661303200?pwd=TFdRd0c2MTJUem8xa252UGJHTE1Mdz09

Passcode: 209247

2021 TSC Recordings	2021 TSC Decisions	Anti-Trust Policy

We will start our meetings by mentioning the project's Antitrust Policy, which you can find linked from the LF and project websites. The policy is important where multiple companies, including potential industry competitors, are participating in meetings. Please review and if you have any questions, please contact your company legal counsel. Members of the LF may contact Andrew Updegrove at the firm Gesmer Updegrove LLP, which provides legal counsel to the LF.

Attended	Proxy (w/ @name)	Gov. Holiday	Did Not Attend

Attendance is taken purely upon #info in Zoom Chat

Alla.Goldner	AMDOCS	djhunt	IBM
Andreas Geißler	DT	Lingli Deng proxy Yuanhong Deng	China Mobile
Bin Yang	WindRiver	Murat Turpçu	Turk Telecom
cl664y@att.com	AT&T	Former user (Deleted)	Reliance Jio
Ciaran Johnston	Ericsson	Olivier Phenix	Bell Canada
TBD	Vodafone	Ranny Haiby	Samsung
Dong Wang	China Telecom	Seshu Kumar Mudiganti	Huawei
Eric Debeau	Orange	Srinivasa Addepalli	Intel
Fernando Oliveira	Verizon	Timo Perala proxy damian.nowak	Nokia

Agenda Items

Presented By

Presos/Notes/Links/

Release Status

David McBride

Weekly status

All Istanbul M2 exceptions resolved, except for the following:

DMAAP-1615 - Getting issue details... STATUS

Wasn't on the ArchSub radar. WIP.

RelEng/Infrastructure

Jessica Gonzalez

Jenkins instability

Need to correlate the data collected by both LF-IT and Integration regarding down times
New tools for identifying problems installed this morning - Jenkins restarted

PTL Updates

Logging service

What is the concern?
Chaker Al-Hakim As-is ONAP only does logging that correlates orchestration events between ONAP components
it does not include tracing service level transactions, network transactions except these information are already captured via VES Collectors (DCAE)
4 Types of logs exist today: Types of EELF Logs
- Audit Log
- Metrics Log
- Debug Log
- Error Log
VNFs, CNFs logs are usually offered by the vendors
Next Steps:
- (David) All the ONAP components should redirect their logs to sdtout - to be assessed with the PTLs
- (Security/Archi/Req subcommittees) Define a security log requirement - WIP in progress from Byung-Woo Jun ONAP Next Generation Security & Logging Architecture
- (Archi/Req Subcommittees) Define a new requirement to track service level log to track orchestration information, etc.

PTL Updates

ESR-GUI & ESR-Server

Revisit support model / decision on status
Dependencies:
- CCVPN: esr-gui and esr-server micro-service
- E2E Network Slicing: esr-server
- CloudRegion creation: esr-server

Unmaintained process covered projects and sub-projects - Project State: Unmaintained

Questions:

Do we need to consider ESR as an independent project since these sub-projectss are interacting with more components than A&AI? Or do we need to assess an alternative solution to keep functionalities but handle it differently i.e. move these capabilities in Python SDK?

Next Step: Requirements/Architecture subcommittees + AAI Project Team to discuss with the ESR sub-projects users (Henry Yu, Ahila P, Guangrong Fu , he.peng6@zte.com.cn) if we can re-architecture these functionalities as a mS or within Python SDK, removing the dependency from the AAI project.

PTL Updates

Thanks to Liexiang Yue - stepping down as Multicloud Committer

Congratulations to Konrad Bańka - new Multicloud Committer

Subcommittee Updates

Arch, Ctrl-Loop, Lab, Modeling, Seccom, Requirements

Kenny Paul

Intel Lab Decommissioning

LF-IT looking for a new host location that can also provide the hands-on support services previously provided by Windriver
8X servers being donated by Intel

Intel 6330 Processor 2.0GHz	2
32Gb RDIMM DDR4 Memory	16
2-port 10GBase-T RJ45	1
Intel S4510 240GB SATA	2
Intel P4510 1TB NVME PCIe3 SSD	4

Switch gear is not provided
Rack and cable map dependent on our needs
CALL TO ACTION:
- Need volunteers to provide support for OS, access management, etc.
- Meeting required between community admins and LF-IT for align usage model and configs so that we can get firm quotes for hosting.

Kenny Paulsend email asking for lab subc participation 01 Jul 2021

Task Force Update
CNF, ONAP Enterprise, Wiki 2.0

Next "ONAP For Enterprise" meeting is scheduled on July 7th, 2021. 14:30 UTC

TSC Activities and Deadlines

cl664y@att.com

Integration Leadership role
Honolulu Awards
Action items from June LFN event identified during the daily summaries calls - Plenary: Daily Summaries
TSC 2.0 - kenny still making edits

Upcoming Events & Housekeeping

Kenny Paul

2021 LFN Operations Survey - dates extended until 16 Jul 2021

PaaS Survey Anuker & XGVela
PTL Meeting Cancelled July 5 (holiday in U.S.)
ONESummit - October 11 - 12
Future DTFs
- Nothing for the remainder of 2021
- vDTF event in Jan 2022
- F2F vDTF in June 2022
- LFN YouTube channel - content can always be shared - please contact Ranny Haibyfor more info.

Zoom Chat Log

07:00:11 From Andreas GEISSLER (DT) to Everyone : #info Andreas Geissler, DT
07:00:29 From Ranny HAIBY (Samsung) to Everyone : #info Ranny Haiby,Samsung
07:00:36 From Alla Goldner to Everyone : #info Alla Goldner, Amdocs
07:01:21 From Dong Wang (China Telecom) to Everyone : #info Dong Wang, China Telecom
07:01:46 From Catherine Lefevre to Everyone : #info, Catherine Lefevre, ATT
07:01:54 From Yuanhong Deng to Everyone : #info proxy Yuanhong Deng, China Mobile
07:02:18 From Eric Debeau to Everyone : #info Eric Debeau, Orange
07:02:35 From Ciaran Johnston (Ericsson) to Everyone : #infor Ciaran Johnston, Ericsson
07:02:55 From Jason Hunt to Everyone : #info Jason Hunt, IBM
07:03:06 From Srini Addepalli (Intel) to Everyone : #info Srini Addepalli, Intel
07:03:24 From SaiSeshu (Huawei) to Everyone : #info Seshu, huawei
07:06:21 From Fernando (Fred) Oliveira to Everyone : #info Fred Oliveira, Verizon
07:09:58 From bin.yang@windriver.com to Everyone : #info Bin Yang, Wind River
07:10:27 From Damian Nowak (Nokia) to Everyone : #info Damian Nowak - Proxy Timo Perala, Nokia
07:37:03 From Ciaran Johnston (Ericsson) to Everyone : https://lf-onap.atlassian.net/wiki/pages/viewpage.action?pageId=16472429
08:13:06 From Kenny PAUL (LFN) to Everyone : https://linuxfoundation.surveymonkey.com/r/2021LFNOPS

