David,
Hi, not good, will try to help you out.
The CD hourly build is running against Beijing/master, I also did a run on a
server for the amsterdam branch – not seeing these 2 pods down.
Without checking your logs on the 2 containers – vnc-portal definitely has
issues under helm 2.5+ where 2.7 is installed by default under rancher 1.6.11+
There are several reasons vnc-portal would fail (infrastructure config, or
dependent pods failing) – but a failure on elasticsearch – which has no
dependencies defined – would likely mean an issue with the undercloud config =
(rancher/helm/kubernetes/docker).
So just to cover off your environment make sure you are running only rancher
1.6.10 (for as Alexis has mentioned – we are bound by the version of Kubernetes
(up to 1.8) that we can use with Rancher (our manager RI). Our version of
docker 1.12.x is dictated by the rancher/kubernetes pair. And we are currently
on Helm 2.3 until an issue with the vnc-portal image is fixed – which is
limiting us to rancher 1.6.10.
/michael
From: [email protected]
[mailto:[email protected]] On Behalf Of Bainbridge, David
Sent: Wednesday, January 3, 2018 14:58
To: Alexis de Talhouët <[email protected]>; BRIAN D FREEMAN
<[email protected]>
Cc: GILBERT, MAZIN E <[email protected]>; onap-discuss
<[email protected]>
Subject: Re: [onap-discuss] [**EXTERNAL**] Re: [integration][oom] OOM readiness
for Beijing release
Just attempting to deploy via OOM. Left with 2 pods not starting correctly:
NAMESPACE NAME READY
STATUS RESTARTS AGE
onap-log elasticsearch-6df4f65775-w48c9 0/1
CrashLoopBackOff 11 1h
onap-portal vnc-portal-845d84676c-jwzf9 0/1
CrashLoopBackOff 7 1h
This is with Amsterdam branch, pulled this morning.
Avèk respè,
/david
From:
<[email protected]<mailto:[email protected]>>
on behalf of Alexis de Talhouët
<[email protected]<mailto:[email protected]>>
Date: Wednesday, January 3, 2018 at 11:42 AM
To: BRIAN D FREEMAN <[email protected]<mailto:[email protected]>>
Cc: "GILBERT, MAZIN E" <[email protected]<mailto:[email protected]>>,
onap-discuss <[email protected]<mailto:[email protected]>>
Subject: [**EXTERNAL**] Re: [onap-discuss] [integration][oom] OOM readiness for
Beijing release
Brian,
Yes, it does sound reasonable. That being said, I’m very much involved in
finishing DCAE support in OOM. Once done, along with the testing the various
use cases, I’ll sync up with Michael to provide a clear, straightforward
documentation, both for setup and usage, providing tips and ways to debug.
Thanks,
Alexis
On Jan 3, 2018, at 2:15 PM, FREEMAN, BRIAN D
<[email protected]<mailto:[email protected]>> wrote:
Gary,
I think Michael was trying to be complete in describing options with a plan to
do a cleaned up documentation after it was working and scripted up. Perhaps its
time to formally start a new page with the final recommended – single path and
scripted install (I agree that it was daunting at the start to find the latest
parts/procedure which was much simpler than the first procedure)
Michael/Alexis – is that reasonable ?
Brian
From:
[email protected]<mailto:[email protected]>
[mailto:[email protected]] On Behalf Of Gary Wu
Sent: Wednesday, January 03, 2018 1:16 PM
To: PLATANIA, MARCO
<[email protected]<mailto:[email protected]>>; Alexis de
Talhouët <[email protected]<mailto:[email protected]>>; GILBERT,
MAZIN E <[email protected]<mailto:[email protected]>>
Cc: onap-discuss
<[email protected]<mailto:[email protected]>>
Subject: Re: [onap-discuss] [integration][oom] OOM readiness for Beijing release
My experience with OOM (excluding DCAE) so far is that SDC fails to pass health
check about half the time. When this happens, the best solution seems to be
re-deploying the entire ONAP again, and if you’re lucky it will work. This
appears to have been observed by others as well. The good thing is that we can
do the re-deploy in about 10 minutes (excluding DCAE).
First impressions:
· The system is highly dependent in using exact versions of everything:
docker, kubernetes, helm, rancher, etc. If anything is slightly off, things
would break, and it would not be obvious exactly what was wrong. As a result,
it makes the system feel fragile.
· The SDC intermittent issue above makes the OOM deployments feel not
fully stable/repeatable yet.
· The OOM “how-to” documentation has almost too much information,
covering multiple options/paths that sometimes overlap (e.g. where to install
Rancher server vs agent, whether to use the cd.sh or manually perform
individual steps). This made it hard for a newcomer to discern which steps
need to be run next given which other choices were made earlier, and it was
also hard to figure out if any mistakes were made performing some prior steps.
I was not able to find any information on how to deploy DCAE via OOM; the most
I got so far was that the DCAE controller container was able to bring up about
half of the expected DCAE VMs, but not enough to be able to pass the DCAE
health check.
Thanks,
Gary
From:
[email protected]<mailto:[email protected]>
[mailto:[email protected]] On Behalf Of PLATANIA, MARCO
(MARCO)
Sent: Wednesday, January 03, 2018 9:15 AM
To: Alexis de Talhouët
<[email protected]<mailto:[email protected]>>; GILBERT, MAZIN E
(MAZIN E) <[email protected]<mailto:[email protected]>>
Cc: onap-discuss
<[email protected]<mailto:[email protected]>>
Subject: Re: [onap-discuss] [integration][oom] OOM readiness for Beijing release
I’ve been playing around with the OOM solution for a while and I was able to
instantiate the vFW, vLB/vDNS, plus another dummy generic VNF that I created. I
installed a minimal version of DCAE (only the collector and one CDAP container
with the Threshold-crossing microservices) to run closed loop, waiting for the
whole DCAE platform to be ready. I wasn’t able to run closed loop yet, perhaps
due to out-of-sync configuration in my environment. I’m trying to fix this and
see how far I can go with closed loop.
Marco
From:
<[email protected]<mailto:[email protected]>>
on behalf of Alexis de Talhouët
<[email protected]<mailto:[email protected]>>
Date: Wednesday, January 3, 2018 at 11:36 AM
To: "GILBERT, MAZIN E (MAZIN E)"
<[email protected]<mailto:[email protected]>>
Cc: onap-discuss
<[email protected]<mailto:[email protected]>>
Subject: Re: [onap-discuss] [integration][oom] OOM readiness for Beijing release
Mazin,
So far, that I’m aware of, a few ppl from various company have tried OOM
Amsterdam, and have successfully deployed vFWCL.
If you want names and email, I guess I can find them for you.
We’re currently testing vFWCL close loop with DCAE being deployed by OOM; I’d
say it’s 80% done.
Thanks,
Alexis
On Jan 3, 2018, at 11:21 AM, GILBERT, MAZIN E (MAZIN E)
<[email protected]<mailto:[email protected]>> wrote:
Thanks Roger.
Have there been other groups who have managed to bring up Amsterdam with OOM
(with and without DCAE)?
Can you point me to any so I can understand their experience.
mazin
On Jan 3, 2018, at 11:15 AM, Roger Maitland
<[email protected]<mailto:[email protected]>> wrote:
Hi Helen,
David is on vacation so I’ll answer - OOM is ready.
