13:59:12 <[1]JonasB> #startmeeting BGS Fuel activities status No9
13:59:12 <collabot> Meeting started Thu Apr 23 13:59:12 2015 UTC.  The chair is [1]JonasB. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:59:12 <collabot> Useful Commands: #action #agreed #help #info #idea #link #topic.
13:59:12 <collabot> The meeting name has been set to 'bgs_fuel_activities_status_no9'
13:59:54 <lmcdasm> #info Daniel Smith
14:00:07 <[1]JonasB> #info Jonas Bjurel
14:00:11 <mskalski> #info Michal Skalski
14:00:14 <fdegir> #info Fatih Degirmenci
14:00:18 <stefan_berg> #info Stefan Berg
14:00:32 <jose_lausuch> #info Jose Lausuch
14:00:55 <[1]JonasB> Cool then we are ready to go, lets start with the baremetal deploy the daniel did
14:01:08 <[1]JonasB> #Topic Baremetal deploy in Eri Lab
14:01:13 <wangxiaoli> #info
14:01:17 <lmcdasm> #info 5 interation was succesful
14:01:20 <[1]JonasB> Daiel shoot info
14:01:26 <lmcdasm> # test and sanity worked - and all is well
14:01:35 <lmcdasm> #info tests and sanity worked - and all is well
14:01:40 <[1]JonasB> Super daniel
14:01:47 <lmcdasm> #info - speaking with Jose just before on strategy for testing setup
14:02:02 <[1]JonasB> Will you fill in the config in your excelent work book?
14:02:10 <[1]JonasB> Or already done?
14:02:33 <lmcdasm> #info yes - today i will update the Lab document with a drawing showing the bldes used (cause i had a blade fail last night and moved some things around)
14:02:45 <lmcdasm> #info and access information, ec
14:02:48 <Szila317> #info Szilard Cserey
14:03:09 <[1]JonasB> lmcdasm: A HW fault?
14:03:12 <lmcdasm> #info test ccase (ping a VM between two blades) passed
14:03:27 <lmcdasm> #info - yes blade 5 is throwing Red Screen of death following POST
14:03:38 <lmcdasm> #info - so the lab is 3 CONTROL and 2 COMPUTE (For now)
14:03:54 <lmcdasm> which is still plenty of HW (80 cores 1/2 TB of RAM)
14:03:58 <[1]JonasB> #info That is enough!
14:04:28 <[1]JonasB> lmcdasm: What about the NTFS MBR?
14:05:06 <lmcdasm> #info - was embeededd internal VIRTUAL BOOK DISK that comes ith HP that was enabled during the t-shooting on the MEzzanine card with an HP guy - turned back off on blade 1 and its fine now
14:05:28 <[1]JonasB> Ok:-)
14:06:09 <[1]JonasB> So lets move on
14:06:22 <[1]JonasB> #topic centos support
14:06:45 <[1]JonasB> #info I have merged Michals patch for centos support
14:06:58 <[1]JonasB> Michal: Anything more we need to do?
14:07:48 <mskalski> #info I tested deployment on no-ha env and everything works there, need to check ha env similar to this used on bare metal for ubuntu
14:09:02 <[1]JonasB> Great
14:10:05 <lmcdasm> do we wanna delete the ENV on the BM and do another run with Centos?
14:10:25 <lmcdasm> since we know the infra setup is fine.. Michal can get int here and just run another deploy? would only take a couple hours
14:10:35 <lmcdasm> and we can say we have done both on both nested and BM?
14:11:20 <mskalski> lmcdasm: yes if I can run deployment on bm it will be helpful
14:11:35 <[1]JonasB> lmcdasm: good point, I think we should do that
14:11:45 <[1]JonasB> Michal: Can you do that?
