0

We have multiple baremetal servers part of our dockers and Openshfit(kubernetes) cluster. For some reason the underlying pods are extremely slow only with BM Nodes, the traditional VMs hosted on exsi servers work flawless. the pods take up very long to come up at all times liveness probes fail often. The BM nodes have 72 cores and 600 GB RAM and 2 n/w ports & are underutilised say Load Average just about 10 ~ 20 and Free RAM over 300 ~ 400 Gigis at all times. sar output looks normal, /var/log/messages have nothing unusual. Not able to nail down what's causing the slowness..

Is there a linux/docker command that will help here & what do i look for? Could this be a noisy neighbour problem? or do I need to tweak some Kernel Parameter(s). The slowness is always there, it's not intermittent. We have closely worked with RH support and got nothing from that exercise. Any suggestions welcome..

Rohit C
  • 19
  • 4
  • 1
    Does the pod work ok after startup? That is, slowness is only during startup, or does it continue to be slow even after the pod is up? – Burak Serdar Nov 14 '19 at 18:07
  • pod works after startup (intermittently killed by OpenShift as liveness probes fail) - but the application hosted is always slow as compared to traditional VMs in the same cluster – Rohit C Nov 14 '19 at 18:11
  • Did you measure how those hosts run non-docker workloads? That is, is the problem related to docker? – Burak Serdar Nov 14 '19 at 18:18
  • simple find, grep commands seem to respond in par with the other non BM servers... I have a feel docker is slowing things down .. – Rohit C Nov 14 '19 at 18:25
  • I suppose one thing to try is to run some docker workloads (without openshift) on those hosts in isolation, to see if it is docker, or OS causing problems. – Burak Serdar Nov 14 '19 at 18:30
  • Do you use a different network for your containers? Or do containers share IP adresses with your host network? This was the cause for same behavior in one of my projects. – Michael Johann Nov 14 '19 at 18:42
  • @BurakSerdar, let me try that... Regarding IP addresses, we don't directly manage that; Openshift uses vxlan_4789 which manages the n/w – Rohit C Nov 15 '19 at 08:37

0 Answers0