16:00:23 <trevor_intel> #startmeeting OPNFV Pharos 16:00:23 <collabot`> Meeting started Wed Jan 13 16:00:23 2016 UTC. The chair is trevor_intel. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:23 <collabot`> Useful Commands: #action #agreed #help #info #idea #link #topic. 16:00:23 <collabot`> The meeting name has been set to 'opnfv_pharos' 16:00:32 <trevor_intel> #topic roll call 16:00:43 <fdegir> #info Fatih Degirmenci 16:00:46 <trevor_intel> #info Trevor (INtel lab) 16:00:49 <jmorgan1> trevor_intel: are you using GTM? 16:00:59 <fdegir> trozet: radez: ping 16:01:02 <trevor_intel> jmorgan1: no 16:01:04 <jmorgan1> #info Jack Morgan 16:01:44 <trevor_intel> ok lets start ... for me lab status, POD usage for rls, documenation are tops 16:01:52 <trevor_intel> waht else? 16:02:13 <trevor_intel> what else? 16:02:28 <jmorgan1> trevor_intel: Intel labs need to migrate some pods to new VPN enviornment 16:02:53 <trevor_intel> jmorgan1: we will get there 16:03:03 <trevor_intel> #topic lab status 16:03:22 <fdegir> to me, things look good 16:03:28 <fdegir> so I should first than jmorgan1 16:03:28 <trevor_intel> #info Intel POD 2 16:03:31 <fdegir> and trevor_intel 16:03:50 <fdegir> with those exceptions 16:03:57 <trevor_intel> jmorgan1: can you give an update on POD 2 16:04:35 <jmorgan1> im waiting to hear back from trozet, need additional info on what to troubleshoot 16:04:58 <trozet> jmorgan1: what are you waiting for from my side? 16:05:25 <jmorgan1> trozet: i'm not sure what is the problem 16:05:44 <trozet> jmorgan1: i can pull our latest RPM and try a run on there to see if it works 16:05:52 <jmorgan1> trozet: we have download/upload bandwith slowness and possibly internode connectivity issues 16:06:00 <jmorgan1> two possible issues 16:06:05 <trozet> jmorgan1: right 16:06:08 <jmorgan1> im waiting to hear about possible second issue 16:06:24 <jmorgan1> but its not been clear if second issue is an issue or not 16:06:27 <trozet> jmorgan1: did you login to the nodes and attempt to ping for a while between them? 16:06:44 <trozet> jmorgan1: watch how many pings fail over an hour, or the response time delays 16:06:58 <jmorgan1> trozet: no troubleshooting done for internode connectivity yet 16:07:17 <jmorgan1> trozet: are you using pod 2 currently? 16:08:11 <jmorgan1> i'd like to migrate it to another set of switches 16:08:14 <trozet> jmorgan1: no im not 16:08:25 <trozet> jmorgan1: feel free to do any troubleshooting you would like 16:08:33 <jmorgan1> do we know of anyone else using pod 2? 16:08:43 <fdegir> jmorgan1: it is reserved for apex 16:08:52 <trevor_intel> trozet: As I understand you will in teh future so we need to resolve 16:08:53 <fdegir> jmorgan1: so noone should be there 16:09:10 <jmorgan1> trozet: i'll solve this problem today and get back to you 16:09:17 <trozet> jmorgan1: thanks! 16:09:32 <trevor_intel> Right now Apex is only using 1 POD (LF POD 1) 16:09:34 <jmorgan1> trozet: you will need to get vpn access to env117 vpn (new vpn enviorment) 16:09:45 <trozet> jmorgan1: ok 16:10:20 <trevor_intel> trozet: ... can you confirm for rls B that yo do / donot intend to use a second POD? 16:10:45 <trevor_intel> trozet: just so everybody has clear understanding 16:10:47 <trozet> trevor_intel: we could definitely use intel pod2 and would use it if we thought it was stable 16:11:36 <trevor_intel> Fatih: POD 7 was planned for Fuel 16:12:03 <trevor_intel> Fatih: It is ready but no response from Fuel team if they need it for rls B 16:12:04 <fdegir> trevor_intel: I think you can take it back 16:12:20 <fdegir> trevor_intel: we didn't have ericsson pod available during november but we have it now 16:12:21 <jmorgan1> trevor_intel: let us get pod 2 working, I think we can do it today 16:12:22 <trevor_intel> Fatih: So it could be allcoated to Apex? 16:12:32 <fdegir> trevor_intel: can be done 16:12:43 <trevor_intel> jmorgan1: ok so keep as backup plan 16:12:52 <jmorgan1> trevor_intel: agreed 16:13:55 <trevor_intel> jmorgan1: can you summarise changes you are making to the lab config ... if you need anything leading up to rls B 16:14:21 <jmorgan1> trevor_intel: ok 16:14:30 <trevor_intel> jmorgan1: meaning any downtime needed? 