TSC 2018-03-08
- Kenny Paul
- Hui Deng
- Michael O'Brien
Date
Recording
IRC Minutes
Full IRC Log
Zoom Chat Log
| Agenda Item | Requested by | Notes / Links |
|---|---|---|
| Housekeeping | ||
| CLA Process | Phil Robb | |
| HPA Modeling In ONAP | Alex Vul | |
| Release Status M3 Milestone | ||
VNF package security recomendation from seccom (for endorsement) | Stephen Terrill | not covered - moving to next week |
| Follow-up on project definition review | Stephen Terrill | not covered - moving to next week 2018-01-25 TSC approved to initiate a review of the project definitions - Chair/LF took action to comeback with a proposal on how. Looking forward to the proposal update. |
| CII Badging program follow-up | Stephen Terrill | not covered - moving to next week |
| modeling subcommitte update | Hui Deng | not covered - moving to next week modeling subcommittee made concensus on R2 resource DM and resource IM, need not vote, and workshop info update |
self merged code status | not covered - moving to next week https://lists.onap.org/pipermail/onap-discuss/2018-February/008068.html https://lists.onap.org/pipermail/onap-discuss/2018-February/008181.html https://lists.onap.org/pipermail/onap-tsc/2017-May/002341.html https://lists.onap.org/pipermail/onap-sdnc/2017-August/000052.html Committer Best Practices#BestPractices https://lists.onap.org/pipermail/onap-tsc/2017-May/000476.html | |
docker hub nexus3 repos are inconsistent | not covered - moving to next week docker image names are different between our repos. This is causing a problem when swapping out the repo URL - we need special handling for particular projects example: POLICY-674 - Getting issue details... STATUS policy-db for nexus3 |
Full IRC Log
13:47:12 <kennypaul> #startmeeting tsc-2018-03-08 13:47:12 <collabot`> Meeting started Thu Mar 8 13:47:12 2018 UTC. The chair is kennypaul. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:47:12 <collabot`> Useful Commands: #action #agreed #help #info #idea #link #topic. 13:47:12 <collabot`> The meeting name has been set to 'tsc_2018_03_08' 13:47:39 <kennypaul> #chair phrobb, gildaslanilis, SteveT 13:47:39 <collabot`> Warning: Nick not in channel: gildaslanilis 13:47:39 <collabot`> Current chairs: SteveT gildaslanilis kennypaul phrobb 13:47:57 <kennypaul> #topic rollcall 14:01:27 <gilbert> #info mazin gilbert 14:01:30 <davidsauvageau> #info David Sauvageau, Bell 14:03:06 <SteveT> #info Stephen Terrill, Ericsson 14:03:06 <susana> #info Susana Sabater Vodafone 14:03:10 <xinhuili> #info Xinhui Li, VMware 14:03:13 <frankbrockners> #info Frank Brockners, Cisco 14:03:19 <Zhaoxing_Meng> #info Zhaoxing Meng, ZTE 14:03:34 <kennypaul> #info Chris Donley, Huawei 14:03:34 <JasonHunt> #info Jason Hunt, IBM 14:03:36 <RajeshGadiyar> #info Rajesh GAdiyar Intel 14:03:47 <jamil_> #info jamil for Orange 14:03:48 <RannyHaiby> #info Ranny Haiby, Nokia 14:03:56 <Sanchita> #info proxy Sanchita Pathak, TechMahindra 14:05:33 <Xiaojun> #info Xiaojun Xie, China Telecom 14:08:34 <kennypaul> #topic CLA Process 14:09:48 <kennypaul> #info Phil reviewed restrictions that will be in place 14:09:50 <frankbrockners> quick question on housekeeping: US switches to daylight savings time next week. Though the wiki says that the TSC meeting is at 14:00 UTC. Will the time of the TSC meeting change for folks in the US? 14:10:45 <kennypaul> #info suggestion to enable after code freeze 14:11:12 <kennypaul> #info suggestion to wait until after release 14:11:38 <Sanchita> +1 14:12:36 <kennypaul> #startvote Does the TSC approve enabling thye CLA enforcement after the Beijing release? +1, 0 -1 14:12:36 <collabot`> Begin voting on: Does the TSC approve enabling thye CLA enforcement after the Beijing release? Valid vote options are +1, 0, -1. 14:12:36 <collabot`> Vote using '#vote OPTION'. Only your last vote counts. 14:12:39 <susana> #vote +1 14:12:41 <cdonley> #vote +1 14:12:47 <frankbrockners> #vote +1 14:12:52 <SteveT> #vote +1 14:12:56 <RannyHaiby> #vote +1 14:12:57 <JasonHunt> #vote +1 14:13:01 <RajeshGadiyar> #vote +1 14:13:09 <Zhaoxing_Meng> #vote +1 14:13:10 <xinhuili> #vote +1 14:13:21 <Sanchita> +1 14:13:40 <jamil> #vote +1 14:13:42 <Sanchita> #vote +1 14:13:47 <kennypaul> #endvote 14:13:47 <collabot`> Voted on "Does the TSC approve enabling thye CLA enforcement after the Beijing release?" Results are 14:13:47 <collabot`> +1 (11): susana, jamil, frankbrockners, RajeshGadiyar, JasonHunt, xinhuili, SteveT, cdonley, RannyHaiby, Sanchita, Zhaoxing_Meng 14:14:31 <kennypaul> #agreed the TSC approves enabling CLA enforcement after the Beijing release. 