Guarav, Hi, some more points – and I hope your demo went well and thanks again for raising all the issues that helped us fix up the Amsterdam branh.
1) Kubectl 1.7.0 vs supported 1.8.4 – I suspect this is your issue. I last ran the following bootstrap script at noon on a lean machine and after running cd.sh had a system up after the prepull in 23 min with no issues except for the known aaf container. https://github.com/obrienlabs/onap-root/blob/master/oom_rancher_setup_1.sh Your issue may be the version of kubectl – our currently stable release config is as follows on the page https://wiki.onap.org/display/DW/ONAP+on+Kubernetes#ONAPonKubernetes-QuickstartInstallation Rancher v1.6.10 (not the stable or latest tag) Helm 2.3.0 on the server and client Kubectl 1.8.4 (current stable version) – yours is 1.7.0 (I have seen 1.8.8 running) You should be running curl -LO https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/linux/amd64/kubectl 2) Mixing raw dockers with Kubernetes managed containers You will need a JIRA to add the Kubernetes infrastructure for AAI ESR – I don’t see the containers in HEAT either (aai1/aai2) As Alexis stated: If you need a new component – it must have a yaml set and deleted/created via the scripts or directly with helm – in HELM_APPS – esr is not there. Mixing raw docker run commands or docker-compose up – will be outside the “onap” namespace setup by kubernetes – they can run independently and be accessed - but they will be out of the 30xxx namespace – as in not managed by Kubernetes. ESR would need a config set, a set of template yamls, an all-services.yaml to set the external port mappings, a Chart.yaml for helm and a values.yaml to list the docker images and their version tags. As I remember ESR (External System Registrar) is part of AAI portal onboarding – I did see these two images in the older manifest but currently the 9 container AAI pod does not have a definition for ESR. https://gerrit.onap.org/r/gitweb?p=integration.git;a=blob;f=version-manifest/src/main/resources/docker-manifest.csv;h=35e992adc0678ccd98328d33c9a5ffc88dbb4dfa;hb=refs/heads/master For a reference Amsterdam install to check on – see the live CD server And the healthcheck references (Thanks to Shane Daniel in adding this dashboard to the elk stack) http://jenkins.onap.info/job/oom-cd/606/console http://kibana.onap.info:5601/app/kibana#/dashboard/AWAtvpS63NTXK5mX2kuS?_g=(refreshInterval:(display:Off,pause:!f,value:0),time:(from:now-24h,mode:quick,to:now))&_a=(description:'',filters:!(),options:(darkTheme:!f),panels:!((col:1,id:AWAts77k3NTXK5mX2kuM,panelIndex:1,row:1,size_x:8,size_y:3,type:visualization),(col:9,id:AWAtuTVI3NTXK5mX2kuP,panelIndex:2,row:1,size_x:4,size_y:3,type:visualization),(col:1,id:AWAtuBTY3NTXK5mX2kuO,panelIndex:3,row:7,size_x:6,size_y:3,type:visualization),(col:1,id:AWAttmqB3NTXK5mX2kuN,panelIndex:4,row:4,size_x:6,size_y:3,type:visualization),(col:7,id:AWAtvHtY3NTXK5mX2kuR,panelIndex:6,row:4,size_x:6,size_y:6,type:visualization)),query:(match_all:()),timeRestore:!f,title:'CD%20Health%20Check',uiState:(),viewMode:view) thank you /michael From: Alexis de Talhouët [mailto:[email protected]] Sent: Wednesday, December 6, 2017 11:10 To: Gaurav Gupta (c) <[email protected]> Cc: [email protected]; Michael O'Brien <[email protected]>; Ramesh Tammana <[email protected]>; Ramki Krishnan <[email protected]>; Arun Arora (c) <[email protected]> Subject: Re: Issues with Kubectl setup and newly pulled dockers not showing in kubectl list ---namespace I see, so for point 1 I cannot help much. It worked well for me. For the second point, do kubernetes manifests exist for esr and esr gui? You need the deployment and the service kubernetes manifests as explained in my previous mail. Thanks, Alexis On Dec 6, 2017, at 10:57 AM, Gaurav Gupta (c) <[email protected]<mailto:[email protected]>> wrote: thanks Alexis response in line . with best regards gaurav ________________________________ From: Alexis de Talhouët <[email protected]<mailto:[email protected]>> Sent: 06 December 2017 19:57 To: Gaurav Gupta (c) Cc: [email protected]<mailto:[email protected]>; Michael O'Brien; Ramesh Tammana; Ramki Krishnan; Arun Arora (c) Subject: Re: Issues with Kubectl setup and newly pulled dockers not showing in kubectl list ---namespace Hi, Some answers inline. Thanks, Alexis On Dec 6, 2017, at 8:34 AM, Gaurav Gupta (c) <[email protected]<mailto:[email protected]>> wrote: Hi All , Michal , Alexis We are into 2 very distinct issues while trying vDNS Demo on OOM . issue 1 - kubectl post following below information . 1. curl -LO https://storage.googleapis.com/kubernetes-release/release/v1.7.0/bin/linux/amd64/kubectl<https://urldefense.proofpoint.com/v2/url?u=https-3A__storage.googleapis.com_kubernetes-2Drelease_release_v1.7.0_bin_linux_amd64_kubectl&d=DwMFaQ&c=uilaK90D4TOVoH58JNXRgQ&r=ebJjFMpXijqZjbZCcbF7yJIq2ES6jM0Q-DEcP-qjjeI&m=sMfOd3IsgFDEQjC8M-dExvTijIpkq549dAV5SUgIXqA&s=U5DyMgpJ6JviFTgJx26Y47cLxgUN47vVnIn0wQINhtw&e=> 2. chmod +x ./kubectl 3. mv kubectl /usr/local/bin 4 setting up config in ~/.kube/config 5. while running kubectl cluster-info an dump oot@k8s-node-0-30a414d1-5a13-41a5-a15a-5970f7fb74ed:~/.kube#<mailto:oot@k8s-node-0-30a414d1-5a13-41a5-a15a-5970f7fb74ed:~/.kube#> kubectl cluster-info dump Unable to connect to the server: dial tcp 10.110.208.207:443: i/o timeout root@k8s-node-0-30a414d1-5a13-41a5-a15a-5970f7fb74ed:~/.kube#<mailto:root@k8s-node-0-30a414d1-5a13-41a5-a15a-5970f7fb74ed:~/.kube#> kubectl cluster-info Kubernetes master is running at https://10.110.208.207:443/<https://urldefense.proofpoint.com/v2/url?u=https-3A__10.110.208.207-3A443_&d=DwMFaQ&c=uilaK90D4TOVoH58JNXRgQ&r=ebJjFMpXijqZjbZCcbF7yJIq2ES6jM0Q-DEcP-qjjeI&m=sMfOd3IsgFDEQjC8M-dExvTijIpkq549dAV5SUgIXqA&s=J_jRQV7D3-6OX_JlBX4KkAVZGq0KIka2-kSq-bZfJlw&e=> AdT: I’m not sure to understand the issue been raised here. Can you clarify? [Gaurav ] - I think I have some issues with kubectl setup so if some one has attempted the kubectl installation and faced similar issues and resolved them , It will help me as well . This is needed so that OOM based deployment can be done and health check scripts refers to kubectl commands . issue 2 - We pulled 2 esr dockers manually inside our OOM deployment . However the new dockers are not getting listed in kubectl commands . So, essentially these containers dont have any specific IP address attached to them along with the port. questions - Are we missing something - I mean is this even valid operation in OOM ? . AdT: Pulling docker images and starting them will not make them appear in Kubernetes UI. For that, you need to create Kubernetes manifests defining the deployment, e.g. how the docker container should looks like (port, entrypoint, volume, pre-conditions, …) and the service, exposing the ports defined in the deployment to the external world (when using NodePort). [Gaurav ] - The use case is - we had OOM Deployement done with few dockers , But we wanted to pull 2 more dockers i.e esr and esr gui . We did pull the dockers from nexsus repo but we donot see them getting listed in the kubectl list --namespaces .Due to this we are also assuming that No Port /IP Address is getting assigned to those . Can this use work or we have to do a complete OOM Deployment fresh ? . - is there anything else that we should be doing in this situation ? . thanks in advance with best regards gaurav This message and the information contained herein is proprietary and confidential and subject to the Amdocs policy statement, you may review at https://www.amdocs.com/about/email-disclaimer <https://www.amdocs.com/about/email-disclaimer>
_______________________________________________ onap-discuss mailing list [email protected] https://lists.onap.org/mailman/listinfo/onap-discuss
