volumes#694
config: relax specification of Config.Volumes#694vbatts merged 2 commits intoopencontainers:masterfrom
Conversation
config.md
Outdated
| A set of directories which SHOULD be created as data volumes in a container running this image. | ||
| If a file or folder exists within the image with the same path as a data volume, that file or folder will be replaced by the data volume and never be merged. | ||
| A set of directories describing where the process is likely write data specific to a container instance. | ||
| Implementations SHOULD provide mounts for these locations such that application data is not written to the container's root filesystem. |
There was a problem hiding this comment.
In this sentence, "implementations" seems to be directed at runtime implementations, right?
There was a problem hiding this comment.
I was explicitly vague on purpose. Should I qualify this further?
There was a problem hiding this comment.
IMO if we want to specify what a runtime should do we should put it in conversion.md, to avoid the mixing of semantics between the two.
config.md
Outdated
| If a file or folder exists within the image with the same path as a data volume, that file or folder will be replaced by the data volume and never be merged. | ||
| A set of directories describing where the process is likely write data specific to a container instance. | ||
| Implementations SHOULD provide mounts for these locations such that application data is not written to the container's root filesystem. | ||
| If a _new_ image is created from a container based on the image described by this configuration, data in these paths SHOULD NOT be included in the _new_ image. |
There was a problem hiding this comment.
So we are excluding the use case of seeding a volume?
There was a problem hiding this comment.
No, this is covering the case where an image is created from a running container. This is pretty awkward.
There was a problem hiding this comment.
I'll add a line above this indicating that seeding volumes with image data is okay.
|
needs a rebase to pass tests. |
stevvooe
force-pushed
the
relax-volumes-specification
branch
from
f211f5f to
950dcf5
Compare
Relaxes the specification of `Config.Volumes` by avoiding references to the concept of "data volumes". Implementors are merely instructed to provide mounts outside the container's root filesystem. Signed-off-by: Stephen J Day <stephen.day@docker.com>
Signed-off-by: Stephen J Day <stephen.day@docker.com>
stevvooe
force-pushed
the
relax-volumes-specification
branch
from
950dcf5 to
8f42721
Compare
|
@vbatts PTAL. Moved to conversion. Added clause about seeding the mount. |
|
Thanks for listening to my carping, I know it's a bit much at times. LGTM! |
I think this language is much better. Thanks for the feedback! |
|
i think this covers this loosely enough. I would love to get @cyphar review as well |
|
LGTM |
|
@cyphar please... review... resisting... urge... to... merge... with... two... LGTMs. |
|
Haha, thanks for pinging me @stevvooe. LGTM, sorry for the delay (exams are fun). |
|
👐 |
5 participants
Relaxes the specification of
Config.Volumesby avoiding references tothe concept of "data volumes". Implementors are merely instructed to
provide mounts outside the container's root filesystem.
Signed-off-by: Stephen J Day stephen.day@docker.com
Closes #687
Supersedes #504
cc @vbatts @cyphar