intelsdi-x / kubesnap Goto Github PK

This is the place to discuss any aspects regarding demo flow. All decision will be recorded here: https://github.com/intelsdi-x/kubesnap/wiki/Demo-flow

I enjoyed your recent post on Medium The benefits of using Snap with Kubernetes, but I'm left wondering what the differences are with the other popular Kubernetes metrics framework: prometheus.io?

When using workload (src/workload) which is running stress on dedicated number of cores metrics for CPU utilization seems to be incorrect. For a container with CPU limit of 5k milicores and request of 1k milicores when stress is run on 2 cores (2k milicores) it's seen that metrics for CPU utilization have strange values (like negative values, or very high values - like 15k milicores). This causes horizontal pod autoscaler to behave incorrectly - logic seems right but the data are causing unnecessary pods creation or deletion.

This is the place to discuss any aspects regarding work plan and work status. All decision will be recorded here: https://github.com/intelsdi-x/kubesnap/wiki/Work-Plan

This is the place to discuss any aspects regarding PoC design. All decision will be recorded here: https://github.com/intelsdi-x/kubesnap/wiki/Design

What is the difference between network stats under "network" and those under "network"->"interfaces"? Why this one interface is reported in section "network"?

 "network": {
  "name": "enp4s0f1",
  "rx_bytes": 12124592136,
  "rx_packets": 21675552,
  "rx_errors": 0,
  "rx_dropped": 0,
  "tx_bytes": 3935774766,
  "tx_packets": 5334828,
  "tx_errors": 0,
  "tx_dropped": 0,
  "interfaces": [
   {
    "name": "enp4s0f1",
    "rx_bytes": 12124592136,
    "rx_packets": 21675552,
    "rx_errors": 0,
    "rx_dropped": 0,
    "tx_bytes": 3935774766,
    "tx_packets": 5334828,
    "tx_errors": 0,
    "tx_dropped": 0
   },
   {
    "name": "flannel.1",
    "rx_bytes": 6263507721,
    "rx_packets": 4596302,
    "rx_errors": 0,
    "rx_dropped": 0,
    "tx_bytes": 2353793829,
    "tx_packets": 608371,
    "tx_errors": 0,
    "tx_dropped": 8
   },
   {
    "name": "eth2",
    "rx_bytes": 0,
    "rx_packets": 0,
    "rx_errors": 0,
    "rx_dropped": 0,
    "tx_bytes": 0,
    "tx_packets": 0,
    "tx_errors": 0,
    "tx_dropped": 0
   },