[onap-discuss][oom][sdc] Problem in installing SDC with the latest oom code

[Ying, Ruoyu]

Hi all,

I tried to install SDC through OOM. But after several tries, I found the status 
of several SDC ports kept shown as  'Init' after 10hrs' build.

vagrant@master:~$ kubectl get pods -n onap | grep demo-sdc
demo-sdc-sdc-be-7f58cf786f-zwtd5   0/2   Init:0/2   
77 14h
demo-sdc-sdc-cs-7db546cbc9-7zmn6   0/1   Running
16514h
demo-sdc-sdc-es-59f6bdd74d-5wsnb   1/1   Running
0  14h
demo-sdc-sdc-fe-748848bc86-8l2xs   0/2   Init:1/2   
68 14h
demo-sdc-sdc-kb-866bd45469-svbwz   1/1   Running
0  14h
demo-sdc-sdc-onboarding-be-6cbf947dcb-grqqf0/2   Init:0/1   
69 14h
demo-sdc-sdc-wfd-be-7474f78865-sqghq   0/1   Init:0/1   
65 14h
demo-sdc-sdc-wfd-fe-847c65b9b6-zfhfh   0/2   Init:0/1   
69 14h

And when I check the log of these pods, almost all of them shown as this:
Error from server (BadRequest): container "" in pod "" is waiting to 
start: PodInitializing

Has anyone met situations like this? Any suggestion is appreciated.Thanks.

Best Regards,
Ruoyu

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12745): https://lists.onap.org/g/onap-discuss/message/12745
Mute This Topic: https://lists.onap.org/mt/26402078/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[onap-discuss][oom][multicloud] Image not found for pod demo-multicloud-multicloud-windriver

[Ying, Ruoyu]

Hi all,

When I install multicloud through OOM, one pod named 
'demo-multicloud-multicloud-windriver' always got status 'ImagePullBackOff'. 
And I check the events inside, found that it is pulling the image named 
"nexus3.onap.org:10001/onap/multicloud/openstack-windriver:1.2.0-STAGING". 
However, when I search inside the nexus repo, there's no image labelled with 
that version. Any suggestions? Thanks.

Details:
vagrant@master:~$ kubectl describe pods -n onap 
demo-multicloud-multicloud-windriver-747974f5b-v4m8v
Name:   demo-multicloud-multicloud-windriver-747974f5b-v4m8v
Namespace:  onap
Node:   worker1/192.168.0.21
Start Time: Fri, 28 Sep 2018 16:59:28 +0300
Labels: app=multicloud-windriver
pod-template-hash=303530916
release=demo-multicloud
Annotations:sidecar.istio.io/inject=true
Status: Pending
IP: 10.244.1.239
Controlled By:  ReplicaSet/demo-multicloud-multicloud-windriver-747974f5b
Containers:
  multicloud-windriver:
Container ID:
Image:  
nexus3.onap.org:10001/onap/multicloud/openstack-windriver:1.2.0-STAGING
Image ID:
Port:   9005/TCP
State:  Waiting
  Reason:   ImagePullBackOff
Ready:  False
Restart Count:  0
   Liveness:   http-get 
http://:9005/api/multicloud-titaniumcloud/v1/swagger.json delay=30s timeout=10s 
period=10s #success=1 #failure=5
Environment:
  MSB_ADDR:msb-iag.onap
  MSB_PORT:80
  AAI_ADDR:aai.onap
  AAI_PORT:8443
  AAI_SCHEMA_VERSION:  v13
  AAI_USERNAME:AAI
  AAI_PASSWORD:AAI
Mounts:
  /opt/windriver/titanium_cloud/pub/config/log.yml from windriver-logconfig 
(rw)
  /var/log/onap from windriver-log (rw)
  /var/run/secrets/kubernetes.io/serviceaccount from default-token-w465t 
(ro)
  filebeat-onap:
Container ID:   
docker://1ef84d1c7b62a5232dbeba284cb17f6704ebf1c09b43c427e794c28a8b986050
Image:  docker.elastic.co/beats/filebeat:5.5.0
Image ID:   
docker-pullable://docker.elastic.co/beats/filebeat@sha256:fe7602b641ed8ee288f067f7b31ebde14644c4722d9f7960f176d621097a5942
Port:   
State:  Running
  Started:  Fri, 28 Sep 2018 17:50:24 +0300
Ready:  True
Restart Count:  0
Environment:
Mounts:
  /usr/share/filebeat/data from windriver-data-filebeat (rw)
  /usr/share/filebeat/filebeat.yml from filebeat-conf (rw)
  /var/log/onap from windriver-log (rw)
  /var/run/secrets/kubernetes.io/serviceaccount from default-token-w465t 
(ro)
Conditions:
  Type   Status
  InitializedTrue
  Ready  False
  PodScheduled   True
Volumes:
  windriver-log:
Type:EmptyDir (a temporary directory that shares a pod's lifetime)
Medium:
  windriver-data-filebeat:
Type:EmptyDir (a temporary directory that shares a pod's lifetime)
Medium:
  filebeat-conf:
Type:  ConfigMap (a volume populated by a ConfigMap)
Name:  multicloud-filebeat-configmap
Optional:  false
  windriver-logconfig:
Type:  ConfigMap (a volume populated by a ConfigMap)
Name:  demo-multicloud-multicloud-windriver-log-configmap
Optional:  false
  default-token-w465t:
Type:Secret (a volume populated by a Secret)
SecretName:  default-token-w465t
Optional:false
QoS Class:   BestEffort
Node-Selectors:  
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
 node.kubernetes.io/unreachable:NoExecute for 300s
Events:
  Type Reason   Age   From  Message
   --       ---
  Normal   Pulling  34m (x118 over 11h)   kubelet, worker1  pulling image 
"nexus3.onap.org:10001/onap/multicloud/openstack-windriver:1.2.0-STAGING"
  Warning  Failed   24m (x2396 over 10h)  kubelet, worker1  Error: 
ImagePullBackOff
  Normal   BackOff  4m (x2483 over 10h)   kubelet, worker1  Back-off pulling 
image "nexus3.onap.org:10001/onap/multicloud/openstack-windriver:1.2.0-STAGING"

vagrant@worker1:~$ sudo docker search 
nexus3.onap.org:10001/onap/multicloud/openstack-windriver
NAME   DESCRIPTION  
 STARS OFFICIAL   AUTOMATED
nexus3.onap.org:10001/onap/multicloud/openstack-windriver:1.1.2 
 0
nexus3.onap.org:10001/onap/multicloud/openstack-windriver:latest
 0
nexus3.onap.org:10001/onap/multicloud/openstack-windriver:v1.0.0
 0

Best Regards,
Ruoyu

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12744): https://lists.onap.org/g/onap-discuss/message/12744
Mute This Topic: https://lists.onap.org/mt/26388044/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=

Re: [onap-discuss] [ONAP Helpdesk #61551] RE: Competing priorities in onap - CLM vs Code

[Michael O'Brien]

Update: I looked more into the failures just now and yes it is "banned 
snapshots" the "banned dependencies" and the " Dependency 
ch.qos.logback:logback-classic:jar:1.1.3 located at Module 
org.onap.logging-analytics:logging-docker-root:pom:1.2.2-SNAPSHOT" stuck out 
before I realized it was the forced release of my jars.

Releasing my jars should fix all of these - I'll release as-is - Prudence and I 
need to release the 8 release projects this week.
Issue closed and adjusted as a release exercise.


Yes, when I looked back at recent docker and release jobs – started seeing 
these – which look to be enforced CLM rules – blocking Jenkins jobs success – I 
didn’t save the link I got the following from – I may have mixed up “banned 
dependencies” with enforce-no-snapshots – which should say “banned snapshots” 
as the failure message
Looking for logs – can’t seem to find them now - before or after I took out 
1.1.3

Dependency ch.qos.logback:logback-classic:jar:1.1.3 located at Module 
org.onap.logging-analytics:logging-docker-root:pom:1.2.2-SNAPSHOT
Dependency ch.qos.logback:logback-classic:jar:1.1.3 located at Module 
org.onap.logging-analytics:logging-kubernetes:pom:1.2.2-SNAPSHOT
Dependency ch.qos.logback:logback-classic:jar:1.1.3 located at Module 
org.onap.logging-analytics:logging-reference:pom:1.2.2-SNAPSHOT

This one is not CLM it is SNAPSHOTS – or unreleased code – I will need to 
release my demo code as-is before RC0 to fix these
https://jenkins.onap.org/view/logging-analytics/job/logging-analytics-master-release-version-java-daily/244/

Identified problems
Banned dependencies

Some Enforcer rules have failed. Look in the log for specific messages 
explaining why the rule failed 
Indication 1 


2:01:55 [INFO] --- maven-enforcer-plugin:3.0.0-M1:enforce 
(enforce-no-snapshots) @ logging-demo ---
12:01:55 [WARNING] Rule 0: org.apache.maven.plugins.enforcer.RequireReleaseDeps 
failed with message:
12:01:55 Found Banned Dependency: 
org.onap.logging-analytics:logging-slf4j:jar:1.2.2-SNAPSHOT
12:01:55 Found Banned Dependency: 
org.onap.logging-analytics:logging-mock-service:jar:1.2.2-SNAPSHOT
12:01:55 Found Banned Dependency: 
org.onap.logging-analytics:logging-library:jar:1.2.2-SNAPSHOT
12:01:55 Use 'mvn dependency:tree' to locate the source of the banned 
dependencies.


-Original Message-
From: Jessica Wagantall via RT  
Sent: Friday, September 28, 2018 7:36 PM
To: gildas.lani...@huawei.com
Cc: Michael O'Brien ; onap-discuss@lists.onap.org; 
onap-...@lists.onap.org
Subject: [ONAP Helpdesk #61551] RE: Competing priorities in onap - CLM vs Code

Dear Michael

Can you please point me to an example? 
The only dependencies I know we are banning are the ones described in this 
example:
https://wiki.onap.org/display/DW/Release+Versioning+Strategy#ReleaseVersioningStrategy-IssueofdependencyonSnapshot

Thanks!
Jess

On Fri Sep 28 19:00:30 2018, gildas.lani...@huawei.com wrote:
> Looping in Helpdesk.
> 
> I am not aware the Jenkins fails on banned dependencies you stated 
> below.
> @Helpdesk, Your thoughts are welcome.
> 
> Thanks,
> Gildas
> ONAP Release Manager
> 1 415 238 6287
> 
> From: onap-discuss@lists.onap.org [mailto:onap-discuss@lists.onap.org]
> On Behalf Of Michael O'Brien
> Sent: Friday, September 28, 2018 5:09 AM
> To: onap-discuss@lists.onap.org; onap-...@lists.onap.org
> Subject: [onap-discuss] Competing priorities in onap - CLM vs Code
> 
> Team,
>Just a comment on the vulnerabilities work and its impact on 
> everything else.
>Since the release Jenkins jobs now fail on banned dependencies and 
> we are forced to fix the CLM critical issues - just be aware that 
> fixing each of these takes time away from development as - I have 
> fixed one of the remaining 9 in my project last night and am working 
> through the rest.
>Fixing a CLM issue is not trivial as you know - half the time there 
> is an override brought in with another library - fixing one usually 
> illuminates a common pattern so the rest should go faster now - 
> however I am blocking developers waiting on my changes in onap and 
> acumos because of this.
>/michael
> 
> 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
> 



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 


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12743): https://lists.onap.org/g/onap-discuss/message/12743
Mute This Topic: https://lists.onap.org/mt/26387899/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

[Vijay VK]

Hi Mike,  Yep definitely not before RC0..:) On the same boat.

As all DCAE services are deployed through cloudify blueprint, the rolling 
upgrade will utilize corresponding cli as well.  Following is the upgrade 
command syntax -

cfy executions start -d  -p .yaml execute_operation
Where the  is assigned per service and input.yaml includes the 
new image to be used.

Looking forward to the discussion for deployment harmonization. One aspect to 
be considered also will be how the artifacts for deployment can be generated 
dynamically. The current workflow is for SDC (by designer) to generate the 
service template and generate Tosca blueprints - which DCAE Controller 
monitors, fetches and deploy on demand on request from CLAMP.  What we are 
discussing on the thread is more around the deployment aspect – however the 
onboarding and dynamic deployment artifact generation (by user/operator) are 
critical functions for ONAP.   We will need SDC and CLAMP team input here as 
well.

Srini – Regarding the DCAE/SDC/CLAMP/Policy flow, they are well captured under 
control-loop subcommittee wiki links below. These explains the different phases 
involved with demo video of testing done for R2. Hope this helps.
https://wiki.onap.org/display/DW/Control+Loop+Sub+Committee+-+Beijing+Flow+Diagrams
https://wiki.onap.org/display/DW/Control+Loop+Sub+Committee+-+Beijing+Integration+Testing+Plan

Regards,
Vijay

From: onap-discuss@lists.onap.org  On Behalf Of 
Mike Elliott
Sent: Friday, September 28, 2018 2:22 PM
To: Addepalli, Srinivasa R ; VENKATESH KUMAR, 
VIJAY ; onap-discuss@lists.onap.org; fbroc...@cisco.com; JI, 
LUSHENG ; GUPTA, ALOK 
Cc: Fabien Andrieux (fandrieu) ; Donald Hunter (donaldh) 
; Trevor Smith (trsmith2) ; Cristina 
Precup (cprecup) ; Chenxi Du (chenxdu) ; 
Roger Maitland 
Subject: Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

Hi Vijay & Srini,

I apologize Vijay if I was unaware that rolling upgrades are now a part of 
Casablanca. That is fantastic news!
So, am I now able to do the following to upgrade ONAP and if DCAE (or any of 
its components) have changed, they will be updated too?


   helm deploy demo staging/onap --namespace onap -f ~/override.yaml

Regarding the need for static and dynamic services. @Vijay, have you looked at 
using the Tiller API (server-side of Helm)?  It can be used to dynamically 
deploy Helm charts and provide full Helm and k8s integration. It can also be 
used to spin up short-lived K8s Jobs that perform all sorts of different 
actions (eg. call remote server endpoint, update DB, etc.).  As far as the 
dynamic policy configuration aspect is concerned, this is not an area we are 
familiar with. Perhaps this is a good delineation point where Helm and k8s 
handle LCM and DCAE Controller handles policy configuration? Worth discussing 
at the F2F I think.

It definitely sounds like we have an opportunity to collaborate (and harmonize) 
on a single deployment solution that addresses all needs. With RC0 looming, it 
would seem that meeting before the architecture meeting to discuss such topics, 
may be unrealistic.  At least for me 😉
A few OOM team members, including myself, will be present at the F2F in 
Montreal. Looking forward to some great discussions!

Thanks,
Mike

From: "Addepalli, Srinivasa R" 
mailto:srinivasa.r.addepa...@intel.com>>
Date: Friday, September 28, 2018 at 12:32 PM
To: "VENKATESH KUMAR, VIJAY" mailto:vv7...@att.com>>, 
"onap-discuss@lists.onap.org" 
mailto:onap-discuss@lists.onap.org>>, Mike Elliott 
mailto:mike.elli...@amdocs.com>>, 
"fbroc...@cisco.com" 
mailto:fbroc...@cisco.com>>, "JI, LUSHENG" 
mailto:l...@research.att.com>>, "GUPTA, ALOK" 
mailto:ag1...@att.com>>
Cc: "Fabien Andrieux (fandrieu)" 
mailto:fandr...@cisco.com>>, "Donald Hunter (donaldh)" 
mailto:dona...@cisco.com>>, "Trevor Smith (trsmith2)" 
mailto:trsmi...@cisco.com>>, "Cristina Precup (cprecup)" 
mailto:cpre...@cisco.com>>, "Chenxi Du (chenxdu)" 
mailto:chen...@cisco.com>>, Roger Maitland 
mailto:roger.maitl...@amdocs.com>>
Subject: RE: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

Hi Mike and Vijay,

Good discussion.  I plan to be Montreal.

It would be great if Kubernetes is used for all kinds of services – 
Infrastructure Services and dynamic spark applications (analytics applications).

With respect to OOM : So far, OOM is used to deploy services that live 
throughout ONAP life.  Can OOM be used to deploy services that are ephemeral 
too.  For example, spark-submit command internally communicates with Kubernetes 
API server to bring up spark applications on containers. Can that be used 
still? In this case, there could be multiple clients to the K8S API server – 
One is OOM via Helm, second one  is spark-submit or some controller that bring 
up services dynamically. I don’t’ see any challenges on top of my head, but 
want to get that confirmed with respect to OOM.

Hi Vijay

Re: [onap-discuss] [it-infrastructure-alerts] Gerrit and Jenkins unavailable Friday 1:00pm to 3:00pm PT time

[Jessica Wagantall]

Dear team

Gerrit and Jenkins are back available.

Thanks!
Jess

On Fri, Sep 28, 2018 at 12:30 PM Jessica Wagantall <
jwagant...@linuxfoundation.org> wrote:

> Dear ONAP team,
>
> In preparation for this maintenance,  Jenkins has been set to quietDown
> mode now.
>
> Thanks so much for your patience!
> Jess
>
> On Thu, Sep 27, 2018 at 2:34 PM Jessica Wagantall <
> jwagant...@linuxfoundation.org> wrote:
>
>> What: The Linux Foundation will be performing a migration of Gerrit to AWS
>>
>> When: Friday Sep 28 at 1:00pm Pacific
>>
>> Why: Gerrit needs to be migrated to AWS. Moving to aws will improve
>> reliability and
>> scalability of gerrit system
>>
>> Impact: Gerrit will be unavailable and Jenkins will need to be set to
>> quiet mode for 2:00 hrs
>>
>> Thanks so much!
>> Jess
>>
> 
>

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12741): https://lists.onap.org/g/onap-discuss/message/12741
Mute This Topic: https://lists.onap.org/mt/26339203/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[onap-discuss] AWS JIRA rest environment available now

[Jessica Wagantall]

Dear ONAP team,

Ryan has helped us creating https://jira-new.onap.org/ which is a copy of
our JIRA but in an
AWS server. I will be working on testing this instance during the times
that most failures to generate reports occurred.

Could I please ask the teams to also give it a shot if you find
jira.onap.org having issues?

Please notice that this is a test environment for now and any changes into
jira-new.onap.org
will not be propagated into jira.onap.org.

Once we gather some examples of this test environment working better than
the production one,
we can do the switch to AWS.

Thanks a ton for your help as always!
Jess

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12740): https://lists.onap.org/g/onap-discuss/message/12740
Mute This Topic: https://lists.onap.org/mt/26380269/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] [it-infrastructure-alerts] Gerrit and Jenkins unavailable Friday 1:00pm to 3:00pm PT time

[Jessica Wagantall]

Dear ONAP team,

In preparation for this maintenance,  Jenkins has been set to quietDown
mode now.

Thanks so much for your patience!
Jess

On Thu, Sep 27, 2018 at 2:34 PM Jessica Wagantall <
jwagant...@linuxfoundation.org> wrote:

> What: The Linux Foundation will be performing a migration of Gerrit to AWS
>
> When: Friday Sep 28 at 1:00pm Pacific
>
> Why: Gerrit needs to be migrated to AWS. Moving to aws will improve
> reliability and
> scalability of gerrit system
>
> Impact: Gerrit will be unavailable and Jenkins will need to be set to
> quiet mode for 2:00 hrs
>
> Thanks so much!
> Jess
>

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12739): https://lists.onap.org/g/onap-discuss/message/12739
Mute This Topic: https://lists.onap.org/mt/26339203/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

[Mike Elliott]

Hi Vijay & Srini,

I apologize Vijay if I was unaware that rolling upgrades are now a part of 
Casablanca. That is fantastic news!
So, am I now able to do the following to upgrade ONAP and if DCAE (or any of 
its components) have changed, they will be updated too?


   helm deploy demo staging/onap --namespace onap -f ~/override.yaml

Regarding the need for static and dynamic services. @Vijay, have you looked at 
using the Tiller API (server-side of Helm)?  It can be used to dynamically 
deploy Helm charts and provide full Helm and k8s integration. It can also be 
used to spin up short-lived K8s Jobs that perform all sorts of different 
actions (eg. call remote server endpoint, update DB, etc.).  As far as the 
dynamic policy configuration aspect is concerned, this is not an area we are 
familiar with. Perhaps this is a good delineation point where Helm and k8s 
handle LCM and DCAE Controller handles policy configuration? Worth discussing 
at the F2F I think.

It definitely sounds like we have an opportunity to collaborate (and harmonize) 
on a single deployment solution that addresses all needs. With RC0 looming, it 
would seem that meeting before the architecture meeting to discuss such topics, 
may be unrealistic.  At least for me 😉
A few OOM team members, including myself, will be present at the F2F in 
Montreal. Looking forward to some great discussions!

Thanks,
Mike

From: "Addepalli, Srinivasa R" 
Date: Friday, September 28, 2018 at 12:32 PM
To: "VENKATESH KUMAR, VIJAY" , "onap-discuss@lists.onap.org" 
, Mike Elliott , 
"fbroc...@cisco.com" , "JI, LUSHENG" 
, "GUPTA, ALOK" 
Cc: "Fabien Andrieux (fandrieu)" , "Donald Hunter 
(donaldh)" , "Trevor Smith (trsmith2)" , 
"Cristina Precup (cprecup)" , "Chenxi Du (chenxdu)" 
, Roger Maitland 
Subject: RE: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

Hi Mike and Vijay,

Good discussion.  I plan to be Montreal.

It would be great if Kubernetes is used for all kinds of services – 
Infrastructure Services and dynamic spark applications (analytics applications).

With respect to OOM : So far, OOM is used to deploy services that live 
throughout ONAP life.  Can OOM be used to deploy services that are ephemeral 
too.  For example, spark-submit command internally communicates with Kubernetes 
API server to bring up spark applications on containers. Can that be used 
still? In this case, there could be multiple clients to the K8S API server – 
One is OOM via Helm, second one  is spark-submit or some controller that bring 
up services dynamically. I don’t’ see any challenges on top of my head, but 
want to get that confirmed with respect to OOM.

Hi Vijay,

In your email, you mentioned this : “however for PNDA app services to be able 
to leverage SDC/Policy/CLAMP flows, I believe should be Tosca model driven”
Can you elaborate on this?  Please give an example of a flow.   Trying to 
understand if TOSCA is helping on workload bring up or anything else? If it is 
only workload bring up, isn’t Helm/spark-submit/direct K8S yaml good enough?

Thanks
Srini


From: VENKATESH KUMAR, VIJAY [mailto:vv7...@att.com]
Sent: Friday, September 28, 2018 8:11 AM
To: onap-discuss@lists.onap.org; mike.elli...@amdocs.com; fbroc...@cisco.com; 
Addepalli, Srinivasa R ; JI, LUSHENG 
; GUPTA, ALOK 
Cc: Fabien Andrieux (fandrieu) ; Donald Hunter (donaldh) 
; Trevor Smith (trsmith2) ; Cristina 
Precup (cprecup) ; Chenxi Du (chenxdu) ; 
Roger Maitland 
Subject: RE: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

Hi Mike,
One of the main distinction for DCAE Controller is need for supporting both 
static and dynamic services. Besides LCM -  the DCAE controller also provides 
dynamic policy configuration update capabilities for all the services deployed. 
If OOM Team has plans to extend any of these capabilities, it would be great to 
discuss any harmonization opportunity here.
With regard to rolling upgrade of DCAE services – it is supported. This 
capabilities existed for some components in R2; in Casablanca – enhancement was 
done to support this feature for all services components as well 
(DCAEGEN2-543).
For PNDA, I would approach platform deployment and service differently. The 
platform deployment itself can be triggered via OOM, however for PNDA app 
services to be able to leverage SDC/Policy/CLAMP flows, I believe should be 
Tosca model driven.  Being new integration -  it will be great to discuss 
overall onboarding and deployment flow for PNDA and identify any optimization 
opportunities. We could possibly have this for discussion in one of the DCAE or 
OOM weekly meetings (post RC0) if not all are making for F2F meetup?
Regards,
Vijay

From: onap-discuss@lists.onap.org 
mailto:onap-discuss@lists.onap.org>> On Behalf Of 
Mike Elliott
Sent: Friday, September 28, 2018 10:02 AM
To: onap-discuss@lists.onap.org; 
fbroc...@cisco.com<

[onap-discuss] Scheduling of Daily Jobs

[Gary Wu]

Hi PTLs,

Currently all of the Daily Jenkins jobs (Java/Docker release jobs) are 
scheduled for around noon UTC (8 AM Eastern), which imposes severe resource 
contention on nexus and nexus3 for the next several hours and causing some of 
them to fail due to timeouts.

To fix this, I will be submitting a change to spread the load throughout the 
day instead of having them all run at the same time.  Please let me know if you 
have any concerns.

Thanks,
Gary



-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12737): https://lists.onap.org/g/onap-discuss/message/12737
Mute This Topic: https://lists.onap.org/mt/26378993/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

[Srini]

Hi Mike and Vijay,

Good discussion.  I plan to be Montreal.

It would be great if Kubernetes is used for all kinds of services – 
Infrastructure Services and dynamic spark applications (analytics applications).

With respect to OOM : So far, OOM is used to deploy services that live 
throughout ONAP life.  Can OOM be used to deploy services that are ephemeral 
too.  For example, spark-submit command internally communicates with Kubernetes 
API server to bring up spark applications on containers. Can that be used 
still? In this case, there could be multiple clients to the K8S API server – 
One is OOM via Helm, second one  is spark-submit or some controller that bring 
up services dynamically. I don’t’ see any challenges on top of my head, but 
want to get that confirmed with respect to OOM.

Hi Vijay,

In your email, you mentioned this : “however for PNDA app services to be able 
to leverage SDC/Policy/CLAMP flows, I believe should be Tosca model driven”
Can you elaborate on this?  Please give an example of a flow.   Trying to 
understand if TOSCA is helping on workload bring up or anything else? If it is 
only workload bring up, isn’t Helm/spark-submit/direct K8S yaml good enough?

Thanks
Srini


From: VENKATESH KUMAR, VIJAY [mailto:vv7...@att.com]
Sent: Friday, September 28, 2018 8:11 AM
To: onap-discuss@lists.onap.org; mike.elli...@amdocs.com; fbroc...@cisco.com; 
Addepalli, Srinivasa R ; JI, LUSHENG 
; GUPTA, ALOK 
Cc: Fabien Andrieux (fandrieu) ; Donald Hunter (donaldh) 
; Trevor Smith (trsmith2) ; Cristina 
Precup (cprecup) ; Chenxi Du (chenxdu) ; 
Roger Maitland 
Subject: RE: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

Hi Mike,
One of the main distinction for DCAE Controller is need for supporting both 
static and dynamic services. Besides LCM -  the DCAE controller also provides 
dynamic policy configuration update capabilities for all the services deployed. 
If OOM Team has plans to extend any of these capabilities, it would be great to 
discuss any harmonization opportunity here.
With regard to rolling upgrade of DCAE services – it is supported. This 
capabilities existed for some components in R2; in Casablanca – enhancement was 
done to support this feature for all services components as well 
(DCAEGEN2-543).
For PNDA, I would approach platform deployment and service differently. The 
platform deployment itself can be triggered via OOM, however for PNDA app 
services to be able to leverage SDC/Policy/CLAMP flows, I believe should be 
Tosca model driven.  Being new integration -  it will be great to discuss 
overall onboarding and deployment flow for PNDA and identify any optimization 
opportunities. We could possibly have this for discussion in one of the DCAE or 
OOM weekly meetings (post RC0) if not all are making for F2F meetup?
Regards,
Vijay

From: onap-discuss@lists.onap.org 
mailto:onap-discuss@lists.onap.org>> On Behalf Of 
Mike Elliott
Sent: Friday, September 28, 2018 10:02 AM
To: onap-discuss@lists.onap.org; 
fbroc...@cisco.com; Addepalli, Srinivasa R 
mailto:srinivasa.r.addepa...@intel.com>>; 
VENKATESH KUMAR, VIJAY mailto:vv7...@att.com>>; JI, LUSHENG 
mailto:l...@research.att.com>>; GUPTA, ALOK 
mailto:ag1...@att.com>>
Cc: Fabien Andrieux (fandrieu) mailto:fandr...@cisco.com>>; 
Donald Hunter (donaldh) mailto:dona...@cisco.com>>; Trevor 
Smith (trsmith2) mailto:trsmi...@cisco.com>>; Cristina 
Precup (cprecup) mailto:cpre...@cisco.com>>; Chenxi Du 
(chenxdu) mailto:chen...@cisco.com>>; Roger Maitland 
mailto:roger.maitl...@amdocs.com>>
Subject: Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

Hi Frank,

It would be great to discuss options with you for deploying PNDA components in 
OOM/K8s. One thing that is missing right now from DCAE is the ability to 
perform an upgrade of DCAE components after its initial install. DCAE operates 
under a different controller environment. It’s kind of its own island that is 
treated special from the rest of the components within an ONAP deployment. It 
would be really great if we could have consistent behavior with the rest of 
ONAP. As of now, it is possible to perform a non-disruptive rolling upgrade of 
all components in ONAP, except for DCAE.

If you (or anyone on this email thread) is going to the architecture meetup in 
Montreal, perhaps that would be a good time to meet F2F and discuss how we 
evolve to such a common deployment model? Otherwise, the OOM team is available 
at any time to discuss via Zoom.

Thanks,
Mike.

--
Mike Elliott
ONAP OOM PTL
Senior Architect - Amdocs


From: mailto:onap-discuss@lists.onap.org>> on 
behalf of "Frank Brockners via Lists.Onap.Org" 
mailto:fbrockne=cisco@lists.onap.org>>
Reply-To: "onap-discuss@lists.onap.org" 
mailto:onap-discuss@lists.onap.org>>, 
"fbroc...@cisco.com

[onap-discuss] Remove unused Beijing Jenkins jobs

[Gary Wu]

Hi PTLs,

There are still a large number of Jenkins jobs building beijing artifacts 
daily.  If you no longer need them, please remove the unused beijing jobs from 
Jenkins so we can lighten the load.

Thanks,
Gary



-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12735): https://lists.onap.org/g/onap-discuss/message/12735
Mute This Topic: https://lists.onap.org/mt/26378268/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

[Vijay VK]

Hi Mike,
One of the main distinction for DCAE Controller is need for supporting both 
static and dynamic services. Besides LCM -  the DCAE controller also provides 
dynamic policy configuration update capabilities for all the services deployed. 
If OOM Team has plans to extend any of these capabilities, it would be great to 
discuss any harmonization opportunity here.
With regard to rolling upgrade of DCAE services – it is supported. This 
capabilities existed for some components in R2; in Casablanca – enhancement was 
done to support this feature for all services components as well 
(DCAEGEN2-543).
For PNDA, I would approach platform deployment and service differently. The 
platform deployment itself can be triggered via OOM, however for PNDA app 
services to be able to leverage SDC/Policy/CLAMP flows, I believe should be 
Tosca model driven.  Being new integration -  it will be great to discuss 
overall onboarding and deployment flow for PNDA and identify any optimization 
opportunities. We could possibly have this for discussion in one of the DCAE or 
OOM weekly meetings (post RC0) if not all are making for F2F meetup?
Regards,
Vijay

From: onap-discuss@lists.onap.org  On Behalf Of 
Mike Elliott
Sent: Friday, September 28, 2018 10:02 AM
To: onap-discuss@lists.onap.org; fbroc...@cisco.com; Addepalli, Srinivasa R 
; VENKATESH KUMAR, VIJAY ; JI, 
LUSHENG ; GUPTA, ALOK 
Cc: Fabien Andrieux (fandrieu) ; Donald Hunter (donaldh) 
; Trevor Smith (trsmith2) ; Cristina 
Precup (cprecup) ; Chenxi Du (chenxdu) ; 
Roger Maitland 
Subject: Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

Hi Frank,

It would be great to discuss options with you for deploying PNDA components in 
OOM/K8s. One thing that is missing right now from DCAE is the ability to 
perform an upgrade of DCAE components after its initial install. DCAE operates 
under a different controller environment. It’s kind of its own island that is 
treated special from the rest of the components within an ONAP deployment. It 
would be really great if we could have consistent behavior with the rest of 
ONAP. As of now, it is possible to perform a non-disruptive rolling upgrade of 
all components in ONAP, except for DCAE.

If you (or anyone on this email thread) is going to the architecture meetup in 
Montreal, perhaps that would be a good time to meet F2F and discuss how we 
evolve to such a common deployment model? Otherwise, the OOM team is available 
at any time to discuss via Zoom.

Thanks,
Mike.

--
Mike Elliott
ONAP OOM PTL
Senior Architect - Amdocs


From: mailto:onap-discuss@lists.onap.org>> on 
behalf of "Frank Brockners via Lists.Onap.Org" 
mailto:fbrockne=cisco@lists.onap.org>>
Reply-To: "onap-discuss@lists.onap.org" 
mailto:onap-discuss@lists.onap.org>>, 
"fbroc...@cisco.com" 
mailto:fbroc...@cisco.com>>
Date: Thursday, September 27, 2018 at 2:31 AM
To: "Addepalli, Srinivasa R" 
mailto:srinivasa.r.addepa...@intel.com>>, 
"'VENKATESH KUMAR, VIJAY'" mailto:vv7...@att.com>>, "'JI, 
LUSHENG'" mailto:l...@research.att.com>>, "'GUPTA, 
ALOK'" mailto:ag1...@att.com>>
Cc: "'onap-discuss@lists.onap.org'" 
mailto:onap-discuss@lists.onap.org>>, "Fabien 
Andrieux (fandrieu)" mailto:fandr...@cisco.com>>, "Donald 
Hunter (donaldh)" mailto:dona...@cisco.com>>, "Trevor Smith 
(trsmith2)" mailto:trsmi...@cisco.com>>, "Cristina Precup 
(cprecup)" mailto:cpre...@cisco.com>>, "Chenxi Du (chenxdu)" 
mailto:chen...@cisco.com>>
Subject: Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics


Hi Srini,

sorry for the delay – ONS.EU got into our way. PNDA 5.0 – which is expected to 
come out later in October – will include HDP 2.6.5 which contains Spark 2.3, 
see 
https://wiki.onap.org/display/DW/Integrating+PNDA#IntegratingPNDA-PNDA5.0Componentsversions.
 There are various options to run applications on Spark – and containers are 
one option.
PNDA apps are currently not deployed using containers – and integration of 
DCAE’s catalogue with PNDA’s deployment manager is targeted for post Casablanca.
With the closer integration of DCAE and PNDA, we should also ask the question 
of how we deploy DCAE apps moving forward.
Doing that as containers via OOM/K8s would be an interesting option to consider 
for DCAE as well.
Cc’ing a few more folks.

Cheers, Frank

From: Addepalli, Srinivasa R 
mailto:srinivasa.r.addepa...@intel.com>>
Sent: Freitag, 21. September 2018 17:59
To: 'VENKATESH KUMAR, VIJAY' mailto:vv7...@att.com>>; 'JI, 
LUSHENG' mailto:l...@research.att.com>>; 'GUPTA, ALOK' 
mailto:ag1...@att.com>>; Frank Brockners (fbrockne) 
mai

Re: [onap-discuss] [**EXTERNAL**] [Onap-usecasesub] MobileEdgeX open source announcement

[Margaret Chiosi]

I think we need to define what we mean by application management (e.g. 
functions) to make sure we aren’t having different understanding and therefore 
conflict in our discussions.

Thank You,
Margaret Chiosi
VP Open Ecosystem Team

Admin: Sophie Johnson
sophie.johns...@huawei.com
+1 (908) 541-3590

Futurewei Technologies, Inc.
Fixed Network Solution CC
400 Crossing Blvd
Bridgewater, NJ 08807
(cell) +1-732-216-5507

[cid:image001.png@01CCE115.88302080]

From: onap-usecase...@lists.onap.org [mailto:onap-usecase...@lists.onap.org] On 
Behalf Of Arash Hekmat
Sent: Tuesday, September 25, 2018 10:07 AM
To: BEGWANI, VIMAL ; Margaret Chiosi (A) 
; Ramki Krishnan ; Srini 
; Ranganathan, Raghu 
Cc: onap-usecase...@lists.onap.org; onap-discuss@lists.onap.org
Subject: Re: [**EXTERNAL**] [Onap-usecasesub] MobileEdgeX open source 
announcement

To add to what Vimal said.

Even if an operator owns all the Edge clouds, the operator would have many 
tools operating on those Edge clouds at different levels. ONAP is one of those 
tools not the only one. ONAP operates at the Network level not at the user 
application level. The operator would have separate tool(s) operating at the 
user application level. The rule of the separation of concerns. To try to add 
user application management in ONAP is to mix two separate domains and to 
pollute ONAP with layer 7+ logic which overloads ONAP with complexity and 
reduces ONAP’s focus and usability.

The application management framework(s) should run independent of ONAP but 
there needs to be horizontal interfaces defined between app. frameworks and 
ONAP so that the app. frameworks could request or discover Network services 
from ONAP.

Best Regards,
Arash

From: BEGWANI, VIMAL mailto:vb1...@att.com>>
Sent: Tuesday, September 25, 2018 9:21 AM
To: Margaret Chiosi 
mailto:margaret.chio...@huawei.com>>; Ramki 
Krishnan mailto:ram...@vmware.com>>; Srini 
mailto:srinivasa.r.addepa...@intel.com>>; 
Ranganathan, Raghu mailto:rra...@ciena.com>>
Cc: Arash Hekmat mailto:arash.hek...@amdocs.com>>; 
onap-usecase...@lists.onap.org; 
onap-discuss@lists.onap.org
Subject: RE: [**EXTERNAL**] [Onap-usecasesub] MobileEdgeX open source 
announcement

There are several reasons why we might want to keep Edge Cloud and ONAP 
separate.  Enterprise customer might want to deploy their applications / 
content at the right edge location and pay only for the cloud services instead 
of paying for a fully ONAP managed applications or content.  While back AT&T 
had fully managed hosting and basic co-location services (we provided servers, 
basic connectivity check, customers managed their applications).  Co-location 
services was widely more popular than fully managed hosting services.  Customer 
might not want to go through the trouble of packing their applications that SDC 
can upload.  I can envision a scenario where Edge Cloud is an independent 
services that is being used by more than one operator (just like cell towers in 
US today or AWS or Azure).

Regards,
Vimal

From: onap-usecase...@lists.onap.org 
mailto:onap-usecase...@lists.onap.org>> On 
Behalf Of Margaret Chiosi
Sent: Tuesday, September 25, 2018 9:08 AM
To: Ramki Krishnan mailto:ram...@vmware.com>>; Srini 
mailto:srinivasa.r.addepa...@intel.com>>; 
Ranganathan, Raghu mailto:rra...@ciena.com>>
Cc: HEKMAT, ARASH mailto:arash.hek...@amdocs.com>>; 
onap-usecase...@lists.onap.org; 
onap-discuss@lists.onap.org
Subject: Re: [**EXTERNAL**] [Onap-usecasesub] MobileEdgeX open source 
announcement

But it may be a passthrough for certain functions. Maybe it would be good to 
create a table with all the different cloud edge ‘management’ functions and 
talk through what ONAPs role is.
e.g. optimized location for app; scale out/in; availability; optimized server 
for a specific set of SLA (CPU speed or RAM speed or NIC specific or Linux 
version or CPU type….)
If cloud edge software covers all of this – then what is ONAP’s role. If cloud 
edge only does subset, then what is ONAP’s role.

Thank You,
Margaret Chiosi
VP Open Ecosystem Team

Admin: Sophie Johnson
sophie.johns...@huawei.com
+1 (908) 541-3590

Futurewei Technologies, Inc.
Fixed Network Solution CC
400 Crossing Blvd
Bridgewater, NJ 08807
(cell) +1-732-216-5507

[cid:image001.png@01CCE115.88302080]

From: Ramki Krishnan [mailto:ram...@vmware.com]
Sent: Monday, September 24, 2018 9:36 PM
To: Margaret Chiosi (A) 
mailto:margaret.chio...@huawei.com>>; Srini 
mailto:srinivasa.r.addepa...@intel.com>>; 
Ranganathan, Raghu mailto:rra...@ciena.com>>
Cc: Arash Hekmat mailto:arash.hek...@amdocs.com>>; 
onap-usecase...@lists.onap.org; 
onap-discuss@lists.onap.org
Subject: RE: [**EXTERN

Re: [onap-discuss] [Onap-usecasesub] Edge Application Provisioning

[Arun Gupta]

With regard to edge application provisioning, I'd like to draw your attention 
to Eden Rozin's presentation at ONS Europe 2018.
https://events.linuxfoundation.org/wp-content/uploads/2017/12/Edgility-Modelling-and-Orchestration-in-Edge-Cloud-Environments-Eden-Rozin-ATT.pdf

Thank you,
Arun Gupta

Arun Gupta
Lead Member of Technical Staff
Domain 2 Architecture and Planning
200 S Laurel Ave – Bldg B, B5-3Z09
Middletown, NJ 07748
P: 732 420 6597
ag7...@att.com ( arungu...@att.com )

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12732): https://lists.onap.org/g/onap-discuss/message/12732
Mute This Topic: https://lists.onap.org/mt/26257544/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics

[Mike Elliott]

Hi Frank,

It would be great to discuss options with you for deploying PNDA components in 
OOM/K8s. One thing that is missing right now from DCAE is the ability to 
perform an upgrade of DCAE components after its initial install. DCAE operates 
under a different controller environment. It’s kind of its own island that is 
treated special from the rest of the components within an ONAP deployment. It 
would be really great if we could have consistent behavior with the rest of 
ONAP. As of now, it is possible to perform a non-disruptive rolling upgrade of 
all components in ONAP, except for DCAE.

If you (or anyone on this email thread) is going to the architecture meetup in 
Montreal, perhaps that would be a good time to meet F2F and discuss how we 
evolve to such a common deployment model? Otherwise, the OOM team is available 
at any time to discuss via Zoom.

Thanks,
Mike.

--
Mike Elliott
ONAP OOM PTL
Senior Architect - Amdocs


From:  on behalf of "Frank Brockners via 
Lists.Onap.Org" 
Reply-To: "onap-discuss@lists.onap.org" , 
"fbroc...@cisco.com" 
Date: Thursday, September 27, 2018 at 2:31 AM
To: "Addepalli, Srinivasa R" , "'VENKATESH 
KUMAR, VIJAY'" , "'JI, LUSHENG'" , 
"'GUPTA, ALOK'" 
Cc: "'onap-discuss@lists.onap.org'" , "Fabien 
Andrieux (fandrieu)" , "Donald Hunter (donaldh)" 
, "Trevor Smith (trsmith2)" , "Cristina 
Precup (cprecup)" , "Chenxi Du (chenxdu)" 
Subject: Re: [onap-discuss] #DCAEgen2 - DCAE - Adding ML based network analytics


Hi Srini,

sorry for the delay – ONS.EU got into our way. PNDA 5.0 – which is expected to 
come out later in October – will include HDP 2.6.5 which contains Spark 2.3, 
see 
https://wiki.onap.org/display/DW/Integrating+PNDA#IntegratingPNDA-PNDA5.0Componentsversions.
 There are various options to run applications on Spark – and containers are 
one option.
PNDA apps are currently not deployed using containers – and integration of 
DCAE’s catalogue with PNDA’s deployment manager is targeted for post Casablanca.
With the closer integration of DCAE and PNDA, we should also ask the question 
of how we deploy DCAE apps moving forward.
Doing that as containers via OOM/K8s would be an interesting option to consider 
for DCAE as well.
Cc’ing a few more folks.

Cheers, Frank

From: Addepalli, Srinivasa R 
Sent: Freitag, 21. September 2018 17:59
To: 'VENKATESH KUMAR, VIJAY' ; 'JI, LUSHENG' 
; 'GUPTA, ALOK' ; Frank Brockners 
(fbrockne) 
Cc: 'onap-discuss@lists.onap.org' 
Subject: RE: DCAE - Adding ML based network analytics

+Frank

Hi Vijay and Frank,

Few basic questions on PNDA+DCAE.

I see that PNDA supports both Spark batch and Spark streaming. Spark streaming 
is of interest. We also see that Kafka based streaming is supported by PNDA.


  1.  Does PNDA use spark version 2.3 or above to allow Kubernetes based 
scheduling of spark jobs?
  2.  If I had implemented my analytics application (with spark executor, 
MLlib, Kafka way of getting data stream and Kafka way of outputting the 
analysis output) as a container  that work on Spark 2.3.0,  does it require any 
changes to work with PNDA and DCAE?  If it requires changes, what kind of 
changes are required?

Thanks
Srini


From: Addepalli, Srinivasa R
Sent: Wednesday, September 19, 2018 6:59 AM
To: VENKATESH KUMAR, VIJAY mailto:vv7...@att.com>>; JI, LUSHENG 
mailto:l...@research.att.com>>; GUPTA, ALOK 
mailto:ag1...@att.com>>
Cc: onap-discuss@lists.onap.org
Subject: RE: DCAE - Adding ML based network analytics

Thanks Vijay.

Understood. We are moving away from CDAP to PNDA and Apache Flink as big data 
platform.
As I understand from PNDA website, it seems to be based on Spark. Since Spark 
supports Machine learning analytics (via MLlib as well as Bring your own 
models), I think PNDA/Spark may be what we need to look into.

Thanks for pointers Vijay. Will go through the link you have provided and 
understand the onboarding of analytics micro-services.

Thanks
Srini



From: VENKATESH KUMAR, VIJAY [mailto:vv7...@att.com]
Sent: Tuesday, September 18, 2018 5:30 PM
To: Addepalli, Srinivasa R 
mailto:srinivasa.r.addepa...@intel.com>>; JI, 
LUSHENG mailto:l...@research.att.com>>; GUPTA, ALOK 
mailto:ag1...@att.com>>
Cc: onap-discuss@lists.onap.org
Subject: RE: DCAE - Adding ML based network analytics

Hi Srini – Thanks for your interest into DCAE. Apologize for the delay in 
response (was tied up with upcoming M4 delivery).

As you might know, DCAE provides a platform for onboarding different type of mS 
(collectors, analytics, CE’s). The TCA instance was build to support the 
original OpenECOMP and ONAP usecases and it was chosen as CDAP application as 
that was the analytics platform supported then. However since Amsterdam 
release, based on community feedback (and lack of support in CDAP)-  we are 
working toward integrating newer analytics platform into DCAE. PNDA is one such 
platform – there is work currently under way in R3 by Cisco team to inte

[onap-discuss] Competing priorities in onap - CLM vs Code

[Michael O'Brien]

Team,
   Just a comment on the vulnerabilities work and its impact on everything else.
   Since the release Jenkins jobs now fail on banned dependencies and we are 
forced to fix the CLM critical issues - just be aware that fixing each of these 
takes time away from development as - I have fixed one of the remaining 9 in my 
project last night and am working through the rest.
   Fixing a CLM issue is not trivial as you know - half the time there is an 
override brought in with another library - fixing one usually illuminates a 
common pattern so the rest should go faster now - however I am blocking 
developers waiting on my changes in onap and acumos because of this.
   /michael

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 


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12729): https://lists.onap.org/g/onap-discuss/message/12729
Mute This Topic: https://lists.onap.org/mt/26375697/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] What are the meanings of some parameters in DG file update-network-to-aai?

[Gaurav Agrawal]

I would suggest you to look at AAI schema file for the field type. Anyways let 
me give you a example:

  "speed-value": "10" (we are storing bps to this field, there 
is a field called “speed-units” to specifically mention the unit of speed)

  "network-ref": "providerId/aaa/topologyId/bbb/clientId/ccc"

  "transparent": "true" (it’s a Boolean)
Thanks and Regards,
Gaurav

From: onap-discuss@lists.onap.org [mailto:onap-discuss@lists.onap.org] On 
Behalf Of wang.senx...@zte.com.cn
Sent: 28 September 2018 06:48
To: Gaurav agrawal 
Cc: menglin...@chinamobile.com; huang.zhuo...@zte.com.cn; 
onap-discuss@lists.onap.org
Subject: Re: [onap-discuss] What are the meanings of some parameters in DG file 
update-network-to-aai?




Thanks, Gaurav, this is really a big help. And could you please give me an 
example in the following form? I don't know the format

and unit of the values of these three parameters.

  "speed-value": ""

  "network-ref": ""

  "transparent": ""





Best Regards!

王森枭 Wang Senxiao



软件开发工程师 Software Development Engineer
软件平台南京开发一部/有线研究院/有线产品经营部 SDN Nanjing Development Dept. Ⅰ/Wireline Product R&D 
Institute/Wireline Product Operation Division


[cid:image001.gif@01D4574F.6B90C670]

[cid:image002.gif@01D4574F.6B90C670]
武汉市东湖高新技术开发区华师园路6号中兴通讯


ZTE Corporation, 6 Huashiyuan Rd,

Donghu Hi-Tech Administrative area,Wuhan, P.R.China,430223

T: +86 027 51811060 M:+86 15988865324
E: wang.senx...@zte.com.cn
www.zte.com.cn

原始邮件
发件人：GauravAgrawal mailto:gaurav.agra...@huawei.com>>
收件人：王森枭10200860;
抄送人：menglin...@chinamobile.com 
mailto:menglin...@chinamobile.com>>黄卓垚10112215;onap-discuss@lists.onap.org
 mailto:onap-discuss@lists.onap.org>>
日 期 ：2018年09月27日 16:10
主 题 ：Re: [onap-discuss] What are the meanings of some parameters in DG file 
update-network-to-aai?
Hi Wang Senxiao

Please find the description of asked parameter.

speed-value: It is used to store bandwidth, it can hold Electric/Ethernet  
layer bandwidth depending on native topology type this tp belongs to.
network-ref: It is supposed to contain reference to the native  topology to 
which this termination points belongs.
transparent: This indicates whether termination-point can support  a 
transparent service, basically its used to identify which service type(s) this 
tp can support.
in-maint: used to indicate whether or  not this object is in maintenance mode 
(maintenance mode = true). This field you can find in some other AAI resources 
too.

Hope it will be helpful.

Thanks and Regards,
Gaurav Agrawal
Senior System Architect
Network Software Platform Dept, HTIPL [2012 Laboratories]

Huawei Technologies India Pvt. Ltd
Survey No. 37, Next to EPIP Area, Kundanahalli, Whitefield
Benguluru-560066, Karnataka
Tel: +91-80-49160700, Ext 70126 || Mob : 7838700296, Email :  
gaurav.agra...@huawei.com

From: wang.senx...@zte.com.cn 
[mailto:wang.senx...@zte.com.cn]
Sent: 2018年9月27日  8:07
To: Gaurav agrawal
Cc: menglin...@chinamobile.com; 
huang.zhuo...@zte.com.cn; 
onap-discuss@lists.onap.org
Subject: What are the meanings of some parameters in DG file 
update-network-to-aai?


Hi, Gaurav

I see this DG file(in the attachment) in the project sdnc/oam about 
updating network to aai, and I know its function.

But when I study the code of storing node termination-point information to 
resource 'p-interface',  I do not understand

the meaning of the parameters network-ref, transparent, in-maint represented, 
and also whether the bandwidth

represented by the parameter speed-value  refers to the Electrical layer or the 
Ethernet layer, do you have a value

of parameter speed-value? Please tell me their meaning, thanks!





Best Regards!

王森枭 Wang  Senxiao



软件开发工程师 Software Development  Engineer
软件平台南京开发一部/有线研究院/有线产品经营部 SDN Nanjing Development Dept. Ⅰ/Wireline  Product R&D 
Institute/Wireline Product Operation Division


[cid:image001.gif@01D4574F.6B90C670]

[cid:image002.gif@01D4574F.6B90C670]
武汉市东湖高新技术开发区华师园路6号中兴通讯

ZTE Corporation, 6 Huashiyuan Rd,

Donghu Hi-Tech Administrative area,Wuhan, P.R.China,430223

T: +86 027 51811060 M:+86  15988865324
E: wang.senx...@zte.com.cn
www.zte.com.cn






-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12728): https://lists.onap.org/g/onap-discuss/message/12728
Mute This Topic: https://lists.onap.org/mt/26289029/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] [SO][Camunda][Unauthorized][HeatBasedInstallation][Windriver]

[Steve Smokowski]

Camunda:Auth, Admin:admin is only used for the cockpit login.  The other 
passwords are stored below.  Inside apihandler you will see the encrypted 
password, in Camunda override.yaml you will see the websecurity section.

password1$, which is encrypted.  Generally you can use the cryptoUtils unit 
test to test the decryption to see if it is correct.

Thanks

-Steve


From:  on behalf of "Sliwka, Wojciech (Nokia - 
PL/Wroclaw)" 
Reply-To: "onap-discuss@lists.onap.org" , 
"wojciech.sli...@nokia.com" 
Date: Friday, September 28, 2018 at 3:04 AM
To: "onap-discuss@lists.onap.org" 
Cc: "Krysiak, Piotr (Nokia - PL/Wroclaw)" , "Krysiak, 
Adam (Nokia - PL/Wroclaw)" , "Koblosz, Sandra (Nokia - 
PL/Wroclaw)" , "Golabek, Tomasz (Nokia - PL/Wroclaw)" 
, "Darosz, Piotr (Nokia - PL/Wroclaw)" 

Subject: [onap-discuss] 
[SO][Camunda][Unauthorized][HeatBasedInstallation][Windriver]


Hi,



during testing on Windriver I noticed following exception in SO:


2018-09-27T11:23:47.457Z|1a22b244-c94f-487c-985b-77e8325fcff9| 
org.onap.so.apihandler.common.CamundaClient - Camunda url is: 
http://bpmn-infra:8081/mso/async/services/CreateGenericALaCarteServiceInstance
2018-09-27T11:23:47.462Z|1a22b244-c94f-487c-985b-77e8325fcff9| 
org.onap.so.apihandler.common.RequestClient - Security exception
javax.crypto.BadPaddingException: Given final block not properly padded. Such 
issues can arise if a bad key is used during decryption.
at com.sun.crypto.provider.CipherCore.doFinal(CipherCore.java:991)
at com.sun.crypto.provider.CipherCore.doFinal(CipherCore.java:847)
at com.sun.crypto.provider.AESCipher.engineDoFinal(AESCipher.java:446)
at javax.crypto.Cipher.doFinal(Cipher.java:2164)
at org.onap.so.utils.CryptoUtils.decrypt(CryptoUtils.java:68)




It looks like provided mso.camundaAuth in override.yaml (which is
F3473596C526938329DF877495B494DC374D1C4198ED3AD305EA3ADCBBDA1862) is not valid.
Due to that admin:admin is used which ends up with '401 Unauthorized'.

Do you know how mso.camundaAuth should be filled?


Attached more logs.
Env: Windriver Integration-SB-07


BR,
Wojtek

wojciech.sli...@nokia.com




-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12727): https://lists.onap.org/g/onap-discuss/message/12727
Mute This Topic: https://lists.onap.org/mt/26374190/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] [SDC] Questions about the ways to trigger SDC distribute the NS csar

[Lianhao Lu]

Hi SDC team,

Thanks for the info. But now I got another problem and hopefully somebody can 
give me some clues. Thanks very much!

I first used the sdc-sim to create a service, then I used restful API to query 
the service,

"GET /sdc/v1/catalog/services" replied with the service UUID 
db7840a9-6bea-4a9b-8427-fc61a1ce3fe5

Then I tried to use resful api " POST 
/sdc/v1/catalog/services/db7840a9-6bea-4a9b-8427-fc61a1ce3fe5 
distribution/{opEnvId}/activate"  to distribute it, however I got a 500 error.

I noticed I can use sdc-sim to distribute the service successfully though web 
GUI,  whenI track the http request of sdc-sim, I found that it issued a 
different restful api with a different service UUID to activate the service:

POST 
sdc2/rest/v1/catalog/services/dd25daf3-a60c-4d12-ba79-c309472dbeb0/distribution/AUTO/activate

So my question here is:
Where does the serviceUUID dd25daf3 comes from? Here I only created a service 
with UUID db7840a9 . I noticed that in the response body of the POST, it 
contains the following things:
...
uniqueId": "dd25daf3-a60c-4d12-ba79-c309472dbeb0",
"uuid": "db7840a9-6bea-4a9b-8427-fc61a1ce3fe5",


How can I get that uniqueId if I only knows the service uuid?

Best Regards,
-Lianhao

From: onap-discuss@lists.onap.org [mailto:onap-discuss@lists.onap.org] On 
Behalf Of Brian
Sent: Wednesday, September 26, 2018 9:06 PM
To: onap-discuss@lists.onap.org; Lu, Lianhao ; 
'onap-...@lists.onap.org' 
Subject: Re: [onap-discuss] [SDC] Questions about the ways to trigger SDC 
distribute the NS csar

Look in the robot code for the sdc_interface.robot.

There is an example of how to initiate a distribution sincne robot does it 
through the SDC API.

Brian


From: onap-discuss@lists.onap.org 
mailto:onap-discuss@lists.onap.org>> On Behalf Of 
Lianhao Lu
Sent: Wednesday, September 26, 2018 1:20 AM
To: onap-discuss@lists.onap.org; Lu, 
Lianhao mailto:lianhao...@intel.com>>; 
'onap-...@lists.onap.org' 
mailto:onap-...@lists.onap.org>>
Subject: Re: [onap-discuss] [SDC] Questions about the ways to trigger SDC 
distribute the NS csar

I found this RESTful API: POST 
https://{serverRoot}/sdc/v1/catalog/services/{serviceUUID}/distribution/{opEnvId}/activate

Is that the correct one to activate the service distribution? Can anyone please 
tell me what is {opEnvId}? I went through the SDC catalog API reference, and 
this is the only place need operational environment ID.  What kind of value 
should we put here in this RESTful API call?