Zoom auto-transcript service - These are often translated incorrectly and can be misleading. They are NOT Authoritative! Information as to why .
They are included here as a time stamp cross-reference for the recording only! The notes above this line and the actual recordings are authoritative.

07:03:45 Okay, transcript has been turned on.
07:03:49 Recording has been turned on.
07:03:52 And as always, I need to move that little thing.
07:03:58 If you're coming in everybody's muted, please keep yourself muted, and unless you need to speak, please. You're on a dial in you can hit star six.
07:04:20 Again, if you happen to send me a private message it will become part of the public record when I cut and paste the minutes into
07:04:17 cut and paste the zoom, log into the minutes.
07:04:22 And as always, we will start our meeting by mentioning, our anti trust policy and find this link from the elephant project websites policies important, where we got multiple companies including potential industry competitors competing participating in
07:04:34 these meetings, please if you put if you have any questions, contact your company's legal counsel.
07:04:41 Members of the LS can also contact Andy Grove at the firm up Grove LLP, which provides legal counsel to the Linux Foundation says you see that you came in.
07:04:54 Thank you very much.
07:04:57 Okay, stop that share start this share
07:05:09 weekly status has been posted.
07:05:13 We're going to talk a little bit about the Jenkins instability.
07:05:18 I'm talking about logging Asr, I want to spend a few minutes talking about the Intel lab decommissioning
07:05:29 enterprise deadlines and upcoming events and housekeeping anything else that needs to be added.
07:05:44 Okay.
07:05:46 David any comments about the
07:05:50 weekly status, other than it's been posted.
07:05:55 Yeah, Kenny, I just wanted to point out that today was the due date for resolving exceptions associated with stumble milestone to.
07:06:09 You'll recall it the TASC approved that milestone with exceptions, and at the time I think we had seven remaining tasks, release management tasks that needed to be completed.
07:06:24 So we're down to one. And I was, could you open up that one issue.
07:06:33 Please.
07:06:42 Oh, I think if you just click on it, it'll give you a link.
07:06:49 Okay, so this is a.
07:06:59 So I was little puzzled by this one because my understanding was that all of the. The remaining architectural reviews, we're going to wrap up on the 29th.
07:07:05 Two days ago.
07:07:07 But this one, remain open, and it can't. can you scroll down and see if
07:07:15 if we got any comments, so I was hoping that Shukor or Shakira would respond to this
07:07:27 Shakur Are you on the, on the call.
07:07:34 Are you okay, Yeah I flagged you in this, in this issue. Just trying to understand. Do we still have some architectural reviews that need to be completed.
07:07:49 David, I sent the, the weekly report on either Wednesday or Thursday and I just was sent the report.
07:08:13 This morning, all the review completed.
07:08:04 Okay, and they they tracked on weekly basis.
07:08:09 So, um, so as far as the architecture subcommittee is concerned, all the reviews at least component reviews for the nine release have been successfully completed so we're, we're done.
07:08:24 So is there a reason that the that the subcommittee didn't close the on a park issue,
07:08:32 because I wasn't aware of it.
07:08:35 Right, so, so I think what we need to do moving forward, is because we have so many things that we're tracking.
07:08:45 When I send the report, the assumption is that the report is being looked at and status or being updated accordingly.
07:08:58 So, so let's make sure that we understand, because I don't want to be looking at. Too many issues that are open that are being tracked and at the same time I'm sending status out, telling everybody what the statuses of all the architectural reviews.
07:09:12 So let's make sure that we understand how we want to track it moving forward, so that we don't end up with this gap where status reports are being sent but no one is actually looking at.
07:09:25 So for the purposes of release tracking, I use a release management tasks to track the architectural reviews and we linked on a park issue to that review so that if I see that, if that has been closed, then either myself or the PTO can close the related
07:09:46 release management task. Okay.
07:09:51 I mean, we just need to understand the process so we don't end up in this area of, you know, uncertainties, when, when things are have been closed.
07:10:03 And yet the proper boxes if you will have not been checked.
07:10:09 Kenny Can you open on a park issue for a minute, please.
07:10:20 So,
07:10:23 so that I mean, the architectural subcommittee also uses these JIRA issues to track these Right, exactly. Exactly.
07:10:30 So it seems like like part of your process would be just to do a simple query for issues that are assigned to the stumble release that haven't been closed that are their own a park issues.
07:10:46 I'm fine just.
07:10:49 Let's see what did we.
07:10:52 That's the link to the email. Okay.
07:11:00 Yeah, so I think it's just that.
07:11:04 This one was still open.
07:11:08 And those are the queries that the David runs, I think.
07:11:15 Yeah, I mean, this is, this gets linked to the release management task and so I if I see that the the architectural subcommittee has closed this itch issue, which presumably the architectural subcommittee is using to track architectural reviews, then,
07:11:34 then I know that it's been done.
07:11:37 But I, I went looked at this issue and there weren't any it wasn't closed and there weren't any comments or anything so I'm not sure what you know what's going on.
07:11:47 So, so So bottom line, it's been done it just needs to be closed right.
07:11:52 I'm not sure that's true, this, this one is this one for some reason it's not ringing the bell.
07:12:03 even though it's.
07:12:06 Let me take a look at this one.
07:12:08 Okay.
07:12:16 For some reason.
07:12:19 Yeah, well I, you know, I mean not to put you on the spot shocker but I see that a it's within the it's part of the architectural subcommittee JIRA project, it's assigned to you, it's assigned to this stumble release.
07:12:38 So it seems like you know part of your process would be just to do a query on those issues and to see what what's there what's open. No no I do that for this one is over here, this is the one that I'm that I'm missing I'm not sure why I'm missing it.
07:12:58 Okay, so something must be yeah some reports or something yeah this was overlooked and you don't know why but that's it.
07:13:03 going on, and maybe coordinate with Viagra, and maybe you know drop me a note or update the JIRA issue.
07:13:23 Just so I understand what we need to do.
07:13:28 Yeah, that's fine, I'll take a look at it. Okay processor.
07:13:31 Yeah, I'll take a look at.
07:13:32 Okay. All right.
07:13:34 Thank you so anyway. The upshot is that, that all of them two exceptions are resolved except for this one issue regarding an architectural review and and Chicago is going to get that sorted out.
07:13:51 So, and then the remainder of the.
07:13:56 release status, you can check the weekly status report.
07:14:01 I think scanning.
07:14:02 I noticed that that one's a task, and not a story, so maybe that's why you missed it, I don't know, like all the other ones are stories, it looks like, but that one's a task.
