Update documentation for v2 release #276
Conversation
k8s-ci-robot
added
release-note-none
| @@ -0,0 +1,137 @@ | |||
| # Usage | |||
There was a problem hiding this comment.
Thanks a lot for creating this doc! I think it could be part of README, what do you think?
There was a problem hiding this comment.
It's a bit long and has a lot of nuanced details re. containerd and HostProcess containers. I think it might be easier to keep this in a separate doc and point to it. I like to keep READMEs a hub of pointers to relevant information.
docs/USAGE.md
Outdated
| - [FileSystem](/pkg/filesystem/) | ||
| - [iSCSI](/pkg/iscsi/) (experimental) | ||
| - [SMB](/pkg/smb/) | ||
| - [System](/pkg/system/) |
There was a problem hiding this comment.
is this also experimental? I remember that there were 2 API Groups that didn't move from alpha in v1
There was a problem hiding this comment.
You're right. System is in alpha. Also, FileSystem and Volume are both v2alpha1, but we don't really have to consider them experimental, since we can just bump Go versions for API changes.
For System, at least we have tests running. I will also mark System as experimental.
docs/USAGE.md
Outdated
| image: k8s.gcr.io/sig-storage/csi-node-driver-registrar:v2.1.0 | ||
| args: | ||
| - --v=5 | ||
| - --csi-address=unix://c:\var\lib\kubelet\plugins\pd.csi.storage.gke.io\csi.sock |
There was a problem hiding this comment.
let's use placeholders for the CSI Driver instead of using PD CSI Driver as an example
docs/USAGE.md
Outdated
|
|
||
| Using HostProcess pods have a few important consequences on the deployed containers, as noted in the HostProcess docs. | ||
| - HostProcess containers have no file system or resource isolation, so they have a complete view of the host machine’s file system. | ||
| This means that paths passed in **arguments need to be complete host machine paths**. |
There was a problem hiding this comment.
| This means that paths passed in **arguments need to be complete host machine paths**. | |
| Paths in the PodSpec must be absolute paths with respect to the host. |
docs/USAGE.md
Outdated
| On the other hand, depending on the containerd and the Windows OS version, Kubernetes volume mounts are either mounted relative to a subdirectory of the host machine specified by an environment variable `%CONTAINER_SANDBOX_MOUNT_POINT%` or **mounted relative to the host process root**. See [HostProcess Caveats](#hostprocess-caveats). | ||
| - HostProcess pods can only contain HostProcess containers. | ||
| Often, the CSI node registrar is deployed in the same pod as the driver, so file paths for both containers need to be updated. | ||
| - Named pipes and Unix domain sockets are not supported and should be accessed via their path on the host. |
There was a problem hiding this comment.
| - Named pipes and Unix domain sockets are not supported and should be accessed via their path on the host. | |
| - Named pipes and Unix domain sockets are not supported and should be accessed via their absolute path with respect to the host. |
alexander-ding
force-pushed
the
enh/migrate-v2
branch
from
5f22bee to
857ef50
Compare
alexander-ding
force-pushed
the
enh/migrate-v2
branch
from
857ef50 to
098ffd6
Compare
|
/lgtm Thanks! |
k8s-ci-robot
added
the
lgtm
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: alexander-ding, mauriciopoppe The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
k8s-ci-robot
added
the
approved
What type of PR is this?
What this PR does / why we need it:
Which issue(s) this PR fixes:
Part of #243.
Special notes for your reviewer:
Does this PR introduce a user-facing change?:
/cc @mauriciopoppe