-
# Summary
The calleido in GKE gave excessive authority when defining Service Account named "calleido-cert-manager-cainjector". Besides, this Service Account is mounted in a Pod named "calleido-cert…
-
### 🚀 The feature, motivation and pitch
For my particular use case, I need the dataset to perform some operations when the data loaders workers are terminated. When the `iteration_end` value in the…
-
Expected Outcome:
pod/weave-net-lxzfx 2/2 Running 7 (26h ago) 3d3h
Actual Outcome:
cyx@node2:/$ kubectl get all -n kube-system
NAME READY STATUS …
-
We are using Longhorn in our cluster with around 50 PVCs. When the cluster is idle at night, the Longhorn CPU usage still sits around 2-4 cores constantly.
This is mainly caused by the longhorn-ins…
-
**Describe the bug:**
**Media Links Used:**
**Expected behavior**
**Screenshots:**
**StackTrace:**
```
Paste Stacktrace here if available
```
**Device Info (please compl…
-
## Description
Provide a way for end users to consume and be charged for a pre-defined set of hyperscaler resources:
- specialized node types. For example GPU and ARM nodes, network, memory and CPU …
-
**Tell us about your request**
Add the ability to get information about the docker swarm when requested through the Docker Engine API to workers
**Which service(s) is this request for?**
Docker S…
-
Following on from discussions in dask/distributed#3300 it would be great if workers gracefull closed when they recieve a spot termination notification. This will be even more useful with #47.
- `EC…
-
Currently the device type is hard-coded. When a worker joins, it needs to have a device type assigned to it so that work can be dispatched.
-
### What happened + What you expected to happen
Script
---
```python
import os
import ray
os.sched_setaffinity(os.getpid(), set(range(10, os.cpu_count())))
ray.init(include_dashboard=Fals…