07:14:13 I don't know if you feel it that's absolutely i mean that's what I'm looking at right now, right. So, so thanks for for the comment but let me take a look at it for some reason this one.
07:14:27 This one is not I'm looking at my list, it's not on my screen but that's fine. Let me, let me let's let me work offline.
07:14:35 Yeah, that could that could be the maybe the issue. Okay.
07:14:40 Cool. Thank you.
07:14:43 Anything else release related.
07:14:49 No, as I said, for the remainder of the release I I emailed the weekly status report yesterday and it's also linked there and the agenda.
07:14:57 Okay.
07:14:59 Okey dokey
07:15:05 Jenkins instability.
07:15:06 This has been kind of an ongoing issue.
07:15:13 As part of this.
07:15:17 Morgan gave a very good presentation yesterday at the tack meeting, just like on on IT support kind of in general.
07:15:27 But we need to basically get the data being correlated by everybody to figure out what this is because it's it's an intermittent issue and not sure why.
07:15:42 Jessica this morning you put in some new tools, some new hooks for monitoring. Yeah. So I installed the tools yesterday, and that's why I did the restart yesterday, um, We need, I need to give it a day or two to collect more data.
07:16:06 But basically I had a conversation with Vanessa from my team and they did something similar when FDA was facing the same certain and stability issues, and that they, they even had to do anything with loading more jobs into the system or doing something
07:16:24 different in in the in the in the tooling itself it just started having a stability so when they install these monitoring tools they discover few in their case, I'm not saying that this is our case but in their case they discover a few plugins that the
07:16:43 newest version of the plugins was causing some memory leaks, for example, and it can it can help us investigative our case is similar. At least, is it can you give us
07:16:59 a root cause for what why this is happening.
07:17:02 But yeah, I'm gonna give it a day or two to collect more data and look at the monitoring results and see if I can spot anything strange and let you guys know.
07:17:16 And then, yeah, if we can correlate that with what the integration team has in place for monitoring.
07:17:26 We might be able to track it down Catherine I, and David were talking yesterday that it could be a situation where there's some other job that we are aware of.
07:17:43 That's also getting triggered simultaneously.
07:17:48 It's getting kicked off somewhere that's that's causing high usage or something I, you know, no idea but that.
07:17:56 Yeah, so hopefully this monitoring tool can tell us some of that too.
07:18:02 Yes, definitely. It will help us find out a little bit better. The root cause.
07:18:10 Why because one of the things that I thought was interesting.
07:18:17 I'm
07:18:17 looking at this is.
07:18:20 We got a couple of these big spikes
07:18:25 in and and this doesn't seem to correspond necessarily looking at the, the times it doesn't necessarily correspond to where I would expect there to be a lot of release work taking place anyway so trying to figure that out will be will be a good thing.
07:18:44 I guess I'll let you guys know if I find anything suspicious.
07:18:53 Any other questions or thoughts on that.
07:19:05 Okay, moving on to logging service.
07:19:16 So the big question here is,
07:19:24 what exactly
07:19:28 is the concern that the community has with what is currently in place for logging.
07:19:44 Just speak up.
07:19:46 Last week we were discussing that, indeed, the logging project bomba. We are provided to see sometimes ago to be moved to maintain.
07:19:58 They are still to Java libraries, and also one Python libraries that are consumed by some component to, to provide some to comply to specification which were developed by logging team, and which is used.
07:20:18 These are maintained we got the confirmation that the three libraries doing Java one invite don't continue to be, maintain, but what it is not clear it's what what is the next step that the committee would like to do.
07:20:35 Considering these two light, these three libraries.
07:20:39 And what will be the, the new announcement which is expected to, but what, what is the purpose of
07:20:50 the new logging service so whatever. So can you, can somebody
07:20:56 could bring additional clarification.
07:21:02 Hey Kevin busy shocker So, if I may.
07:21:06 I do want to clarify the point that I made last week.
07:21:10 Regarding logging. And I just want to make sure it could be just me, but I just want to make sure that the.
07:21:17 The point is, at least my point is well understood.
07:21:21 When we're talking about logging within own app. We are only talking about logging in correlating the orchestration events between all the own app components.
07:21:38 However, though. What we're not talking about is correlating the actions at the service level.
07:21:49 And this is critical. So for example, If you have a service.
07:21:54 And that service consists of load balancer, a firewall and a database.
07:22:04 The logging that we're talking about here will allow us to correlate the orchestration event, the orchestrating events for.
07:22:17 When the firewall was orchestrated when the load balancer world was orchestrated, and when the database was orchestrated. However, it's not going to correlate, if a firewall rule was to change on the service that was just orchestrated.
07:22:43 So this is a critical, this is a key distinction between being able to correlate a transaction at the service level, versus in login versus logging, a correlate in an event, an orchestration event at the orchestration level.
07:23:08 So, so, yeah. My concern was, and I don't want to speak for me and I hope Amy is on a. My concern is that the. There isn't.
07:23:20 There is a, an implied assumption that own app is going to orchestrate to log and correlate at the service level
07:23:39 was discussion here about logging of actions that have been performed towards the network as opposed to logging generally the, the number of logs that each container and know enough purchases.
07:23:52 So it's more like an audit log as a chapter.
07:23:55 So, so I think in this is what this is the point that at least I'm trying to clarify my point. Right. Me, from a from an owner perspective.
07:24:06 You know, I think what we, what we can do is to log when the, the recipe gets executed within sou log the fact that you updated. I'm just making up an example, the ANA I was updated that SDN controller received a request to orchestrate all the resources
07:24:29 networking resources. These are all the events that get correlated.
07:24:35 And then, now you have the service up and running. But once that services up and running in the service provider sends a provisioning request to that service own app is no longer part of the equation.
07:24:56 Yes, is what you're saying here that we that that own app itself is not a log management system for the services that it orchestrates. Exactly.
07:25:08 It's not yet, right now. I don't know if we want to make it because I don't know if I would ever know.
07:25:16 The or have access to a transaction. the example that I just gave you.
07:25:22 If a firewall in update to firewall rule or sent to the firewall own app is out of the loop at that point.
07:25:37 I don't know that it may know that it sent a request but it doesn't know what the firewall that so it has no idea.
07:25:41 I just wanted to make sure that what you were, what you're saying is your point is that you don't see an app as being a log management system for the services that it orchestrates.
07:25:52 However, it must, it must be collecting and doing login correctly for itself for its own components and a ID map so on and so forth.
07:26:07 Equally, it would do that right in. Also, it would log, the events that trigger the orchestration, right, for example, the example that I made earlier would be one example, but it would also log and correlate events that are triggered by the collector,
07:26:27 the micro service analytics in the policy. And going back to.
07:26:34 So, and the SDN controller these events would be orchestrated would be correlated.
07:26:40 But in relative to the orchestration of a, of a service. However, the service level transaction that gets sent to that service owner. It does not have any exposure to it.
07:27:05 If you you you check already the different log checker to confirm that because no money, at least on the Java side, I cannot confirm on the Python site.
07:27:20 They are already full of different existing lock with different couples we have the audit lock the matrix clock. The clock and you have log and I put the reference here, under the under what you at share.
07:27:38 So, there is a unique coalition ID, which is transmitted from one component to another component so we can trace a flow.
07:27:48 When we have a particular event across multiple components we have at least one unique coalition ID. So did you already look at what it is, offer so we can notify the gap, especially the audit log maybe.
07:28:08 So, so I have not looked at in great detail.
07:28:13 This is me. But what I'm trying to clarify, is the understanding.
07:28:19 So right.
07:28:22 Yeah, so, so what I can say is the logging service within own up was not used whatnot. It was not planned or is that never planned to be an audit or a logging service of the, of the, let's say, and deployed services or instantiated services outside like
07:28:41 a firewall or something like that it was only for logging internal logging of the owner components themselves storing the locks and as Catherine explained, they are have been enhancements done like this kind of correlation ID so that you can combine the
07:28:58 locks together when the, the whatever as an SL call is received, it started and it will forward then the call to the CMC. So this kind of transaction log or ID is then used everywhere so it can be correlated together but it's still within the owner parts,
07:29:18 and the owner services is only within the owner component, not, it's not logging anything from the, from the network. Yeah, so this is part of another story which is not part in this logging service here.
07:29:33 So I think that we have to understand I think this is more.
07:29:37 This is really only internal logging in internal auditing, of the, of the the the the stuff which is ongoing within own app itself, not about the services which are instantiated.
07:29:49 So this is something completely different. Yeah, so that would be new is usually done by the operator themselves. And then in connecting them to the services and getting the logs from there, or getting events usually so so best events or something like
07:30:04 that which they can store, and put a process but not in that way that this logging service is not used for that.
07:30:12 That's right.
07:30:17 Okay. There is no free with underlying No, understand better, but do we not use the west to collect such kind of information. Yes, of course. Yeah, you get events like transaction locks or something like events you receive from the network Yeah, from
07:30:34 the net elements. So, if you would like to, to, usually you would need to use them, or to tell the network element or DCCNFOVNFQF you have launched, please send me all the blocks or something by adverse events or something.
07:30:53 Yeah, that's usually done and this is them collected by the CIE and then can be processed can be of course then in an operator case story stored in some some data lake or whatever, post processing or it for for logging.
07:31:08 It is different this is not part of owner.
07:31:12 But it's not part of the logging. Right, it's not part of the data collection that part of the logging service. Yeah, it's a data collection issue.
07:31:21 But doing me.
07:31:25 In, because we will have hidden information, because we already capturing that that divest via DC as a KPI. Right. So we are capturing even Do we really need to do have that.
07:31:39 I'm just asking because for the moment I don't see the value but it's not because I don't see it it's not important for you. So, do we really need to capture in the locker because Don't, don't forget, each time we write something in the locker.
07:31:54 It takes some, some bandwidth as well.
07:31:59 But there is a there is at least in my mind, though, Catherine, there is a clear differences between logging orchestration events, vs collecting
07:32:17 metrics from the actual VNF.
07:32:23 And, and having a micro service a dedicated micro service or set of micro services within own app to actually analyze the data and provide a yet another level of clarity to the service provider.
07:32:40 But that's a custom data collection and analytics components, not a logging component
07:33:07 The dean for me is due for no particular reason so again if you believe we should duplicate the information in the log, then we will need to create a new logo.
07:33:06 But typically, it requires some
07:33:10 equipment and architecture discussion right because I don't think we should speak so the architecture and requirement of these new requirement here
07:33:22 is the team believe it is something useful in addition to the west, why not.
07:33:29 We just need to be clear if we have this new requirement. We need to define the paper will implement the new requirement.
07:33:39 So, in my opinion, So, I mean we don't, for example, we are also not using the logging service let's say the database for example of the logging service or something like that.
07:33:52 In our case, of course we are collecting and that was the issue that every, so the target that every pot for examples of own up but I'm just talking about the owner pods are logging on the standard out so that we have, let's say kind of whatever default.
07:34:10 Let's say mechanism, we have in our environment to fetch the logs from the Kubernetes pots and store them in our own monitoring system. So we don't need any, let's say support for logging database within own app, because this is usually something which
07:34:29 has the service provider already has or it has to put it is it's not another edition of feature of owner, I would say so, the biggest point is, are the important point is that all the output, all the components are logging, to the to their to their standard
07:34:49 out lock, and this is I think implemented in most of the cases already and including then the right or cover kind of common locks drag or lock event information you receive, including this kind of transaction ID or request ID, which was implemented by
07:35:14 Tech Mahindra most in most cases, because we were also requesting this and, and they have implemented it.