The OOM team has been working with the Integration team and the entire ONAP
community to create the
amsterdam<https://urldefense.proofpoint.com/v2/url?u=https-3A__gerrit.onap.org_r_gitweb-3Fp-3Doom.git-3Ba-3Dshortlog-3Bh-3Drefs_heads_amsterdam&d=DwMGaQ&c=LFYZ-o9_HUMeMTSQicvjIg&r=IKSC5mg8GeOiSar1dax3GQ&m=_dlaJi5cUYGfdQpkrTvOLmwKf9nHldPQl5dmZYtz2iQ&s=uYM-V22L0V1EhonUacGlX4_i-MF4XdWNLtYWzReGYA4&e=>
release of OOM which is now available. There is a continuous delivery system
that deploys the master branch of all of the ONAP components with OOM –
here<https://urldefense.proofpoint.com/v2/url?u=http-3A__kibana.onap.info-3A5601_app_kibana-23_dashboard_AWAtvpS63NTXK5mX2kuS-3F-5Fg-3D-28refreshInterval-3A-28display-3AOff-2Cpause-3A-21f-2Cvalue-3A0-29-2Ctime-3A-28from-3Anow-2D24h-2Cmode-3Aquick-2Cto-3Anow-29-29-26-5Fa-3D-28description-3A-2527-2527-2Cfilters-3A-21-28-29-2Coptions-3A-28darkTheme-3A-21f-29-2Cpanels-3A-21-28-28col-3A1-2Cid-3AAWAts77k3NTXK5mX2kuM-2CpanelIndex-3A1-2Crow-3A1-2Csize-5Fx-3A8-2Csize-5Fy-3A&d=DwMGaQ&c=LFYZ-o9_HUMeMTSQicvjIg&r=e3d1ehx3DI5AoMgDmi2Fzw&m=TEGneiYu3X69Tbakkf-0jbhukaSEN5O-Dj-rwTm4eXY&s=HXzlpENhYHSnNMtU61IUybmlPUBxj2VxcRYHbRcCRJw&e=>
is the dashboard. As this system is effectively continuously evaluating ONAP
health this should be very useful for the Integration team to quickly identify
component submissions that result in a degradation.
The initial version of DCAE support is under final testing. This version
brings up the DCAE controller which then brings up the rest of the DCAE VMs.
There is a discussion scheduled for tomorrow between the OOM and DCAE team on a
more fully containerized solution. The OOM team is also working on deploying
OpenStack DNS Designate to simply the deployment of OpenStack infrastructure
required for DCAE. We’ll let the community know when this is ready.
Cheers,
Roger
From:
[email protected]<mailto:[email protected]>
[mailto:[email protected]] On Behalf Of Yunxia Chen
Sent: Tuesday, January 2, 2018 5:25 PM
To: onap-discuss
<[email protected]<mailto:[email protected]>>; Sauvageau,
David <[email protected]<mailto:[email protected]>>
Subject: [onap-discuss] [integration][oom] OOM readiness for Beijing release
Hi, David,
Happy new year and it was very nice to talk with you at Santa Clara. Is OOM
ready for Integration team to take over to test the Beijing release, maybe
start with Amsterdam project first? From last meeting, it seems have a little
bit issue on DCAE, has it resolved yet?
Regards,
Helen Chen
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://urldefense.proofpoint.com/v2/url?u=https-3A__www.amdocs.com_about_email-2Ddisclaimer&d=DwMGaQ&c=LFYZ-o9_HUMeMTSQicvjIg&r=IKSC5mg8GeOiSar1dax3GQ&m=_dlaJi5cUYGfdQpkrTvOLmwKf9nHldPQl5dmZYtz2iQ&s=EJl3g98o-LmIGfk4sjynukqHSJ3epLjqaYO4anfC954&e=>
_______________________________________________
onap-discuss mailing list
[email protected]<mailto:[email protected]>
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.onap.org_mailman_listinfo_onap-2Ddiscuss&d=DwICAg&c=LFYZ-o9_HUMeMTSQicvjIg&r=IKSC5mg8GeOiSar1dax3GQ&m=_dlaJi5cUYGfdQpkrTvOLmwKf9nHldPQl5dmZYtz2iQ&s=j7WSvxdz-fKgmH76NuG-x2cW3E_oZL4nU8OuXiW1ylM&e=
_______________________________________________
onap-discuss mailing list
[email protected]<mailto:[email protected]>
https://lists.onap.org/mailman/listinfo/onap-discuss<https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.onap.org_mailman_listinfo_onap-2Ddiscuss&d=DwQFaQ&c=LFYZ-o9_HUMeMTSQicvjIg&r=KgFIQiUJzSC0gUhJaQxg8eC3w16GC3sKgWIcs4iIee0&m=R31H22kxjYwICQ8ZD3yEbIihWWgllI9uTVKv-rEmq_M&s=1xdaNYoF-XWCfc_fUVLiqQGdG44FeNm_001LZk7EDBE&e=>
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