14:12:13 <mskalski> [1]JonasB: yes
14:12:44 <lmcdasm> cool! thx Michal! - im logged out so the lab is all yours as of right now :)
14:12:57 <mskalski> lmcdasm: thx
14:13:35 <lmcdasm> all the IP info is on there, etc - but if you need something else (like NTP for pre-deploy is 172.20.1.253, etc) let me know
14:13:40 <lmcdasm> im online on IRC today
14:14:08 <mskalski> lmcdasm: ok
14:14:46 <[1]JonasB> #info Michal will perform a manual bare-metal Fuel/CentOS deployment in Eri Lab
14:15:11 <[1]JonasB> Lets move forward, maybe func test
14:15:28 <[1]JonasB> #topic Functest
14:15:39 <[1]JonasB> Jose - shoot
14:15:53 <jose_lausuch> #info a new VM in the BL9 is being created to run all the rally installation framework
14:16:12 <[1]JonasB> Thats a good approach
14:16:25 <jose_lausuch> #once that works and it gets connected to Jenkins, it will be cloned by Daniel to the other BL's
14:16:29 <[1]JonasB> Where will we put the Jenkins slave?
14:16:37 <jose_lausuch> #info once that works and it gets connected to Jenkins, it will be cloned by Daniel to the other BL's
14:16:47 <jose_lausuch> in the VM
14:16:53 <jose_lausuch> that makes it portable
14:17:00 <[1]JonasB> Great!
14:17:08 <fdegir> have a question
14:17:20 <jose_lausuch> #info and as Daniel suggested, it can be even "dockerized" to run inside Fuel/Foreman withouth hacking too much fuel
14:17:21 <fdegir> we will have one VM to start testing, right?
14:17:30 <[1]JonasB> So you do portforwarding to that VM or do that VM consume a new public IP?
14:17:30 <lmcdasm> each lab will get its own VM
14:17:30 <jose_lausuch> that is the idea
14:17:44 <fdegir> ok
14:17:47 <lmcdasm> here .let me explain the thinking a bit (from networking side)?
14:17:47 <jose_lausuch> exactly, sorry, I Wasnt so precise
14:18:08 <[1]JonasB> lmcdasm: Screen is yours
14:18:12 <lmcdasm> the test VM (in the blade/lab enviorments) will have an external NIC on the 10.118.34.0/27 IP
14:18:29 <lmcdasm> and as well a presence on all the (5) networks (ADMIN, MGMT, STORAGE, ETC)
14:18:45 <lmcdasm> so you can have jenkins talk direct on one side and execute out the other in an isolatd manner
14:18:54 <lmcdasm> and my thinking (at least for esxi its already there)
14:19:10 <lmcdasm> is we can OVF this test rig  as well so we have an "autodeployment" of the test rig too
14:19:19 <lmcdasm> (ideally :P).
14:19:39 <lmcdasm> does that work for you Fatih?
14:19:46 <lmcdasm> (and in line with what we talked abotu Jose?).
14:20:14 <fdegir> i think so
14:20:16 <jose_lausuch> yep
14:20:29 <lmcdasm> well.. jose is gonna get the VM up and ill do the networking
14:20:41 <lmcdasm> and we will try and see and either win or tinker and then win later! :)
14:20:42 <fdegir> just let me know when you have the vm ready so I can as well try jenkins connectivity
14:21:01 <lmcdasm> will do - maybe a hour or two would you say Jose?
14:21:27 <fdegir> jose_lausuch: can you install java 1.7 while you're at it?
14:21:30 <[1]JonasB> Thats a great plan, everyone buys in to it?
14:21:40 <jose_lausuch> yes, less than 1 hour, the VMis up, im just installing the OS
14:21:42 <jose_lausuch> yes
14:21:46 <lmcdasm> one esc
14:21:46 <fdegir> goo
14:21:46 <fdegir> d
14:21:47 <lmcdasm> sec*
14:22:01 <lmcdasm> there is another option as well - just to contrast
14:22:07 <[1]JonasB> Anything more on func-test?