16:15:22 <trevor_intel> We are also looking into teh bandwidth issue ... options to increase it 16:15:44 <jmorgan1> currently, Intel lab has each pod in its own isolated VPN enviornemnt, using lab switches (not managed by our team). I will be moving all PODs to a single VPN environment and our own switches we manage 16:15:53 <trevor_intel> The infrastructure is there to support it but do not haev costs yet 16:16:16 <jmorgan1> this means that a person only needs to connect via vpn and can ssh to any pod in the Intel Lab. 16:16:38 <trevor_intel> So it will ease support alot 16:16:40 <fdegir> jmorgan1: this is good 16:16:47 <jmorgan1> I have pods 7,8,9 on this new enviornment, I will move POD 1,2,3 first 16:17:08 <fdegir> jmorgan1: will the public IPs be impacted? 16:17:18 <jmorgan1> POD 2 today, then POD 3 and then POD 2. I only need an hour of downtime to change the IP and cable/label systems 16:17:51 <trevor_intel> jmorgan1: you mention POD 2 twice :) 16:17:59 <jmorgan1> POD 4,5,6 will need to pull current switches and install our switches, re-ip, cable/lable --> this will take 1/2 day of downtime 16:18:12 <jmorgan1> trevor_intel: pod 1 rather ;) 16:18:26 <jmorgan1> i will need to work with POD users to schedule down time 16:18:43 <jmorgan1> POD 4,5,6 will need to be done at the same time due to sharing same switch 16:19:00 <jmorgan1> every Intel POD will have 10gbe, and 1gbe network connectivity 16:19:10 <trevor_intel> jmorgan1: Fatih's question re public IP? 16:19:35 <fdegir> jmorgan1: I mean jumphost IPs 16:19:44 <jmorgan1> see https://wiki.opnfv.org/get_started/intel_pod7 as example 16:20:33 <jmorgan1> fdegir: yes, all POD systems will be on a 10.2.117.0/24 network 16:20:51 <jmorgan1> fdegir: I will work with each pod as we go to make this transition smooth 16:21:12 <fdegir> jmorgan1: then coordination with aricg is needed 16:21:24 <fdegir> jmorgan1: for LF firewal openings for jumphost public IPs 16:22:13 <jmorgan1> fdegir: no problem, we have time to coordiate when we make transition 16:22:30 <fdegir> jmorgan1: just a note for you 16:22:34 <jmorgan1> fdegir: do you want me to send you the list of every node's public IP to you? 16:22:45 <fdegir> jmorgan1: the new ones? 16:22:51 <jmorgan1> for LF firewall 16:23:00 <fdegir> jmorgan1: that would be good 16:23:01 <jmorgan1> fdegir: yes, new ones 16:23:10 <jmorgan1> fdegir: ok, will send today 16:23:13 <fdegir> jmorgan1: please include aricg as well 16:23:33 <jmorgan1> fdegir: ok 16:23:39 <fdegir> jmorgan1: another note 16:23:44 <jmorgan1> any other concerns with this change at Intel labs? 16:23:52 <jmorgan1> fdegir: shoot 16:23:53 <fdegir> jmorgan1: in case if I'm not around when you do the changes 16:24:08 <fdegir> jmorgan1: you can connect jumphost back to jenkins yourself 16:24:28 <fdegir> jmorgan1: once the firewall is confirmed to be open by aric 16:24:56 <fdegir> jmorgan1: just can check to see if slave.jar process is running as jenkins user or not 16:25:08 <fdegir> jmorgan1: if it is not running, you can issue sudo monit start jenkins command 16:25:26 <fdegir> jmorgan1: please see the link https://etherpad.opnfv.org/p/jenkins_slave_connection 16:25:37 <fdegir> jmorgan1: step 12.5 16:25:54 <fdegir> jmorgan1: I'll reply to your ip mail and include this info there as well 16:26:11 <jmorgan1> fdegir: ok, no problem 16:26:20 <trevor_intel> Anything else re. Intel PODs? 16:26:23 <fdegir> one last question regarding Intel PODs 16:26:39 <trevor_intel> keep going if there is more :) 16:26:39 <fdegir> pod8 is operational and compass people got the access 16:26:48 <fdegir> I haven't heard any feedback 16:26:55 <fdegir> jmorgan1: did you hear anything? any problem? 16:27:27 <jmorgan1> fdegir: no feedback nor problems stated. chigang did ping me earlier but not sure why 16:27:38 <fdegir> jmorgan1: ok 16:27:54 <fdegir> jmorgan1: I'll try to catch them early next morning and ask them to send feedback 16:28:21 <fdegir> that's all from me 16:28:24 <fdegir> for PODs 16:28:24 <jmorgan1> fdegir: we ddi confirm that we could do a fuel deployemnt in our new switch/vpn enviornment setup on pod 7 yesterday. so its looks good 16:28:44 <fdegir> good to hear 16:29:10 <fdegir> it could be repurposed for someone else as I mentioned 16:29:33 <trevor_intel> is it ok to talk docs now? Need help 16:29:38 <fdegir> yes 16:29:52 <trevor_intel> #topic Pharos documentation for Rls B 16:30:11 <trevor_intel> This is waht I see for docs ... 16:30:58 <trevor_intel> 1) project stuff "pharos.rst" 16:31:22 <trevor_intel> 2) pharos spec "pharos-spec.rst" 16:31:53 <trevor_intel> 3) jump server config 16:32:05 <trevor_intel> 4) remote connectivity requirements 16:32:22 <trevor_intel> 5) inventory template 16:32:32 <trevor_intel> 6) lab-topology template 16:32:43 <trevor_intel> 7) pod-description template 16:33:11 <trevor_intel> Wnat to verify understanding and who is workign on these 16:34:00 <trevor_intel> First ... do you agree Pharos is responsible for the inventory config file? 16:34:47 <fdegir> I think it can be pushed back to genesis/installers 16:35:00 <fdegir> if I'm not terribly mistaken 16:35:18 <trevor_intel> https://gerrit.opnfv.org/gerrit/#/c/4079/ 16:35:46 <fdegir> yes, this was it 16:35:56 <trevor_intel> Who owns the template/example ... which repo? 16:36:21 <trevor_intel> Or should own it? 16:36:24 <fdegir> it can perhaps reside in pharos 16:36:50 <trevor_intel> Fatih: how to close on that? 16:37:01 <fdegir> would be good to hear what frankbrockners says 16:37:23 <trevor_intel> frankbrockners: ping 16:37:41 <frankbrockners> yup - on a call... hence a bit slow to respond 16:37:55 <fdegir> the reason is that the example can be embedded to relevant pharos doc 16:38:22 <trevor_intel> frankbrockners: np ... trying to close on ownership of the inventory config file 16:38:25 <frankbrockners> how can I help 16:38:25 <fdegir> and residing next to pharos spec etc 16:38:54 <trevor_intel> frankbrockners: do you agree it should reside on pharos repo? 16:39:20 <trevor_intel> frankbrockners: Also what about network config? 16:39:23 <frankbrockners> probably makes the most sense 16:39:40 <frankbrockners> that said, afaik/ the "early versions" are in installer repos for now 16:39:48 <trevor_intel> frankbrockners: Deploy is Genesis, corect? 16:39:58 <frankbrockners> network and deploy config for sure reside in the installer repos 16:40:03 <frankbrockners> templates are genesis 16:40:09 <trevor_intel> frankbrockners: Test belongs in functest 16:40:18 <frankbrockners> actual instances will be installer specific and reside in the installer repos 16:40:53 <trevor_intel> frankbrockners: yes but mean templates / examples 16:40:59 <frankbrockners> we had this question by installers before 16:41:22 <frankbrockners> IMHO it would be easiest to agree on a location in the installer repos where network and deploy config would be found 16:41:34 <frankbrockners> rather than ask the teams to push things into another repo 16:41:42 <frankbrockners> because the files are installer specific 16:41:58 <frankbrockners> or better scenario specific 16:42:04 <frankbrockners> which are again installer specific 16:42:35 <trevor_intel> frankbrockners: How can we close on this so ... just so everybody is clear? 16:43:11 <frankbrockners> templates should be in genesisreq asap - we tried yesterday but still pending 16:43:31 <frankbrockners> inventory (you had the link above) is there 16:43:38 <frankbrockners> network and deploy are very close 16:43:58 <frankbrockners> apex has already created apex specific instances based on these templates 16:44:13 <frankbrockners> joid is working on the same 16:44:31 <trevor_intel> frankbrockners: ok so Pharos will NOT host inventory config file template 16:44:35 <frankbrockners> so hopefully - but the end of the week we have something 16:44:52 <frankbrockners> trevor_intel: for now it is in genesisreq 16:45:03 <frankbrockners> if you want to move it... I'm fine with that 16:45:03 <trevor_intel> frankbrockners: ok good enough for now 16:45:27 <trevor_intel> frankbrockners: its ok just want to make sure I am not missing 16:45:31 <frankbrockners> trevor_intel - thanks - let's be pragmatic for now 16:45:38 <trevor_intel> frankbrockners: makes sense 16:46:01 <trevor_intel> frankbrockners: thanks, I think we can move on and let you go 16:47:32 <trevor_intel> Fatih: Orange contribution ... 16:47:54 <fdegir> trevor_intel: yes 16:48:02 <trevor_intel> Templates for lab 16:48:22 <trevor_intel> topology and lab desription ... agree? 16:48:30 <trevor_intel> description 16:49:00 <fdegir> yes 16:49:25 <trevor_intel> There was an email and I suggested Orange gusy wirk with Jack 16:49:43 <trevor_intel> Since we haven't had a response, what to do for rls B? 16:50:22 <trevor_intel> I think the templates are really important 16:50:45 <fdegir> perhaps one of the available lab docs could be taken 16:50:50 <fdegir> and converted to a template 16:50:54 <fdegir> as we asked orange guys to do 16:51:29 <trevor_intel> jmorgan1: can we do this together? 16:51:48 <trevor_intel> jmorgan1: will take your time this week 16:51:50 <jmorgan1> trevor_intel: sure, just assign me the AR with details of what needs to be done 16:52:13 <jmorgan1> trevor_intel: via email. thanks 16:52:39 <trevor_intel> Fatih: These will all be .rst files? 16:52:47 <fdegir> trevor_intel: yes 16:53:07 <fdegir> trevor_intel: can you include additional names for review when you submit the patch to gerrit? 16:53:36 <fdegir> trevor_intel: Chris, Sofia Wallin and Ryota Mibu 16:53:46 <trevor_intel> Fatih: yes 16:53:48 <fdegir> trevor_intel: mainly for format requirements and so on 16:53:58 <jmorgan1> trevor_intel: i'll need commit access if submitting .rst, correct? 16:54:36 <trevor_intel> jmorgan1: no you don't haev commit permissions 16:54:49 <trevor_intel> You can submit for review 16:54:58 <trevor_intel> as a contributor 16:55:02 <fdegir> jmorgan1: have you signed CLA on gerrit? 16:55:18 <fdegir> jmorgan1: please go to this link https://gerrit.opnfv.org/gerrit/#/settings/agreements 16:55:24 <fdegir> and you have the agreement there 16:55:25 <jmorgan1> trevor_intel: ok, just to an FYI mostly 16:55:31 <jmorgan1> fdegir: ok, i'll take a look 16:55:47 <fdegir> ping aricg as well to get contribution rights fixed faster 16:55:53 <fdegir> after you signed the CLA 16:56:35 <trevor_intel> Those were my main concerns ... 16:56:57 <trevor_intel> Pharos spec and project file we can work on 16:57:29 <trevor_intel> Does anybody have any other topics or concerns for today? 16:57:36 <trevor_intel> Fatih: ? 16:57:42 <trevor_intel> jmorgan1: ? 16:57:52 <fdegir> just last question 16:57:56 <fdegir> trevor_intel: about doc 16:58:17 <fdegir> trevor_intel: you might want to check with ChrisPriceAB to see if they intend to include any pharos info in composite release documentation 16:58:20 <trevor_intel> trozet: ? 16:59:03 <trevor_intel> Fatih: yes I know, thx 16:59:06 <fdegir> ok 16:59:12 <jmorgan1> trevor_intel: can we do a JRIA scrub for the Pharos project, perhaps next week? 16:59:42 <jmorgan1> assign tasks per release, close old items, add new ones, etc 16:59:45 <trevor_intel> jmorgan1: yes for sure needed badly ... once documents are done 16:59:56 <trozet> no other concerns from me 17:00:02 <trevor_intel> jmorgan1: Rls B is job 1 17:00:15 <trevor_intel> thanks all! 17:00:19 <fdegir> thx 17:00:22 <jmorgan1> thanks 17:00:29 <trevor_intel> endmeeting 17:00:36 <trevor_intel> #endmeeting