07:35:17 And I think that's that's that's the that's the important part I think we don't need on logging, in my opinion, are really an integrated logging service because this is usually addicted to this is usually available in or data lake function or something
07:35:34 like a monitoring system which is part of for example the cloud infrastructure and then we can, we pass still only the the events then or the logging information to that one.
07:35:45 I don't think that it's that it's needed to have, let's say get kind of logging database, our specific component for that. The other thing is the logging or the information which we received from the, from the, from the VM SOC and apps or something like
07:35:59 that. And as I said, so this is usually done either by the by the events service events in my opinion.
07:36:10 Or, there are mechanisms also provided by the vendor to which he retrieve logging information from out the DNS but I think I don't know whether you just make sense to also add this then as a new functionality within own up, because I think that's the
07:36:28 standard way to do.
07:36:31 It's my opinion on the next step Sarah, I know, young from Erickson has been working on a page looking at security and logging architecture, and I think that lines up pretty much with that description there of all microservices log into standard out standard
07:36:48 error.
07:36:49 I can share the page here it's a work in progress at the minute but it goes into a little bit of detail on that post in the chat.
07:37:06 Based on what you say and again I'm.
07:37:09 If there is a need, there is a need right but if we define something at one point we need to implement. So, the, the security log.
07:37:18 It is too I've seen some,
07:37:22 some discussion about that so we just if Can you can base because I'm, I'm using my iPad.
07:37:31 And I'm using my laptop to take some notes so it's a little bit tricky, tricky, but you can add the link provided by Qian so the only thing that we might miss.
07:37:50 So far, except if somebody believe we, I mean, check it out just talk about it I bow that we can always define the government, but the only thing is maybe a security log, potentially, that's what we might want to add to the suite, and the team is already
07:38:04 working on it right.
07:38:10 So, the point I was trying to make I wasn't I wasn't trying to take a position one way or the other. I was trying to clarify the point that I made last week, regarding the logging function with the in own app in the, in the, in the benefits or the service
07:38:34 that provides right versus logging, the transaction or the transactions that actually are executed on the Vmf steps, set themselves.
07:38:53 The service level tracing if you will, versus the orchestration level tracing and logging, and they're different.
07:39:04 They're very different.
07:39:11 I'm sensing that there is a need for a, an application or a service level tracing, so that the service providers can conceivably trace a transaction or an update or delete whatever that transaction is across the service in many components or vn F's that
07:39:36 basically form that service.
07:39:39 And that's different that's outside the scope of own app, but I'm saying, I'm not saying that we, we couldn't attend to look into it but that's not what the login function that exists today or at least existed.
07:39:52 That's not the, those are not the functionality that provided that just logged, the orchestration transactions.
07:40:50 Hello.
07:40:53 Chris I was speaking on mute. Yeah, no, sorry.
07:41:00 So, if, if there is a need, I would suggest that the equipment architecture committee sit down, define the needs formalize the needs and make a presentation to the Jesse when the
07:41:19 low level detail about what we should do in any could be discussed, and the product team.
07:41:26 If people wants to implement it.
07:42:13 So it could be a new requirement.
07:42:20 Okay.
07:42:23 I think we would need to definitely address it as a requirement and have the requirements of committee review it and then
07:42:36 we, we certainly could take a look at it from an architectural subcommittee perspective and beyond is looking at it.
07:42:44 Like, Karen just mentioned.
07:42:46 So, okay, I'm sorry.
07:42:49 So, so it is, it is. It has to be driven by a set of requirements to clarify the login function within own app, versus the tracing function at the service level, which are two different functions.
07:43:15 I really don't know at this point in time I didn't know if the latter which is the tracing is more of a data collection analytics function versus like infant login function.
07:43:40 No no no comment no problem at all. So
07:43:47 I don't know if any other RTC members have different opinion but there is no new arm to define new requirement, new challenge will be to find people to equal one.
07:44:00 Yes, yes, yes. Yeah.
07:44:04 Okay. Anybody else wants to add anything, based on the discussion.
07:44:15 There is.
07:44:18 So maybe I need next step
07:44:22 we need to see so that's something it's.
07:44:26 I would put a no no David because we need to check that the PDL call here, its security team is working on it, security team security team put in market the ownership when she, like
07:44:58 we need the requirement team should also be in for my boat.
07:45:07 So everybody's welcome connected, so we can define the requirement and then it should be discussed with a different project team.
07:45:15 Okay.
07:45:17 So, this is only to check where we are on this requirement David that's why I suggest we talked to the PDL. The two other items are something that the subcommittee's need to lead first before we present to the BTR team first.
07:45:38 Okay.
07:45:40 If everything is good.
07:45:43 I would suggest that we move to the next items.
07:45:47 So, nearly one year ago, it was July 9. We took a decision after sending a final call to the committee I guess that maybe the ESRI or the UI, and the ASR server these two super deal of part of the ANSI codec umbrella, there was a decision from the GC
07:46:11 to move them under unmaintained because we didn't get any feedback on it.
07:46:18 The subject has been recently been really open because we never implement the decision for different reasons so nobody should be upset about that. But we get feedback from the community that ESL UI and the SF server I indeed cushion for cc VPN for, or
07:46:41 end to end network slicing and also called the blue the region as well.
07:46:46 So based on the feedback, I would like to be open to discussion with DSC and requests on behalf of all this people to undo the decision to move these two components to maintain.
07:47:02 are still, I understand, kilometers on this component existing, it means, it is still maintained even if there are not a lot of activities by Pete by people from the community.
07:47:23 So the request is to remove the decision double A maintain or change the unmaintained stages to incubation because I don't think when we did the project lifecycle, this component where that as much as well, so there.
07:47:44 There's where there's a lot of confusion because there, we don't have a life cycle for or subcomponents
07:47:55 I think that's one of the, one of the challenges we need to address.
07:48:02 This could be broken out, as I said, it was originally proposed as a separate project.
07:48:08 The decision was that should be under API.
07:48:13 So that's where it, it lives in in that context, then the competitors and the ppl have the jurisdiction to decide what's going to be managed or not.
