Migrate instructions #509
Migrate instructions #509
Conversation
Fork master
k8s-ci-robot
added
release-note-none
|
Hi @mattcary. Thanks for your PR. I'm waiting for a kubernetes-sigs member to verify that this patch is reasonable to test. If it is, they should reply with Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
k8s-ci-robot
added
the
size/L
|
/cc @msau42 |
|
/ok-to-test |
k8s-ci-robot
added
ok-to-test
| cluster with the PD snapshot handle from the alpha snapshot, then bind that to a | ||
| new `VolumeSnapshot`. | ||
|
|
||
| If you are able to set the `deletionPolicy` of any existing |
There was a problem hiding this comment.
Throughout the doc, it would be good to clarify if you're referring to the alpha version or beta version of the object.
| PD CSI driver deployment. This may be a safer option if you are not sure about | ||
| the deletion policy of your alpha snapshot. | ||
|
|
||
| #### Steps to restore from an existing snapshot. |
There was a problem hiding this comment.
These steps are useful even outside the context of trying to migrate from alpha to beta. Maybe this should be pulled out of the migration section and made more top level. Then in the migration section, you can refer to the "import an existing snapshot" section.
There was a problem hiding this comment.
Yeah, that's a good point. I was a little surprised myself that I couldn't find any guide on manually provisioning snapshot contents.
There was a problem hiding this comment.
I've done this, creating a migration section and referring to that from the migration tips section.
mattcary
force-pushed
the
migrate-instructions
branch
2 times, most recently
from
746d42b to
9d8bff0
Compare
| @@ -122,3 +125,150 @@ | |||
| lost+found | |||
| sample-file.txt | |||
| ``` | |||
|
|
|||
| ### Manually Provisioning Snapshot Contents. | |||
There was a problem hiding this comment.
What do you think of calling it something like "Import a pre-existing snapshot"? I think this description aligns better with what a user would want to accomplish without using Kubernetes terms.
| #### Troubleshooting | ||
|
|
||
| If the `VolumeSnapshot`, `VolumeSnapshotContents` and `PersistentVolumeClaim` | ||
| are not all mutually syncrhonized, the pod will not start. Try `kubectl describe |
| snapshotter. In many cases, snapshots created with the alpha will not be able to | ||
| be used with a beta driver. In some cases, they can be migrated. | ||
|
|
||
| *You may lose all data in your snapshot!* |
There was a problem hiding this comment.
Bold and add "Attention" similar to the other notes
| snapshot-controller and PD CSI driver: | ||
|
|
||
| * snapshot-controller:v2.0.1 | ||
| * csi-provisioner:v1.5.0-gke.0 |
|
|
||
| #### Migrating by Restoring from a Manually Provisioned Snapshot | ||
|
|
||
| The idea is to provision a `VolumeSnapshotContents` in your new or upgraded beta |
There was a problem hiding this comment.
clarify this is a beta VolumeSnapshotContent
| the deletion policy of your alpha snapshot. | ||
|
|
||
| After confirming that the PD snapshot is available, restore it using the beta PD | ||
| CSI driver as described above in "Manually Provisioning Snapshot Contents". The |
There was a problem hiding this comment.
Can you make this a relative link?
|
|
||
| After confirming that the PD snapshot is available, restore it using the beta PD | ||
| CSI driver as described above in "Manually Provisioning Snapshot Contents". The | ||
| restoration should in a cluster with the beta PD CSI driver installed. At not |
|
|
||
| #### Migrating by Restoring from a Manually Provisioned Snapshot | ||
|
|
||
| The idea is to provision a `VolumeSnapshotContents` in your new or upgraded beta |
There was a problem hiding this comment.
I think this actually won't work if you upgrade a cluster right? Because the alpha CRD will be still be installed, preventing the beta CRD from being installed. So creation of the beta objects won't succeed until the alpha CRD is deleted.
There was a problem hiding this comment.
I'll clarify that an upgrade means the alpha components are removed.
I don't want to get in to uninstalling everything; I tried it and it's a major PITA due to all the finalizers cross-linked between everything. It's really easy to get into a state where you have to patch out finalizers manually just to get resources to go away. Not to mention the problems of accidentally trying to delete CRDs while there are still resources around (eg, because of the finalizers).
mattcary
force-pushed
the
migrate-instructions
branch
3 times, most recently
from
d8fbc41 to
1491a6c
Compare
|
|
||
| After confirming that the PD snapshot is available, restore it using the beta PD | ||
| CSI driver as described above in [Import a Pre-Existing | ||
| Snapshot](#import-a-pre-existing-snapshot). The restoration should in a cluster |
| then bind that to a new beta `VolumeSnapshot`. | ||
|
|
||
| Note that if using the same upgraded cluster where the alpha driver was | ||
| installed, the alpha driver must be completely removed, including its CRDs. |
There was a problem hiding this comment.
Would it be useful to note somewhere here that uninstalling the CRD will delete all the CR objects, so it's best to save all the snapshot handles before doing that?
There was a problem hiding this comment.
I guess you also should save the information about what workload it was associated with too
There was a problem hiding this comment.
How does that sound?
FWIW I wasn't actually able to uninstall the driver & CRDs without wedging things. That's why I added the advice about spinning up a new cluster.
There was a problem hiding this comment.
If you weren't able to get it to work, then maybe let's just say in the docs that you'll need to migrate to a new cluster. The advice about saving the objects from the old cluster still stand.
There was a problem hiding this comment.
OK, I've made it slightly stronger.
mattcary
force-pushed
the
migrate-instructions
branch
2 times, most recently
from
ff3a466 to
fdf6eb8
Compare
mattcary
force-pushed
the
migrate-instructions
branch
from
fdf6eb8 to
7f402a6
Compare
|
/lgtm |
k8s-ci-robot
added
the
lgtm
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: mattcary, msau42 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
/kind documentation
What this PR does / why we need it:
Add guidance on migrating from alpha snapshot version
Does this PR introduce a user-facing change?: