Commit Graph

5 Commits

Author SHA1 Message Date
Kevin Klues
74d634a028 Update kubelet support for recent changes to resource.k8s.io/v1alpha2
Signed-off-by: Kevin Klues <kklues@nvidia.com>
2023-03-14 22:34:18 +00:00
Moshe Levi
9c57613912 Add ClassName to chekpoint state and in-memory cache
Signed-off-by: Moshe Levi <moshele@nvidia.com>
2023-03-14 19:33:04 +02:00
Kevin Klues
fd7370b84d Update DRAManager checkpoint to store a map for CDIDevices
The key of the map is the KubeletPluginName where the CDIDevices originate.

Signed-off-by: Kevin Klues <kklues@nvidia.com>
2023-03-13 12:52:41 +00:00
Kevin Klues
273a8ffad1 Rename CdiDevices to CDIDevices in dramanager checkpoint
Signed-off-by: Kevin Klues <kklues@nvidia.com>
2023-03-13 12:52:41 +00:00
Moshe Levi
e7256e08d3 kubelet dra: add checkpointing mechanism in the DRA Manager
The checkpointing mechanism will repopulate DRA Manager in-memory cache on kubelet restart.
This will ensure that the information needed by the PodResources API is available across
a kubelet restart.

The ClaimInfoState struct represent the DRA Manager in-memory cache state in checkpoint.
It is embedd in the ClaimInfo which also include the annotation field. The separation between
the in-memory cache and the cache state in the checkpoint is so we won't be tied to the in-memory
cache struct which may change in the future. In the ClaimInfoState we save the minimal required fields
to restore the in-memory cache.

Signed-off-by: Moshe Levi <moshele@nvidia.com>
2023-03-10 12:22:15 +02:00