[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: high tcp retransmissions between pods on same openshift node



We did make changes to the version of openvswitch that we were using. What is the underlying infrastructure?

---

ERIK JACOBS

PRINCIPAL TECHNICAL MARKETING MANAGER, OPENSHIFT

Red Hat Inc

ejacobs redhat com    M: 646.462.3745     @: erikonopen


On Tue, Apr 18, 2017 at 5:06 AM, Roosemeyers Erik <erik roosemeyers arxus eu> wrote:

Hi Erik,

 

Thx for picking up my questions…

 

We’re using OCP v3.1 and the openshift-sdn.

 

Both IP and DNS name of containers are used in the tests.

 

If it can help, I can include more detailed test results in various scenarios.

 

Do you think a newer version of ocp will improve on this ? thx !

 

Kind regards,

Erik

 

From: Erik Jacobs [mailto:ejacobs redhat com]
Sent: maandag 10 april 2017 22:45
To: Roosemeyers Erik <erik roosemeyers arxus eu>
Cc: dev lists openshift redhat com
Subject: Re: high tcp retransmissions between pods on same openshift node

 

Hi Erik,

 

Are you using openshift-sdn or multitenant-sdn or something else? Are you running OpenShift Container Platform 3.1?

 

Are you just communicating with the pod IP directly?


---

ERIK JACOBS

PRINCIPAL TECHNICAL MARKETING MANAGER, OPENSHIFT

Red Hat Inc

ejacobs redhat com    M: 646.462.3745     @: erikonopen

 

On Tue, Mar 28, 2017 at 4:27 AM, Roosemeyers Erik <erik roosemeyers arxus eu> wrote:

Hi all,

 

I'm testing network connectivity (speed and reliability) on a OpenShift cluster 3.1.

 

I do this with 2 pods, each based on iperf3 Docker image.

 

First pod starts iperf3 as server, other one as iperf3 client with: iperf3 -c iperf3 -t 120 (iperf3 name of server pod).

 

What I notice is high throughput but also big retransmission rates:

eg.

duration              Transfer              Bandwidth         Retransmissions (TCP)

0.00-120.00 sec 214 GBytes         15.3 Gbits/sec   170478

 

If the pods are on 2 different nodes, the retransmission rate is smaller (as is throughput):

 

duration              Transfer              Bandwidth         Retransmissions (TCP)

0.00-120.00 sec 22.6 GBytes       1.62 Gbits/sec   3627

 

What I was wondering - what causes this higher retransmission rates ? (Mostly, it's a sign the are network problems but is it ?)

 

(I did some other tests, like deploying the containers via the Docker daemon onto 1 server. And even on a local laptop with only Docker daemon installed - and I always saw these tcp retransmission rates...)

 

 

Thanks in advance !

 

kind regards,

Erik


_______________________________________________
dev mailing list
dev lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/dev

 



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]