07:48:30 The two repositories in question are all are in read only
07:48:40 right now so whatever is there is, is there.
07:48:48 How the TMC would like to handle this doesn't matter to me.
07:48:53 But just.
07:48:57 We have a bad.
07:49:00 I think we have a bad history of talking about subcomponents as if it as if they are a project and expecting the same level of governance
07:49:13 need to open the link of the project on maintain, you will see that the position was covering project and the project.
07:49:27 So we cover from a process perspective.
07:49:40 Using few Google in the definition and maintain project of the project is no part of the official entities, is you go up to the definition of unmaintained.
07:49:53 You go up us caught up Yes, stop stop stop so project of projects. So, the process was covering also see projects.
07:50:09 And that's why we had this discussion about ESRI Asr, because he is in London is a city project so from a process perspective, I'm not too much worry.
07:50:22 And since it is acting, since it is already documented.
07:50:28 What we want to do is the transition from and maintain to incubation and mature.
07:50:35 That's, that's what we need to confirm with the SEC scene, it was stuck as a maintain or do not implemented a search. We need to activate the transition would you define into section three and position for unmaintained want to incubation nature.
07:50:56 I'm transition from unmaintained to incubation mature. That's what the process we need to do.
07:51:05 So, you know, we, we did make put this into unmaintained read only over a year ago.
07:51:15 And at that time it didn't seem like anyone was stepping up no one's committed to this in over a year.
07:51:22 And I see now that you know some folks are saying that it's still needed in the deployment.
07:51:26 But what does that mean like.
07:51:29 So there's a lot of things that go around maintaining a micro service, and at the I already has a lot, right.
07:51:38 So, or what we're saying is, now we have to deal with everything and this even though, you know, to me, and the people on the bridge, every week. No one ever talks about this.
07:51:50 Right.
07:51:53 It, I guess it's just there for this one flow that it's just been there for a while and, you know, we don't have the expertise on these to even know what to do with them.
07:52:03 So,
07:52:07 I'm just trying to understand what's being said here.
07:52:12 So I think the problem is of course that we have a an AI object, but it's not the only one, let's say, which has no user interface and the server does not only the addition to the API, but it's also triggering some registration Event Registration things
07:52:34 on the multi cloud for example.
07:52:37 Yeah, so the. And these use cases have used them in their workflows, or in the use case flows tool for adding then, for example, VNF managers or the cloud region.
07:52:56 so the region where the vn F's get installed.
07:53:01 Therefore, they use the, I think the is our user interface and the server behind.
07:53:08 Although I think, for example, in our case, we are using for the cloud region instance creation, the Python SDK and we using that as a script. Yeah, so we have.
07:53:32 We also indeed it will create let's say for the graduation and integration in our case for our use case also a user interface, but it's not. Let's say it's just for our eternal purpose but generally we, you can use for example for cloud region installation
07:53:43 or integration.
07:53:45 The, the, the Python SDK for doing that in an easier way, you can of course do it always with, with some some commands anyway, so that would be possible to replace the functionality.
07:53:57 So my suggestion was okay, the cloud region installation, including the registry command to the multi cloud is already implemented in Python SDK so my idea would have been a couple of the other ones which is the DNS manager at GSDND.
07:54:18 So the SDN see add on
07:54:22 a possibility would be to add this functionality in the Python SDK as well. And then, if a use case, like the network slicing or whatever, cc VPN would use that they can use the Python SDK and use a script for installing and basically replacing the is
07:54:43 our server functionality or with the with the Python SDK call.
07:54:49 That was my, my, my suggestion. Instead of having a user interface which is not well maintained and it's also it was a bit buggy, I would say.
07:55:02 But, of course, it does the work as a to do.
07:55:05 So then this is as as as already said so, providing them the server and degree would require then of course, held jobs that would require the, the containers and so on, images, It could be replaced by by by fantastic.
07:55:27 Right, so I think it's mainly very connected to what undress it so it's mainly about maintenance of the solution later on right because all of those global requirements applicable for all code.
07:55:39 Become a peak of all so for us our and essentially. Someone has to be found. To do this, simple as that. Right.
07:55:51 And we representative on the team on the call.
07:56:04 So I know they are amazing.
07:56:06 China so that's probably why they need not attend the call, it's already 11 o'clock in China.
07:56:17 But I believe we need to discuss with the team about to understand their plan because if it if the school is beyond the nei.
07:56:32 Maybe we need either to consider it separately from a an AI, or either way, we need to keep the functionality and under the default the suggested by on player.
07:56:40 Right. I mean the topic is, I would probably support what Andreas proposed with partners decay because it looks like this Yes or server my understanding was that it is working as the kind of presentation layer presentation would be really the UI but okay
07:56:56 okay something that allows us to present the DNA object in more user friendly way, right, but now it looks like with any service well some. Let's call it micro workflows right if we have a couple ability to add these things, and then Okay so that is so
07:57:22 What do you have in the object. Now, question is if this would be the right place. In general, to be such micro orchestrator, right.
07:57:29 So either it is a separate product for me. And then it all of those requirements that we are putting to each and every product in, you know, and this would be applicable right release notes, maintenance, security updates.
07:57:45 Right. I mean, Certainly some of that.
07:57:46 Or, or the proposal to move this to Python SDK.
07:57:51 That might be a good proposal, right, because we probably have more adoption, and more potential contributors to Python SDK. Right.
07:57:59 Okay, my two cents.
07:58:47 So, I'm, I'm sorry, so I was on mute. So you don't hear my keyboard so that's what I'm understand from the team I'm just trying to find the, the owner login, because I don't know by hard.
07:59:01 So at least it will remove the dependency on nei because we we have more users than just having in and of ANSI.
07:59:13 And I think we have the, it's only
07:59:18 in also Ali have been
07:59:28 told it seems that we should remove the dependency on on on the on
07:59:37 me, based on what I'm hearing from the team cannot find the.
07:59:47 So Catherine Where are you getting the information on the
07:59:54 phone the email. The email exchange.
