cortexlabs/cortex

Issues

• Persist Prometheus metrics to S3 instead of a persistent volume

  #2321 opened 3 years ago
  1

• Revisit API status

  #2320 opened 3 years ago
  0

• Revisit `cortex get` output in relation to grafana

  #2319 opened 3 years ago
  0

• Add instance count dashboard

  #2318 opened 3 years ago
  0

• Add docs for production / large scale clusters

  #2317 opened 3 years ago
  0

• Fix istio request rate discrepancy

  #2316 opened 3 years ago
  0

• Investigate why pods get stuck in a terminating state

  #2315 opened 3 years ago
  0

• Investigate unresponsive EKS control plane at scale

  #2314 opened 3 years ago
  0

• Support updating `iam_policy_arns` of an existing cluster

  #2312 opened 3 years ago
  1

• Raise cluster autoscaler resource limits

  #2297 opened 3 years ago
  0

• Forward gateway request headers to AsyncAPI

  #2296 opened 3 years ago
  0

• Support for AWS virtual GPU plugin for GPU sharing

  #2295 opened 3 years ago
  3

• GCP support removed?

  #2294 opened 3 years ago
  1

• Properly validate instance quota on `cortex cluster configure`

  #2293 opened 3 years ago
  0

• Investigate 502 responses under high load

  #2290 opened 3 years ago
  1

• Prometheus fails at large cluster size

  #2285 opened 3 years ago
  0

• Discuss the API spec wrapping k8s spec and nodegroup management

  #2283 opened 3 years ago
  0

• Investigate other cluster config fields that can be updated

  #2282 opened 3 years ago
  0

• Add queue length to `cx get`

  #2281 opened 3 years ago
  0

• Move queue metrics from operator to dequeuer

  #2280 opened 3 years ago
  0

• Create a shared async gateway deployment which runs on the operator node group

  #2279 opened 3 years ago
  0

• Allow operator nodegroup to scale to 25

  #2278 opened 3 years ago
  0

• Node group is un-schedulable if it has had a history of it failing to be added (even if a subsequent addition succeeded)

  #2257 opened 3 years ago
  0

• Support updating the priority of an existing node group

  #2254 opened 3 years ago
  0

• Philosophy and goals of the project going forward?

  #2241 opened 3 years ago
  6

• Change async SQS type from FIFO to Standard

  #2240 opened 3 years ago
  0

• Block query params for AsyncAPI

  #2238 opened 3 years ago
  0

• Cortex refresh for AsyncAPI

  #2237 opened 3 years ago
  0

• Check if request is abandoned in the proxy before forwarding to user containers

  #2236 opened 3 years ago
  1

• SSE for S3

  #2235 opened 3 years ago
  0

• AttributeError: 'list' object has no attribute 'get'

  #2223 opened 3 years ago
  2

• Encrypt messages in SQS queue

  #2222 opened 3 years ago
  0

• Death dep is not honoured by kubexit if the container died before kubexit started

  #2217 opened 3 years ago
  2

• Task job gets set to `"unexpected_error"` status sometimes

  #2215 opened 3 years ago
  0

• Submitting task jobs can sometimes fail

  #2214 opened 3 years ago
  1

• Enable the user to configure SMTP for Grafana (to enable email alerts)

  #2210 opened 3 years ago
  0

• Support max_concurrency for async APIs

  #2200 opened 3 years ago
  0

• Support autoscaling async APIs to 0 replicas

  #2199 opened 3 years ago
  1

• ImportError: cannot import name 'UJSONResponse'

  #2193 opened 3 years ago
  7

• Convert Grafana and Prometheus volumes to gp3

  #2188 opened 4 years ago
  0

• Support spot instance capacity rebalancing

  #2186 opened 4 years ago
  0

• Support capacity-optimized spot allocation strategy

  #2185 opened 4 years ago
  0

• Address eksctl warning regarding custom AMIs

  #2184 opened 4 years ago
  0

• Consider adding in-cluster image caching layer

  #2182 opened 4 years ago
  4

• Submitting a batch job with file_path_lister is failing/timing out

  #2178 opened 4 years ago
  0

• implement Proxy Readiness Probes

  #2175 opened 4 years ago
  0

• Support mounting EFS volumes

  #2160 opened 4 years ago
  0

• Consider using NLB IP mode

  #2159 opened 4 years ago
  0

• Pass the cortex_logger as a field into the handler constructor

  #2153 opened 3 years ago
  0

• Support `cortex prepare-debug` for the TensorFlow handler

  #2148 opened 3 years ago
  0