14:14:33 <gilbert> #vote +1 14:15:02 <kennypaul> #topic TSC Composition 14:24:07 <kennypaul> #info discussion of whether LFN Platinum members can appoint reps to the TSC. 14:26:51 <kennypaul> #action kenny work with TSC on new language for the charter regarding "active" TSC member participation. 14:30:46 <phrobb> #action phrobb and kennypaul to start and email vote on when to change the TSC composition 14:31:45 <kennypaul> #info decision on whether to retain the current composition of the TSC structure until after after the Casablanca release will be moved to an email vote. 14:32:21 <kennypaul> #topic HPA Capabilities 14:34:42 <kennypaul> #Alex Vul reviewed his slides 14:35:22 <kennypaul> #info Alex Vul reviewed his slides 15:02:28 <kennypaul> #agreed the TSC accepts the proposed solution. 15:02:57 <kennypaul> #topic M3 Review 15:03:23 <kennypaul> #info Gildas Lanilis reviewed his slides. 15:05:36 <kennypaul> #info action kenny/phil provide info on vulnerability with ODL 15:06:13 <kennypaul> #info No blockers for M3 15:14:03 <kennypaul> #info discussion of SO code merge that is still pending. 15:14:46 <kennypaul> #info AAI scope changed - 15:15:15 <kennypaul> #info will be addressed in Casablanca 15:15:44 <kennypaul> #info DCAEgen2 scope changed. 15:16:22 <kennypaul> #info stretch golas will be descoped 15:16:39 <kennypaul> #info OOF scope changed 15:17:04 <kennypaul> #info vnf scaleout removed from Bejing 15:17:21 <kennypaul> #info SDC scope change 15:17:58 <kennypaul> #info will meet stretch goals 15:18:34 <kennypaul> #info logging - no update 15:18:55 <kennypaul> #info M2 action items closed 15:21:17 <kennypaul> #agreed all projects other than logging and vvp are approved to pass M3 gate. 15:22:28 <kennypaul> #topic VVP inclusion in Beijing 15:24:15 <kennypaul> #info VVP is standalone - ONAP is not dependent upon it 15:24:41 <kennypaul> #info code is packaged as containers. 15:26:08 <kennypaul> #info unclear what S3P requirements would actually apply to VVP 15:27:37 <kennypaul> #info discussion 15:29:03 <kennypaul> #info need to get agreement on what checklist / S3P items apply. 15:31:03 <kennypaul> #info team top meet and discuss - Steven Wright to drive. 15:34:09 <kennypaul> #topic meeting times 15:34:37 <kennypaul> #agreed the TSC meeting will be aligned with UTC 15:35:07 <kennypaul> #action kenny to make time chages for the TSC 15:35:48 <kennypaul> #action kenny check into modifying TSC time for ONS 15:35:55 <kennypaul> #endmeeting
Zoom Chat Log
06:11:02 From Brian : Agree - integration phase time is critical
06:11:11 From Randa Maher : +1 for after release
06:11:13 From Brian : do it at start of Casablanca
06:11:13 From Helen Chen : +1
06:11:32 From Seshu : +1
06:11:51 From Sanchita Pathak : +1
06:12:41 From Tony Hansen : please spell out acronyms
06:12:55 From Zhaoxing : +1
06:18:24 From Lingli : why not?
06:24:17 From Brian : seems like all of the LFN projects will have the same issue
06:24:24 From Brian : opendaylight for example
06:24:53 From Stephen Terrill : At least the ones that are in the "startup phase" the ones that in in the "steady state" phase elect from teh community
06:24:58 From Chris Donley : ODL and OPNFV already switched to meritocratic elections
06:28:59 From Frank Brockners : Minor modification... OPNFV is about to move: Full meritocratic TSC will be in place in September this year
06:29:24 From Stephen Terrill : When does DST start in US? Is that next week? I see in Europe it starts March 25th
06:29:48 From Frank Brockners : Next weekend from what I know
06:29:51 From Helen Chen : 3/11/2018
06:30:04 From Lingli : next week as I was told
06:30:13 From Frank Brockners : This was my earlier quesiton (on TSC): US switches to daylight savings time next week. Though the wiki says that the TSC meeting is at 14:00 UTC. Will the time of the TSC meeting change for folks in the US?
