Hello William,

 

thank you very much. When I properly setup NFS for both nodes all AAF pods are up&running.

The only pod not working properly from k8s level is:

onap          dev-aai-champ-58fb6954cf-z84nw                                    0/1       Running   0          21m 

 

but it's started and no error in log, so need to troubleshoot this further ...

 

I think this NFS part is very crucial, especially when in Beijing it's not possible to deploy all pods within single compute node,

as per my understanding some pods might be spawned on one compute, others on another and there might be gaps in configuration required/expected for them if there is no centralized handling of /dockerdata-nfs/ dir.

 

It has to be added into guide I think ...

 

Thanks,

Michal

 

 

--------- Original Message ---------

Sender : William Kurkian <[email protected]>

Date : 2018-05-15 20:55 (GMT+1)

Title : Re: [onap-discuss] Environment for Deploying ONAP

To : Michal Ptacek<[email protected]>

CC : null<[email protected]>, null<[email protected]>, null<[email protected]>, null<[email protected]>

 

Hi Michel,

 

In our deployment, we have two hosts. We had not set up the network file share slave on the second VM. The first one had it setup by a script we used.

 

We were having that aaf-locate pod fail, and a number of other aaf pods were failing due to this dependency, and possible other issues.

 

Once we set up the network share on the slave it started working. In our case we redeployed, though I don't think you should have to. I used the NFS slave script from this page: https://wiki.onap.org/display/DW/ONAP+on+Windriver+ONAP+Developer+Cloud#ONAPonWindriverONAPDeveloperCloud-SettingupanNFSshareforMultinodeKubernetesClusters

 

it is called slave_nfs_node.sh. I also ran the master_nfs_node.sh, but I think I already had it setup.

 

 

On Tue, May 15, 2018 at 3:05 PM, Michal Ptacek <[email protected]> wrote:

Hello William,

sorry for still hijacking your thread :-) But I would like to understand how did you fixed that AAF pods problems,

I am also having problem you indicated in some of previous mails

 

'/bin/bash: /opt/app/osaaf/logs/locate/stdout`date -I`: No such file or directory ' in dev-aaf-locate pod

other AAF pods are having issues as well,

 

I have setup rancher_server and have 2 kubernetes nodes for onap pods, I think I have everything in place but did not used Michael's script in LOG-325.

You mentioned some "NFS requirement", I have NFS :-) Can you please alloborate little more on how you fix that or you just used that LOG-325 script and simply all problems disappeared ?

 

thank you very much,

Michal

 

--------- Original Message ---------

Sender : William Kurkian -X (wkurkian - INSIGHT GLOBAL INC at Cisco) <[email protected]>

Date : 2018-05-15 18:06 (GMT+1)

 

Title : Re: [onap-discuss] Environment for Deploying ONAP

 

To : null<[email protected]>, null<[email protected]>

CC : null<[email protected]>

 

Thanks Michael,

 

We managed to solve our issues. There were two things. 1) We setup our cloud environment according to kubernetes w/ rancher. When we set up our cloud, we didn’t use any particular script to setup the second host we needed for ONAP, and we were unaware of the NFS requirement. We have updated this, and now have only 4 pods with problems, but they do not affect us.  Previously, most of the AAF pods were failing.

 

2) We were having issues with accessing the portal, specifically, the session was timing out. We found that this was due to us not using the entries we configured in /etc/hosts when we initially navigated to the UI. Once we started using those, our session issues were fixed.

 

Thanks for your help, and the resources provided by the community,

William

 

https://www.cisco.com/c/dam/m/en_us/signaturetool/images/logo/Cisco_Logo_no_TM_Cisco_Blue-RGB_43px.png

William Kurkian

Engineer / Developer

[email protected]

Tel:

 

 

 

 

Cisco Systems, Inc.

 

 

 

United States

cisco.com

http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif

Think before you print.

This email may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply email and delete all copies of this message.

Please  click here for Company Registration Information.

 

 

From: Michael O'Brien <[email protected]>
 Sent: Monday, May 14, 2018 7:01 PM
 To: William Kurkian -X (wkurkian - INSIGHT GLOBAL INC at Cisco) <[email protected]>; [email protected]
 Cc: Raghu Mannam -X (rmannam - INSIGHT GLOBAL INC at Cisco) <[email protected]>
 Subject: RE: Environment for Deploying ONAP

 

Guys,

   Hi, post links to which pages you are using on the wiki or readthedocs, whether you are using the automated scripts off LOG-320 – which ones, and post some of the issues you are having with details to either onap-discuss here – or on the wiki as a comment.

