Re: [kubernetes-users] Re: Running Calico on GKE

Aside: kube-flannel takes care of dropping the CNI plugin on the root filesystem in this way today . But, yes, the kubelet needs to be in CNI mode first. On Wed, Nov 30, 2016 at 10:13 AM 'Tim Hockin' via Kubernetes u

Re: [kubernetes-users] Have anyone successfully registered more than 120 nodes with Kube-master 1.4.x?

On Wednesday, November 30, 2016 at 3:50:31 PM UTC-8, Brandon Philips wrote: > ulimits are applied to processes not users. That process has a 4096 file > limit. > > > How are you running the API server? Systemd? Docker container? Can you > provide a gist of what you are doing so I can suggest a

Re: [kubernetes-users] Have anyone successfully registered more than 120 nodes with Kube-master 1.4.x?

ulimits are applied to processes not users. That process has a 4096 file limit. How are you running the API server? Systemd? Docker container? Can you provide a gist of what you are doing so I can suggest a fix? On Wed, Nov 30, 2016 at 1:22 PM wrote: > On Tuesday, November 29, 2016 at 5:30:53 P

[kubernetes-users] Does kube-apiserver set the limitation for "open files" with value 1024 for 'root'?

Here is why? root@ovrh0301:/home/envi# ulimit -a core file size (blocks, -c) 0 data seg size (kbytes, -d) unlimited scheduling priority (-e) 0 file size (blocks, -f) unlimited pending signals (-i) 772117 max locked memory (kbytes, -

[kubernetes-users] Re: Bare Metal SIG

On Wed, Nov 30, 2016 at 12:09 AM, Sandeep Srinivasa wrote: > precisely. and i would argue that bare metal poses an even bigger > platform-specific challenge. How do you build a cluster without depending > on VPC/ELB/EBS, etc. > We do this on bare metal all the time, I'll be at the cluster-ops m

Re: [kubernetes-users] Have anyone successfully registered more than 120 nodes with Kube-master 1.4.x?

On Tuesday, November 29, 2016 at 5:30:53 PM UTC-8, Brandon Philips wrote: > How are you running the API server? If you are running it in a container your > shell ulimits won't match. Find the API server PID and do: > > > cat /proc/$PID/limits to check > > > On Tue, Nov 29, 2016 at 5:11 PM wro

Re: [kubernetes-users] Re: Running Calico on GKE

The trick is that it has to drop files in the root filesystem, configure a kubelet flag, and restart kubelet. There's really no way to do that from daemonset. On Wed, Nov 30, 2016 at 2:15 AM, mxmxmx wrote: > Thanks for your answer. > Thats mainly why we were sceptical about it. > Do you think it

[kubernetes-users] Persistent Volumes and failover - how does it work?

I have been going through documentation but it does not really go into depth for me to understand how does persistent volumes work with pod/node failovers. Let me explain the questions I am having Let's say I have a cluster on EC2 with 2 minions. Since a single EBS volume can be mounted only on on

Re: [kubernetes-users] Health metrics in Heapster?

Hi Steve, Your use case is out of the scope of Heapster project, which is about gathering system metrics (like resource usage) rather than a part of alerting pipeline. Added a few folks from Stackdriver which can elaborate about potential options with alerts there. Fabian (in cc) should be able to

Re: [kubernetes-users] MYSQL Pod connection

sorry for your trouble I already solved it, it was a mysql privilege error preventing the connection and fix it thanks On Wednesday, November 30, 2016 at 10:48:03 AM UTC-5, David Aronchick wrote: > > Can you paste in your deployment and service definitions? > On Wed, Nov 30, 2016 at 07:14 Ioanni

Re: [kubernetes-users] Re: [GRAYMAIL] Re: Bare Metal SIG

I think SIG-On-Prem makes a lot of sense! +1 from me on that name. Will wait another day for feedback before getting it created hopefully with Sarah's help. Thanks, JJ. > On Nov 30, 2016, at 10:37 AM, Isaac Arias wrote: > > sig-datacenter > > From: Sandeep Srinivasa > Date: Wednesday, Nov

Re: [kubernetes-users] MYSQL Pod connection

Can you paste in your deployment and service definitions? On Wed, Nov 30, 2016 at 07:14 Ioannis Canellos wrote: > An example of how you can use mysql with deployment, service and more can > be found here: > https://github.com/fabric8io/kubernetes-zipkin/tree/master/mysql/src/main/fabric8 > > On W