08:00:00 These are the two people who step up and said the only fly to the TASC that it is using us sitting used.
08:00:18 Okay, well I don't know that. Does that qualify them as that.
08:00:24 When you say sub project team you're talking about the ASR team. Yes, yes, yes.
08:00:30 Okay, well there there's, there's a difference between saying something is being a component is being used, and being the people that actually supported the dear of the people who are supporting it based on their response.
08:00:45 Right, well there's no, the repositories have been locked for over a year so no one is supporting it.
08:00:55 I understand what I mean.
08:00:59 We need to start with these people who are using it.
08:01:03 Right now, I think, referring to them as users, rather than name I don't I'm concerned about assigning response ability, I think it's aspirational Guinea.
08:01:23 The user if you're if you're using it, then potentially supporter.
08:01:44 Yeah, we, we were right now they could tell us the pound sand but it like Catherine says it's a good place to start.
08:01:53 Yeah.
08:01:56 So I'm adding more name.
08:02:06 I cannot find so
08:02:10 we need to start also more white.
08:02:15 So that's, I believe what we should do except based on the feedback. Again, if, if, what I've talked is not the right way to do.
08:02:25 The team can raise their voice and suggest differently.
08:02:30 But I think we need to set up a meeting with, with these folks I could not find them to see what we are doing. Yeah.
08:02:43 Make sense, is it a fair statement. So people have a different idea about how to handle it.
08:02:57 And we should also include
08:03:05 the team because they are impacted by that for you. That's why we have been one is the problem.
08:03:14 Okay.
08:03:16 Okay. Agree for the next step, anybody wants to add anything, make sense.
08:03:24 Okay.
08:03:28 So, there is no commendable that we we move with that, and we when we be back from requirement and architecture, team accordingly.
08:03:37 Some additional update.
08:03:39 We have leaving you was previously one over committed from the multi cloud and being young informers this week. The best one is stepping down so just wanted to thanks for all the contribution to the mythical project.
08:03:56 And I want to congratulate command I don't know if come had it on the bridge. But I want to congratulate come cohort of new multi cloud competitors.
08:04:07 So that's what the announcements, and I give back the floor to Kenny, to give us an update about the interlock.
08:04:16 Okay, thanks Catherine.
08:04:19 So, Intel has.
08:04:23 So just as a quick recap here.
08:04:29 Intel announced that they were going to shut down the lab that we are currently using.
08:04:36 And they would not be setting up something, anything new there.
08:04:41 So this is the one that that Wind River has been providing us.
08:04:46 Hands on support for since the early days of iPhone app.
08:04:51 Anyways, they Intel has committed to donating eight servers with the configuration shown there.
08:05:01 And what else it is doing right now is looking for a host location.
08:05:08 that can not only, you know,
08:05:10 give the hardware at home, but also perform Brock stack cable label ongoing hands on administration.
08:05:23 Generally when you move something to a colo.
08:05:28 The company that's that's renting the space and the coal, oil is also providing the hands on staff to do the work. In this case, we do not have that.
08:05:39 Therefore, there's going to be some incremental cost here, even though we're getting the hardware. Also we need to figure out the switch gear that's, that's going to be needed and get that purchased.
08:05:52 So, getting this all set up and getting the quotes actually requires
08:06:00 the cable mapping and in things that are necessary there.
08:06:06 We need some members of the community to step up because as near as I can tell the lab subcommittee is basically dead.
08:06:17 So this would be appropriate for revitalizing the loves that loves the evil, the lap, lap.
08:06:31 Start over.
08:06:35 Lab subcommittee.
08:06:39 So that we've got some folks that can provide the type of access and support responsibilities that the integration team has been handling.
08:06:50 We need a meeting between, whoever the community admins are going to be an L fit and figure out the exact usage model as that will dictate to some extent, the configuration that we end up with.
08:07:08 So we're in a bit of a, of a block situation here, in terms of us being able to make progress, and get the setup.
08:07:21 Intel has graciously allowed us to keep the lab in place until we can have this new lab set up.
08:07:31 But the assumption there was that the lat, the new lab would be set up within one quarter.
08:07:39 And we are pushing that, and now So, looking for folks to step up and revitalize the lab subcommittee.
08:07:55 I can send a mail out on that, unless there are folks that are willing to raise their hand here today on that.
08:08:11 Okay, I will send out an email just Can you just give me for clarification, does it doesn't mean that Intel will not provide any, any support for. That is correct.
08:08:25 Okay, you're providing the hardware.
08:08:28 Okay.
08:08:31 Okay. But that means we need at least some people from Intel because if we want to manage remotely this outerwear we need some people to get basic information to access the Sarah, how to access them and so on so we need at least someone from internet.
08:08:47 To get access to these several so.
08:08:52 Well, I mean that they're, they're not. I don't think the hardware they're delivering is configured. I think it's Yes.
08:09:01 I think it's just gear pure and simple. But yes, we will when we arrange this meeting we will have people from their sides.
08:09:12 Okay, on the call.
08:09:15 But going, but there's no expectation that they'll be providing any sort of administrative support in any way shape or fashion going forward.
08:09:26 Okay.
08:09:29 Thanks for the clarification.
08:09:35 Okay.
08:09:43 So can, as you said, I think it would be good to send also an email because I don't believe people will reply states.
08:09:53 But I would suggest that maybe we also send an email like you're just written.
08:09:59 And people will will probably look at it as one.
08:10:03 So no opportunity to use some internship at least to set up the lab, as a staff. Yes, so we see.
08:10:19 Okay.
08:10:22 Thank you for the update.
08:10:25 No problem.
08:10:27 Love to get this one,
08:10:31 get this one, put put to bed.
08:10:36 I'm enterprise Task Force next meeting will be on the seventh
08:10:42 or as activities coming up.
08:10:47 Integration leadership role any updates on that Catherine.
08:10:53 No, I just wanted to add it here on the topics to be sure that it's on our radar, but unfortunately have been it jackets on a couple of things.
08:11:04 And in evidence to work on it.
08:11:14 No, I'm okay. So that's still on my plate to start to get organized and to sit down with the team, but I recognize I didn't focus a lot since more than a week.
08:11:18 to, to solve on top of the fact that we have two carriers who have agreed to work together to support the team from a leadership perspective.
08:11:34 Thank you.
08:11:42 Also on the to do list. Honolulu awards. Talked about any action items resulting from the June event.
08:11:48 Last week, don't want to seem to recall any.
08:11:54 If there is no