    I don’t have any detail on what “issues” is.

 

     ONAP runs on anything that supports Ubuntu 16.04 (a Rancher RI requirement) – openstack, VMWare Fusion/Workstation, AWS, Azure, GCE…..

 

    For reference the oom_entrypoint.sh script will work out of the box and get an onap system up (except for the single node 11O pod limit) on the azure instance ARM template defined in OOM-711/LOG-321

 

   We will be able to help you out and benefit the larger community with your version of deployment as well.

 

    Currently the oom_rancher_install.sh in LOG-325 will install on any Ubuntu 16.04 VM and bring up a fully running kubernetes cluster with helm, kubectl, docker and rancher installed.  However the script is being retrofitted to be able to do a cloud-native cluster – as you know we need at least 2 machines.

     The script can bring up the master, then run docker, the nfs share and the rancher client docker container on each host – wait for the cluster to stabilize and then remove the original host – to make a 1 + n cluster.  I have verified this works best on AWS with EFS, currently getting an Azure cluster working, openstack will work as well.  I have a 4 nuc bare metal cluster that I am also building and will post shortly.

 

http://onap.readthedocs.io/en/latest/submodules/oom.git/docs/oom_quickstart_guide.html 

https://wiki.onap.org/display/DW/ONAP+on+Windriver+ONAP+Developer+Cloud 

https://wiki.onap.org/display/DW/ONAP+on+Kubernetes+on+Rancher 

https://wiki.onap.org/display/DW/Cloud+Native+Deployment

 

 

    Thank you

    /michael

 

From: William Kurkian -X (wkurkian - INSIGHT GLOBAL INC at Cisco) [mailto:[email protected]]
 Sent: Monday, May 14, 2018 4:17 PM
 To: Michael O'Brien <[email protected]>
 Cc: Raghu Mannam -X (rmannam - INSIGHT GLOBAL INC at Cisco) <[email protected]>
 Subject: RE: Environment for Deploying ONAP

 

Apologies, I left out that we were using Kubernetes with Rancher, and have followed the instructions for using these.

 

William

 

https://www.cisco.com/c/dam/m/en_us/signaturetool/images/logo/Cisco_Logo_no_TM_Cisco_Blue-RGB_43px.png

William Kurkian

Engineer / Developer

[email protected]

Tel:

 

 

 

 

Cisco Systems, Inc.

 

 

 

United States

cisco.com

http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif

Think before you print.

This email may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply email and delete all copies of this message.

Please  click here for Company Registration Information.

 

 

From: William Kurkian -X (wkurkian - INSIGHT GLOBAL INC at Cisco)
 Sent: Monday, May 14, 2018 4:15 PM
 To: '[email protected]' <[email protected]>
 Cc: Raghu Mannam -X (rmannam - INSIGHT GLOBAL INC at Cisco) <[email protected]>
 Subject: Environment for Deploying ONAP

 

Hello,

 

My name is William Kurkian. My colleage and I are trying to setup ONAP on 2 VM environment using KVM on a linux box. We have followed the install instructions a number of times, and we are pretty confident we are doing it right at this point.

 

However, we are having issues that don’t seem to be shared by the community Windriver Test environment, which appears to use Openstack. We are not using Openstack, and the issues we are seeing are quite strange.

 

Could we be encountering issues due to not using Openstack? Does it work on environments without it?

 

Any help is greatly appreciated.

Thanks,

William Kurkian

https://www.cisco.com/c/dam/m/en_us/signaturetool/images/logo/Cisco_Logo_no_TM_Cisco_Blue-RGB_43px.png

William Kurkian

Engineer / Developer

[email protected]

Tel:

 

 

 

 

Cisco Systems, Inc.

 

 

 

United States

cisco.com

http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif

Think before you print.

This email may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply email and delete all copies of this message.

Please  click here for Company Registration Information.

 

 

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

_______________________________________________

onap-discuss mailing list

[email protected]

https://lists.onap.org/mailman/listinfo/onap-discuss

 

 

  


_______________________________________________
onap-discuss mailing list
[email protected]
https://lists.onap.org/mailman/listinfo/onap-discuss

 

 

 

  

_______________________________________________
onap-discuss mailing list
[email protected]
https://lists.onap.org/mailman/listinfo/onap-discuss

Reply via email to