March 2019 - openSUSE Kubic - openSUSE Mailing Lists

[opensuse-kubic] MicroOS custom image build
by Tardif, Christian 04 Apr '19

04 Apr '19

Hi, We're starting to look at MicroOS as the base OS for our Kubernetes stacks. And as we're now using RKE to deploy k8s, we're stuck with Docker right now. Added to this, we have other things to customize in order to ease the OS deployment (mostly targeting, but not limited to bare metal deployment) Two questions here: 1) Can I (I assume yes) leverage kiwi to build a raw image (as our metal deployer requires raw image)? If so, I haven't found how to do it in the available docs 2) Providing that everything is automated and standardized, using cloud-init NoCloud datasource is not an option (as it's unlikely we'll be having a connected unpartitioned disk to hold the cloud-init configuration), and I haven't been able to make the LocalDisk datasource (again, I most probably just haven't the docs required to make it work). Thanks! Christian -- To unsubscribe, e-mail: opensuse-kubic+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-kubic+owner(a)opensuse.org

4 32

[opensuse-kubic] High IO activity for master node
by Robert Munteanu 27 Mar '19

27 Mar '19

Hi, I have taken kubic for a spin with a local QEMU/KVM setup - 1 master node and 3 worker nodes. The cluster is up and running (ingress, load balancer, storage, monitoring ). It is basically stale - except the management pods nothing is deployed. I'm using the qcow2 files from [1], and launching the machines with qemu-system-x86_64 \ -machine type=pc,accel=kvm \ -cpu host \ -smp $cpus \ -drive file=$disk,if=virtio,l2-cache-size=3145728 -cdrom $cdrom \ -m 1024 \ -netdev bridge,id=hostnet0 \ -device e1000,netdev=hostnet0,mac=$mac \ -nographic $cpus is 2 for the master node and 1 for the others. I have one recurring problem with the master becoming unresponsive after some time. Looking at the grafana charts I can see that there is steadily increasing Disk I/O for the master nodee. At about ~30 minutes after launching all 4 nodes I see the load average is 6 on the master node, with disk read I/O increasing steadily. The 5 minute average as collected by prometheus is at about 350 MB/s. The situation seems to be "fixed" when the kube-scheduler-master and kube-controller-manager-master crash and are not scheduled. Looking at those pods' logs does not show match, they seem to exit because of a failed leader election, e.g. for kube-controller-manager-master I0326 15:59:39.521403 1 leaderelection.go:249] failed to renew lease kube-system/kube-controller-manager: failed to tryAcquireOrRenew context deadline exceeded for kube-scheduler-master I0326 15:59:39.589619 1 leaderelection.go:249] failed to renew lease kube-system/kube-scheduler: failed to tryAcquireOrRenew context deadline exceeded And looking at the I/O grafs it seems that there is a cycle - cluster is ok - IO increases - kube-*-master pods fail - IO decreases - kube-*-master pods recover Right now I'm at a loss about where to look next. Does anyone have any ideas about what might cause this or how to better debug this kind of situation? Thanks, Robert [1]: https://download.opensuse.org/repositories/devel:/kubic:/images/openSUSE_Tu… -- To unsubscribe, e-mail: opensuse-kubic+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-kubic+owner(a)opensuse.org

4 8

[opensuse-kubic] AutoYast2 issue
by Tardif, Christian 26 Mar '19

26 Mar '19

Hi, I'm presently experimenting with AutoYast deployment for MicroOS and I'm facing a strange issue. Did my installation the usual way on a machine, specifying NIC bonding, vlans, and bridges as needed, then created required users, ... all the funny things you don't want to repeat over and over. Then installed autoyast2 software, and then did the yast2 clone_system. It created the autoinst.yml which was looking OK to me when I validated it. Then tried the reinstall the same machine with the yml file on a usb stick and... it didn't work. Complaining that the autoyast* packages weren't available (missing repos ?) then complained that it couldn't mount volumes on a read-only filesystem. then stopped. Is there any other steps I need to perform before installing (or modifying the autoyast file to include some other stuff) ? Thanks, -- Christian Tardif -- To unsubscribe, e-mail: opensuse-kubic+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-kubic+owner(a)opensuse.org

2 1

[opensuse-kubic] manifestes for kubernetes workload?
by Thorsten Kukuk 24 Mar '19

24 Mar '19

Hi, most of the time, pods/services/deployments for kubernetes have their source on github.com, the containers somewhere in a registry and you can deploy them using a yaml manifest on github.com, e.g. kubectl apply -f https://raw.githubusercontent.com/paulbouwer/hello-kubernetes/master/yaml/h… Now we have already packaged some stuff, build a RPM from it in our buildservice, build a container from it in our buildservice, publish the container in our registry. How do we want to provide the manifest? We cannot point to OBS or an RPM. So my idea was, to build a sub-package, which contains the manifest with the adjusted image url pointing to our registry, install them by default on kubic and the user can use that to deploy the containers. I only would like to specify this, so that not everybody is using something else. 1. I think it should be easy to identify the RPM, so e.g. something like hello-kubernetes-<???>.rpm Which suffix could we use for this? 2. What would be a good location in the filesystem? /usr/share/defaults/manifests? One directory for everything? Every RPM having it's own directory? Any ideas or opinions? Any better ideas? Thorsten -- Thorsten Kukuk, Distinguished Engineer, Senior Architect SLES & MicroOS SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany GF: Felix Imendoerffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nuernberg) -- To unsubscribe, e-mail: opensuse-kubic+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-kubic+owner(a)opensuse.org

4 7

Re: [opensuse-kubic] root login on Hyper-V image*s* of MicroOS KO
by Thorsten Kukuk 01 Mar '19

01 Mar '19

Hi, On Thu, Feb 28, Jimmy PIERRE wrote: > On Thu, 28 Feb 2019 at 12:51, Thorsten Kukuk <kukuk(a)suse.de> wrote: > > That's described on the Kubic:Installation page: > https://en.opensuse.org/Kubic:Installation# > Images_for_x86_64_virtualisation_environments > > You need to set a password or add your ssh keys with cloud-init. > An example cloud-init config file can be found here: > https://en.opensuse.org/Kubic:MicroOS/cloud-init# > Example_configuration_for_openSUSE_Micro > Thanks for stepping in! I undersrand. If I create my own vm with the iso, is > there a reason why the images would be better (size, tuning, updates etc.)? If you create install your virtual machine or bare metal hardware with YaST, you are more flexible what and how to install and you don't need to fiddle around with cloud-init to be able to login. If you use the images, they are preconfigured, so faster to deploy, but you are less flexible and you need to play with cloud-init. If you use KVM or VMware, it shouldn't matter what you use, but since cloud-init is really slow, I would install with YaST and not install cloud-init. On OpenStack I would use the image. > And the documentation to survive do exist please? Which documentation are you missing? There is the Kubic portal in the openSUSE wiki and were you can find many documentation (only look at the above examples), enough to setup a kubernetes cluster at your own. Thorsten -- Thorsten Kukuk, Distinguished Engineer, Senior Architect SLES & MicroOS SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany GF: Felix Imendoerffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nuernberg) -- To unsubscribe, e-mail: opensuse-kubic+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-kubic+owner(a)opensuse.org

3 3