08:11:58 nothing near then then we will let when I would play that you keep it because I know from the daily minutes, we have, we have joined on device or action item I don't think they aren't necessarily all for the DC.
08:12:15 But again, I didn't do my due diligence.
08:12:19 But it's related to the daily minutes and the action item that I've been capture.
08:12:24 So leave it for there, because I want to be sure there is no thing really striking that we need to implement so I want to do the due diligence on that, if you don't mind.
08:12:36 Okay.
08:12:37 And yes, I am still making edits to the PSC to Dotto document. My apologies.
08:12:52 Okay, moving on.
08:12:54 The elephant operations survey.
08:13:00 I want to copy the link, and I want to paste it into the chat window.
08:13:08 This is not the event survey.
08:13:13 This is our annual survey of elephant operations. This is what we use to make budget decisions. This is what we use for the LFN governing board to make strategy decisions, and things of that nature.
08:13:31 Out of the thousands of people that we have in the community.
08:13:37 57 people have filled out the survey.
08:13:54 So that's not a statistically significant number of folks for any decision making, given the breadth of our community, no decision making could be based upon that.
08:13:55 So, We have extended the date for two more weeks.
08:14:02 Please fill this out. This will directly impact.
08:14:07 What LFN does and how we budget, and will have direct impact on.
08:14:17 Now all y'all are able to contribute and things like that.
08:14:22 So, I can't stress the importance of that survey enough.
08:14:28 It Kenny. Yep, it's sincere, just to reinforce it in case anybody you know, is it listening to Kenny, you know, this is, this is really important because, you know, unfortunately some of these decisions are made, you know, at levels in the elephant where
08:14:45 people don't have day to day experience with operations, and I'll even include myself in that, in that bucket. And so we really need folks input so you know I know there are a lot of people here that, you know, work in this space and deal with issues,
08:15:03 day in and day out. So, you know, please just take a few minutes, click on the link, fill out the survey, it would be very helpful.
08:15:13 So, would you mind to update, so we see the decline, but I have a question in my mind that I would like to race. So first of all, I think it's important that the maximum.
08:15:25 People really participate to the surveys, based on what you were sharing with all of us, but then I'm also concerned because each open source community, they are providing their needs.
08:15:40 Right, which should be considered as part of the budget.
08:15:45 So, how the Linux Foundation is assessing the feedback that they are getting from the survey versus the clear demand of each open source community. When we are preparing the widget because I want to be sure that it's also consider in the violence, not
08:16:11 the so V, otherwise we can end up with a gap. Right. Um, well we will do as we have, as we always do every year, with regards to the budget planning.
08:16:20 So the, the call will go out to each of the individual TLC to state what their specific operational needs are.
08:16:34 But in terms of the, the operation survey, and its impact to the budget.
08:16:46 It could be reflected in terms of how
08:16:56 the, the governing board decides to allocate budgets in general.
08:17:08 So that's that's where the differences. But to answer your question, when it comes actual budget negotiation time.
08:17:21 With regards to each individual community filling out the spreadsheet.
08:17:28 Okay. So somehow.
08:17:31 You have the input from the open source community which will be the priority because we have already known by our needs, and you complement, additional needle up of this message is being forced by the fact that people should raise their individual voice
08:17:48 via the survey.
08:17:50 That's my understanding of how it will work. I hope I'm right.
08:17:56 That's, that's basically it Catherine.
08:18:00 But basically it so
08:18:03 the
08:18:07 that what gets reported up is what the community sees is issues.
08:18:16 And so there may be, you know, budget allocated in in more of a general aspect.
08:18:25 There could be, you know, there's always the potential for community asks for a particular community saying this is our priority, and the governing board looking at it and saying, well, We think this other thing is the strategic priority.
08:18:43 That's where we're going to put our plans that is always a possibility.
08:18:50 But the operation survey is going to drive a ton of that decision making process.
08:19:08 So please fill it out.
08:19:14 Yeah, We can't we can't do much with 57 people.
08:19:19 Um, and, certainly, as Jason pointed out, the people on the governing board aren't involved in the day to day operations of all of the projects, so that the back is he need to report that up.
08:19:45 Okay, so I will put a pin in that one.
08:19:52 There's also a survey for platform as a service that is out there that was launched by the Xg Vela community.
08:20:05 So I would encourage folks to take a look at that
08:20:12 as well.
08:20:15 The PTO meeting on Monday is cancelled
08:20:23 due to holiday here in the US.
08:20:26 As far as events coming up one summit is in October, and for future ETFs discussed yesterday at the tech meeting.
08:20:36 We are not planning anything for the remainder of 2021.
08:20:42 We think that folks are going to be relatively burnt out on virtual events.
08:20:49 The one summit will be a hybrid event.
08:20:54 Both face to face and and virtual those details are being worked out.
08:21:00 I know there was a question on that last week I still don't have any kind of answer to that.
08:21:06 Anyways for our community events, nothing planned for the remainder of 2021.
08:21:12 We are looking, looking at our first virtual DTF being in January of 2022, and then a face to face planned for June of 2022.
08:21:30 Is there anyone in the community that
08:21:35 has a great deal of of interest in pushing for anything
08:21:44 else than what I have here.
08:21:47 Can you just took an action passive is not just securities and locate an extra put together the surveys joint venture from locate an extra the border left.
08:22:16 Kenny about the not having community events this year. That is correct, but I want to remind people that we recently set up in our YouTube channel. So if people have content they would like to share with the world or the larger community that is an option.
08:22:26 And if you're wondering how to do that technically, please contact me and I'll get you in touch with the people in charge of this channel.
08:22:52 I know for example, some people from mostly from Orange we're preparing, we heard this morning, are preparing some Santa Fe orchestration demos they would like to share so I think the YouTube channel can be a good platform for that.
08:23:10 Thank you for reminding us so that ran appreciate it.
08:23:15 And I appreciate the coordination.

Agenda Items

Release Status

RelEng/Infrastructure

PTL Updates

PTL Updates

PTL Updates

Subcommittee Updates

Task Force UpdateCNF, ONAP Enterprise, Wiki 2.0

TSC Activities and Deadlines

Upcoming Events & Housekeeping

2021 LFN Operations Survey - dates extended until 16 Jul 2021

Zoom Chat Log

Task Force Update
CNF, ONAP Enterprise, Wiki 2.0

Browser not supported