You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
* I will remove the outdated OADP 1.1.x and 1.2.x documentation.
* restore upgrade notes from 1.1 through
* keep restic troubleshooting
---------
Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>
### Issue with Backup/Restore of DeploymentConfig with volumes or restore hooks
252
243
253
-
- (OADP 1.3+) **Error:** `DeploymentConfigs restore with spec.Replicas==0 or DC pods fail to restart if they crash if using DC with volumes or restore hooks`
254
-
255
-
**Solution:**
256
-
257
-
Solution is the same as in the (OADP 1.1+), except it applies to the use case if you are restoring DeploymentConfigs and have either volumes or post-restore hooks regardless of the backup method.
258
-
259
-
- (OADP 1.1+) **Error:** `DeploymentConfigs restore with spec.Replicas==0 or DC pods fail to restart if they crash if using Restic/Kopia restores or restore hooks`
260
-
261
-
**Solution:**
262
-
263
-
This is expected behavior on restore if you are restoring DeploymentConfigs and are either using Restic or Kopia for volume restore or you have post-restore hooks. The pod and DC plugins make these modifications to ensure that Restic or Kopia and hooks work properly, and [dc-post-restore.sh](../docs/scripts/dc-post-restore.sh) should have been run immediately after a successful restore. Usage for this script is `dc-post-restore.sh <restore-name>`
264
-
265
-
- (OADP 1.0.z) **Error:** `Using Restic as backup method causes PartiallyFailed/Failed errors in the Restore or post-restore hooks fail to execute`
244
+
- **Error:** `DeploymentConfigs restore with spec.Replicas==0 or DC pods fail to restart if they crash if using DC with volumes or restore hooks`
266
245
267
246
**Solution:**
268
247
269
-
The changes in the backup/restore process for mitigating this error would be a two step restore process where, in the first step we would perform a restore excluding the replicationcontroller and deploymentconfig resources, and the second step would involve a restore including these resources. The backup and restore commands are given below for more clarity. (The examples given below are a use case for backup/restore of a target namespace, for other cases a similar strategy can be followed).
270
-
271
-
Please note that this is a temporary fix for this issue and there are ongoing discussions to solve it.
272
-
273
-
Step 1: Initiate the backup as any normal backup for restic.
This is expected behavior on restore if you are restoring DeploymentConfigs and have either volumes or post-restore hooks. The pod and DC plugins make these modifications to ensure that Restic or Kopia and hooks work properly, and [dc-post-restore.sh](../docs/scripts/dc-post-restore.sh) should have been run immediately after a successful restore. Usage for this script is `dc-post-restore.sh <restore-name>`
287
249
288
250
### New Restic Backup Partially Failing After Clearing Bucket
-`CSI`[Plugins for CSI](https://github.com/vmware-tanzu/velero-plugin-for-csi)
20
20
-`kubevirt`[Plugins for Kubevirt](https://github.com/kubevirt/kubevirt-velero-plugin)
21
-
-`VSM (OADP 1.2 or below)`[Plugin for Volume-Snapshot-Mover](https://github.com/migtools/velero-plugin-for-vsm)
22
21
23
22
Note that only one of `AWS` and `Legacy AWS` may be installed at the same time. `Legacy AWS` is intended for use with certain S3 providers that do not support the V2 AWS SDK APIs used in the `AWS` plugin.
Copy file name to clipboardExpand all lines: docs/credentials.md
-20Lines changed: 0 additions & 20 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -10,7 +10,6 @@
10
10
1.[BSL and VSL share credentials for one provider](#backupstoragelocation-and-volumesnapshotlocation-share-credentials-for-one-provider)
11
11
2.[BSL and VSL use the same provider but use different credentials](#backupstoragelocation-and-volumesnapshotlocation-use-the-same-provider-but-use-different-credentials)
12
12
3.[No BSL specified but the plugin for the provider exists](#no-backupstoragelocation-specified-but-the-plugin-for-the-provider-exists)
13
-
5.[Creating a Secret: OADP with VolumeSnapshotMover](#creating-a-secret-for-volumesnapshotmover)
14
13
15
14
### Creating a Secret for OADP
16
15
@@ -214,22 +213,3 @@ spec:
214
213
If you don't need volumesnapshotlocation, you will not need to create a VSL credentials.
215
214
216
215
If you need `VolumeSnapshotLocation`, regardless of the `noDefaultBackupLocation` setting, you will need a to create VSL credentials.
217
-
218
-
219
-
### Creating a Secret for volumeSnapshotMover (OADP 1.2 or below)
220
-
221
-
VolumeSnapshotMover requires a restic secret. It can be configured as so:
222
-
223
-
```
224
-
apiVersion: v1
225
-
kind: Secret
226
-
metadata:
227
-
name: <secret-name>
228
-
type: Opaque
229
-
stringData:
230
-
# The repository encryption key
231
-
RESTIC_PASSWORD: my-secure-restic-password
232
-
```
233
-
234
-
- *Note:* `dpa.spec.features.dataMover.credentialName` must match the name of the secret.
235
-
Otherwise it will default to the name `dm-credential`.
0 commit comments