-Lianhao

From: onap-discuss@lists.onap.org 
[mailto:onap-discuss@lists.onap.org] On Behalf Of Lianhao Lu
Sent: Wednesday, September 26, 2018 10:19 AM
To: onap-discuss@lists.onap.org; 
'onap-...@lists.onap.org' 
mailto:onap-...@lists.onap.org>>
Subject: [onap-discuss] [SDC] Questions about the ways to trigger SDC 
distribute the NS csar

Hi SDC team,

Currently we're trying to do some stability test of policy which is a SDC 
distribution client user. So I'm wondering besides clicking on the button in 
SDC GUI, is there other way to trigger SDC to distribute the NS csar to all SDC 
distribution client users, e.g. is there a SDC RESTful API we can use?

Best Regards,
-Lianhao


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12726): https://lists.onap.org/g/onap-discuss/message/12726
Mute This Topic: https://lists.onap.org/mt/26226896/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [onap-discuss] [DMAAP] Questions about pairwise tests , DMAAP need preload something for SDC?

[libo zhu]

Hello, Sunil,
   Many thanks for the provided info. After preload these data, 
DMAAP could works with SDC successfully.
   Form the SDC's healthcheck, the DE(Dmaap Eningeer) is UP, there 
are some topics shown from DMAAP APIs.

   Thanks Michael, Sunil and Jorge for all the help.

Thanks
Libo

From: UNNAVA, SUNIL [mailto:su6...@att.com]
Sent: Friday, September 28, 2018 12:54 AM
To: Zhu, Libo ; KOYA, RAMPRASAD ; 
'onap-discuss@lists.onap.org' ; 
'hg0071...@techmahindra.com' 
Cc: Lu, Lianhao ; Liam Fallon ; 
Addepalli, Srinivasa R ; LANDO, MICHAEL 

Subject: RE: [onap-discuss] [DMAAP] Questions about pairwise tests , DMAAP need 
preload something for SDC?

Hi Libo,

SDC leverages the pre-configured Kaka and Zookeeper data directories, so that 
they no need to create the API keys. As you used the script in DMaaP Message 
Router CSIT script, you may have missed that and that may be causing the issue. 
 Refer the following OOM deployment files for mounting the Kafka and ZooKeeper 
data directories in the init containers

https://gerrit.onap.org/r/gitweb?p=oom.git;a=blob;f=kubernetes/dmaap/charts/message-router/charts/message-router-kafka/templates/deployment.yaml;h=d6daf3eefeacf581b934bde57eb182f0884428cb;hb=HEAD
https://gerrit.onap.org/r/gitweb?p=oom.git;a=blob;f=kubernetes/dmaap/charts/message-router/charts/message-router-zookeeper/templates/deployment.yaml;h=6c1f7621186ab126c2f87b612189834f8dfa992c;hb=HEAD
https://gerrit.onap.org/r/gitweb?p=dmaap/messagerouter/messageservice.git;a=tree;f=oom-topics;h=88e469795d2aebfa90944065f88d6f87d1dd1c97;hb=HEAD

Thanks,
Sunil

From: Zhu, Libo mailto:libo@intel.com>>
Sent: Thursday, September 27, 2018 11:28 AM
To: KOYA, RAMPRASAD mailto:rk5...@att.com>>; 
'onap-discuss@lists.onap.org' 
mailto:onap-discuss@lists.onap.org>>; UNNAVA, 
SUNIL mailto:su6...@att.com>>; 'hg0071...@techmahindra.com' 
mailto:hg0071...@techmahindra.com>>
Cc: Lu, Lianhao mailto:lianhao...@intel.com>>; Liam 
Fallon mailto:liam.fal...@ericsson.com>>; Addepalli, 
Srinivasa R 
mailto:srinivasa.r.addepa...@intel.com>>; 
LANDO, MICHAEL mailto:ml6...@intl.att.com>>
Subject: [onap-discuss] [DMAAP] Questions about pairwise tests , DMAAP need 
preload something for SDC?

Hello, DMAAP team,
   Currently we're trying to do some pairwise tests for SDC, DMAAP 
and Policy, met issue about "Error occured during access to U-EB Server. 
Operation: register to topic as producer".
Install the DMAAP Message Router by the script in CSIT, is there something like 
preload DMAAP needs to do for SDC to work?


Best regards.
Libo
Tel: +86-21-6116 7988
MP: +86-134 8204 6855
Mail: libo@intel.com


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12725): https://lists.onap.org/g/onap-discuss/message/12725
Mute This Topic: https://lists.onap.org/mt/26332826/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[onap-discuss] [integration] Benchmark test result statistic

[Hongliang Bian]

 

Hi Brian,

 

Thank you for your instruction last night. Attached please find the test
result statistic and the summary of related errors. Thanks J

 

BRs,

Hongliang

 

 

 

卞宏梁/Bian Hongliang



中国电信股份有限公司北京研究院/China Telecom Corporation Limited Beijing
Research Institute

昌平区北七家镇未来科技城南区中国电信北京信息科技创新园，邮编：102209

China Telecom Beijing Information Science & Technology Innovation Park,
Southern Zone of Future Science & Technology City, Beiqijia Town, Changping
District, Beijing, 102209, China
TEL：  +86 010-5090-2955 
E-mail：   bianhl@chinatelecom.cn

 


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12724): https://lists.onap.org/g/onap-discuss/message/12724
Mute This Topic: https://lists.onap.org/mt/26374361/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-



性能测试结果记录0926.xlsx
Description: MS-Excel 2007 spreadsheet

[onap-discuss] [integration] Benchmark related JIRA reports

[Hongliang Bian]

 

Hi Yang and Bin,

 

According to the discussion last night, we have summarized some issue and
reported on JIRA. We assigned the vcpe script related issue to Yang, and
assigned the openstack related issue to Bin. We don’t know whether it is
proper or not, so please let we know if there is any problem. And you can
also deliver these issues to your related colleagues.

 

Thanks for your patience and expertise helping us with all these issues in
the Benchmark test! J

 

BRs,

Hongliang

 

 

 

卞宏梁/Bian Hongliang



中国电信股份有限公司北京研究院/China Telecom Corporation Limited Beijing
Research Institute

昌平区北七家镇未来科技城南区中国电信北京信息科技创新园，邮编：102209

China Telecom Beijing Information Science & Technology Innovation Park,
Southern Zone of Future Science & Technology City, Beiqijia Town, Changping
District, Beijing, 102209, China
TEL：  +86 010-5090-2955 
E-mail：   bianhl@chinatelecom.cn

 


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12723): https://lists.onap.org/g/onap-discuss/message/12723
Mute This Topic: https://lists.onap.org/mt/26374308/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[onap-discuss] [SO][Camunda][Unauthorized][HeatBasedInstallation][Windriver]

[Sliwka, Wojciech (Nokia - PL/Wroclaw)]

Hi,


during testing on Windriver I noticed following exception in SO:


2018-09-27T11:23:47.457Z|1a22b244-c94f-487c-985b-77e8325fcff9| 
org.onap.so.apihandler.common.CamundaClient - Camunda url is: 
http://bpmn-infra:8081/mso/async/services/CreateGenericALaCarteServiceInstance
2018-09-27T11:23:47.462Z|1a22b244-c94f-487c-985b-77e8325fcff9| 
org.onap.so.apihandler.common.RequestClient - Security exception
javax.crypto.BadPaddingException: Given final block not properly padded. Such 
issues can arise if a bad key is used during decryption.
at com.sun.crypto.provider.CipherCore.doFinal(CipherCore.java:991)
at com.sun.crypto.provider.CipherCore.doFinal(CipherCore.java:847)
at com.sun.crypto.provider.AESCipher.engineDoFinal(AESCipher.java:446)
at javax.crypto.Cipher.doFinal(Cipher.java:2164)
at org.onap.so.utils.CryptoUtils.decrypt(CryptoUtils.java:68)



It looks like provided mso.camundaAuth in override.yaml (which is

F3473596C526938329DF877495B494DC374D1C4198ED3AD305EA3ADCBBDA1862) is not valid.

Due to that admin:admin is used which ends up with '401 Unauthorized'.

Do you know how mso.camundaAuth should be filled?


Attached more logs.
Env: Windriver Integration-SB-07


BR,
Wojtek

wojciech.sli...@nokia.com



-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#12722): https://lists.onap.org/g/onap-discuss/message/12722
Mute This Topic: https://lists.onap.org/mt/26374190/21656
Group Owner: onap-discuss+ow...@lists.onap.org
Unsubscribe: https://lists.onap.org/g/onap-discuss/unsub  
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-



so_exception
Description: so_exception