06:30:31 From Frank Brockners : Haven't seen an answer yet
06:30:52 From Chris Donley : Can we move it back to 7 am pacific?
06:31:49 From Frank Brockners : Chris, this would naturally happen if we 14:00 UTC is the TSC meeting time, wouldn't it?
06:32:14 From Chris Donley : yes
06:47:30 From Tina Tsou @Arm : My first thought is to look at how container orchestration deals with this today and make sure we're not reinventing the wheel. Some other thoughts: - The first level of platform awareness is to pick the VNF compiled for the right architecture (Arm, x86) - Some (such as smt, vCPU pinning) are architecture agnostic and could be common - Some are architecture specific (ISA features such as AVX etc.). Surely we don't need to rediscover that list but there is a list known by the Linux kernel. Why maintain that list here vs referencing Linux? - Some of them don't seem to make any sense. For example 'TXT' (which is an Intel security feature). What does it actually mean for the VNF to have a TXT dependence? - Same question for SGX. What is the use-case for technology like SGX in a VNF and how is calling it out separately useful? The security ones cause the most head scratching. If there was a simple thing to push for to make this Arm agnostic I would say it would be: - The list of key/value pairs needs
06:48:40 From Thinh Nguyenphu (Nokia) : HPA support is in ETSI NFV SOL001.
06:48:55 From Tina Tsou @Arm : to be extensible and not hard-coded anywhere, so that the barrier to enabling new platform features is low - The extensible list needs to be split into arch specific (one for each arch) and non arch-specific. Non arch-specific will need to be agreed jointly and the arch specific stuff gets maintained by an appropriate owner. - Ideally there would be a vendor specific list as well Does that make sense?
06:49:21 From Tina Tsou @Arm : SGX could be an optional feature, but perhaps we should be pushing for a set of generic key-value pairs: Security- Feature 1 Security- Feature 2 … Security-Feature N, where some of these are general security types, applicable to both architectures, but then implementations are underneath. In other words, I do not think SGX or any other vendor feature should be mentioned in any top level model specification
06:54:34 From Srini Addepalli (Intel) : Hi Tina, you are right. There is no mention of hardware features in the model. HW feature itself is value for the TOSCA parameter. Hence, it is generic and extensible. Any features invented in future don't require changes to the model.
06:59:53 From Tina Tsou @Arm : Thanks Srini and Alex, I will send more comments after this meeting
07:00:52 From Pamela Dragosh : Please don’t delay M3
07:01:31 From Steven Wright : This consideres the TOSCA DM, buit for BEijing, ONAP must also support HEAT. IS there some impact tehre, or someone working on this ?
07:03:16 From Alexander Vul : HEAT will not support HPA...
07:03:35 From Alexander Vul : the decission was to make HPA work using TOSCA...
07:06:50 From Tina Tsou @Arm : So are you going to change TXT and SGX?
07:08:24 From Stephen Terrill : I looks like I will not get to present the CII Badging program status. PLease have a look at https://wiki.onap.org/download/attachments/25434810/2018-03-08%20CII%20Badging%20program.pptx?version=2&modificationDate=1520435345000&api=v2 as some projects have a bad url to the GERRIT repo for the project.
07:10:47 From Michael O'Brien : logging M3 - https://lf-onap.atlassian.net/wiki/display/DW/Logging+Beijing+-+M3+API+Freeze+Milestone+Checklist
07:15:23 From Seshu : Hi sir
07:15:51 From Stephen Terrill : You can see the overall status with Tony's tool :-) (working to bring this into nexus): http://tlhansen.us/onap/cii.html
07:20:13 From Brian : helen - do we have RCx week1 environments assigned for folks in windriver or Tlab ?
07:20:51 From Helen Chen : At Windriver, they still have their old env
07:21:10 From Helen Chen : We are trying to setup some maturity testing tools for them
07:21:36 From Brian : cool
07:21:48 From Helen Chen : TLAB is ready for them to apply, need to check with Rich how many people are there
07:22:33 From Helen Chen : Beijing is not stable enough for Integration team to do too much. We are playing with Amsterdam MR for Maturity testing
07:23:19 From Helen Chen : testing our tools and measurement (kind of PoC)
07:29:03 From Michael O'Brien : Helen, yes, when running the vFW - I use amsterdam only so far
07:30:32 From Helen Chen : Expecting all projects pass the health check and vFW/vLB before M4
07:32:48 From Brian : i have been wrestling with amsterdam on OOM only
07:32:58 From Brian : those startup race conditions are a pain :(
07:33:12 From Helen Chen : :-(
07:33:37 From Michael O'Brien : pullPolicy will need to be IfNotPresent - jira
07:33:41 From Brian : Micheal are folks fixing that for beijing master ? helm charts seem to let things come up out of order so AAI, SO and SDNC cant talk to dmaap
07:33:51 From Michael O'Brien : oom-722