node-feature-discovery
node-feature-discovery copied to clipboard
kubernetes-sigs
Node feature discovery for Kubernetes
adds command line flag to configure pagination for nfd-gc for issue: https://github.com/kubernetes-sigs/node-feature-discovery/issues/1998
**What would you like to be added**: GC running using gocron within master pod and leveraging it's informer cache **Why is this needed**: reduce memory overhead from the additional pod...
**What happened**: `nfd-topology-updater` is used for report node's NUMA and `Guaranteed` pods's cpu used. But it missed cpu for `Guaranteed` pod which pod's have any no exclusiveCPU. 1. Node `yj-kubevirtwork-001`...
**What would you like to be added**: add x86-64 microarchitecture level label and blacklist all other cpuid features by default except for cpuid features newer than x86-64v4. **Why is this...
NFD worker POD logs show log entry below: ``` failed to get self pod, cannot inherit ownerReference for NodeFeature. Get https://10.43.0.1:443/api/v1/namespaces/gpu-operator/pods/gpu-operator-node-feature-discovery-worker-xxxx. dial tcp 10.43.0.1:443 I/O timeout ``` The machine in...
NFD worker POD logs show log entry trying to read or open a file for attributes ```operstate``` and ```speed```. I see that ```/host-sys/class/net/bonding_masters``` is a file and NOT a directory....
**What would you like to be added**: With Helm v3.8.0, the OCI support became [GA](https://helm.sh/docs/topics/registries), which is an excellent chance to start publishing Helm charts to OCI-compliant registries. Quay /...
**What would you like to be added**: I think the default/out-of-box behavior when NFD is installed should be that the `NodeFeature` CRs should have their owner reference set to `v1.Node`...
**What happened**: I have installed the 0.17.3 version of nfd using helm. I want to get the numa node topology, so I enabled the topology updater while installing. But numa...