[kubernetes-users] Re: [GRAYMAIL] Re: Bare Metal SIG

sig-datacenter From: Sandeep Srinivasa Date: Wednesday, November 30, 2016 at 7:33 AM To: Tomasz 'Zen' Napierala Cc: Rakesh Malhotra , Kubernetes user discussion and Q&A , Kubernetes developer/contributor discussion , Justin Garrison , Isaac Arias Subject: [GRAYMAIL] Re: Bare Metal SIG I agr

[kubernetes-users] Re: Bare Metal SIG

I agree. sig-on-prem is best suited. On Nov 30, 2016 20:53, "Tomasz 'Zen' Napierala" wrote: I also agree that sig-on-prem better describes the goal. > On 30 Nov 2016, at 16:06, Justin Garrison wrote: > > If sig-aws has a reason to exist, then so does sig-metal. People who use the cloud have

Re: [kubernetes-users] Health metrics in Heapster?

+Filip +Piotr On Wed, Nov 30, 2016 at 5:31 AM, 'Steve Wolter' via Kubernetes user discussion and Q&A wrote: > Hi kubernetes-users, > > I have the following problem: I need a dashboard and alerts for the health > of my Kubernetes Deployment, for example the # of healthy/unhealthy pods in > the re

[kubernetes-users] Re: Bare Metal SIG

I also agree that sig-on-prem better describes the goal. > On 30 Nov 2016, at 16:06, Justin Garrison > wrote: > > If sig-aws has a reason to exist, then so does sig-metal. People who use the > cloud have (justifiably) no interest in mechanics that will be mandatory in > bare metal - like

Re: [kubernetes-users] MYSQL Pod connection

An example of how you can use mysql with deployment, service and more can be found here: https://github.com/fabric8io/kubernetes-zipkin/tree/master/mysql/src/main/fabric8 On Wed, Nov 30, 2016 at 5:06 PM, Montassar Dridi wrote: > Usually I create one mysql pod and connect it to my java applicatio

[kubernetes-users] Re: Bare Metal SIG

> > If sig-aws has a reason to exist, then so does sig-metal. People who use > the cloud have (justifiably) no interest in mechanics that will be > mandatory in bare metal - like choice of load balancers, networking across > data centers, etc > With that being said, I don't think this should

[kubernetes-users] MYSQL Pod connection

Usually I create one mysql pod and connect it to my java application pod buy getting the sql pod ip address and replace my db_uri as follow jdbc:mysql://MYSQL-POD-IP:3306/ THEN THE CONNECTION WILL BE RUNNING FINE WITH NO ISSUE. but then I tried to create MYSQL pod using Deployments and replica

[kubernetes-users] Health metrics in Heapster?

Hi kubernetes-users, I have the following problem: I need a dashboard and alerts for the health of my Kubernetes Deployment, for example the # of healthy/unhealthy pods in the replica sets of the deployment or the observed and actual generation of the deployment. I'm running half a dozen clusters

[kubernetes-users] Binaries in beta build

Hi, I started using kubernetes some weeks ago and think it is great software. Part of the reason is that kubernetes development is driven by so many great people at an incredible pace. I wanted to test the new beta version of the 1.5.0 release. After downloading the package from the github

Re: [kubernetes-users] Current status of GPU support in Kubernetes

Great! I am working on rebase my pr(https://github.com/kubernetes/kubernetes/pull/28216 ). Regards, Hui-Zhi Zhao > On 30 Nov 2016, at 6:28 PM, Vishnu Kannan wrote: > > this -- You received this message because you are subscribed to the Go

Re: [kubernetes-users] Current status of GPU support in Kubernetes

Kubernetes supports only 1 GPU as of now. Take a look at this doc . This PR is attempting to add support for multiple GPUs. We hope to introduce alpha suppor

Re: [kubernetes-users] Re: Running Calico on GKE

Thanks for your answer. Thats mainly why we were sceptical about it. Do you think it could be done using a DaemonSet, or is there other main issues to consider ? Is there some public discussion to follow on this topic somewhere around ? Le mardi 29 novembre 2016 01:17:28 UTC+1, Tim Hockin a écri

[kubernetes-users] Re: Running Calico on GKE

Glad to read it ;) I will join calico's slack to try to catch someone who tested it. Le mardi 29 novembre 2016 00:56:46 UTC+1, Christopher Liljenstolpe a écrit : > > Greetings, > > It is possible to enable Calico on GKE. I haven't personally done it, but > we have tested it here. You need to en

[kubernetes-users] Current status of GPU support in Kubernetes

Hi everyone, We are considering using Kubernetes to manage batch jobs on an on-prem GPU cluster. However, I had a hard time looking for updated information about GPU support in Kubernetes. Here it said only 1 GPU per machine is supported: https://github.com/kubernetes/kubernetes/pull/24836 Is