14:22:14 <lmcdasm> but lets move on *
14:22:15 <lmcdasm> :)
14:22:35 * fdegir goes to another meeting, please action me so I can check later
14:22:41 <[1]JonasB> #topic auto-deployment
14:22:47 <Szila317> #info Hi
14:23:00 <[1]JonasB> So before we start, let me say a few words
14:23:08 <Szila317> ok
14:24:04 <[1]JonasB> So we have two close tracks here - Stefan is doing a prototype hack, while Szilard is doing more of production code version.
14:24:16 <Szila317> yes that's right
14:24:34 <Szila317> basically is the same track
14:24:41 <Szila317> but one is pythonified
14:25:03 <Szila317> #info so in a couple of hours I will push
14:25:09 <[1]JonasB> So I think we should use the prototype to lead and gain experience while Szilard does a polished version where he doesnt have to rush with the speed of light - does that makes sens?
14:25:12 <Szila317> #info a working version of this stuff
14:25:34 <Szila317> #info so you can try it out
14:25:54 <[1]JonasB> Great Szilard, so that one workes on nested env!?
14:26:13 <Szila317> #info, it works on nested Libvirt environment
14:26:22 <[1]JonasB> ;-)))
14:26:40 <lmcdasm> cool once you push i
14:26:46 <lmcdasm> ill drop int he esxi adapter (or start to)
14:26:49 <lmcdasm> or?
14:26:54 <[1]JonasB> #info And Szilard has an HP DHA?
14:26:57 <lmcdasm> is there another focus Jonas?
14:27:18 <[1]JonasB> No - this is perfect
14:27:26 <Szila317> #info I have the HP DHA, maybe I need to add 1 or 2 functions to be able to load ISO file
14:27:50 <stefan_berg> So on my part I've been experimenting and prototyping more with the DEA, DHA and adapter concept and did some rethinking of some things. Maybe worthwhile to take that for a sping as well.
14:27:59 <Szila317> #info and I also plan to convert into Python all the remaining bash script from Stefan
14:28:03 <lmcdasm> Szi? - can you include it (even if its not 100) in your Push (the HP DHA?)
14:28:04 <stefan_berg> "sping!?" spin1
14:28:06 <[1]JonasB> Szilard, do you want help with finalizing the HP DHA?
14:28:14 <Szila317> yes
14:28:43 <lmcdasm> we have a "spare" blade that is set for just doing Power on / off / iso testing
14:28:45 <Szila317> but I need 2 blades later to be able to run tests
14:28:48 <lmcdasm> (blade 8)
14:28:54 <Szila317> okay I can use that
14:28:55 <lmcdasm> 2?
14:29:13 <Szila317> 2 when I will start testing on HP baremetal
14:29:21 <Szila317> 1 is enough for now
14:29:29 <lmcdasm> ok.. for the first part (for ISO coding use blade 8)
14:29:40 <lmcdasm> and when Blade 5 is fixed (ill up the ticket prio today)
14:29:40 <Szila317> ok
14:29:42 <lmcdasm> you can have that.
14:29:50 <[1]JonasB> Daniel, can you help out finalizing the HP DHA?
14:29:58 <lmcdasm> perhasp :)
14:30:09 <Szila317> Jonas I can do that, it's not too much
14:30:24 <Szila317> +1 method to load the ISO that's all
14:30:47 <Szila317> Daniel can you send me ESXi stuff so I can start think on it
14:30:53 <Szila317> how to convert it into python
14:30:54 <[1]JonasB> Ok - maybe you can sort that together - which one has the least to so (didnt say little to do:-)
14:31:12 <lmcdasm> ya ya
14:31:13 <Szila317> sure
14:31:29 <lmcdasm> since everyone else now has their playpens
14:31:40 <Szila317> Thank God :)
14:31:42 <lmcdasm> i will do some of that today (after the lab drawings and the VM setup for Jose )
14:31:50 * lmcdasm thinks he needs a personal ticketing system
14:31:56 <[1]JonasB> #info Daniel will help Szilard coding/finalizing the HP DHA
14:32:25 <[1]JonasB> lmcdasm: Spin up Jira on a VM!
14:32:58 <Szila317> Stefan, I will need some help from you later, to find out how to spin up a Fuel VM in order to be able to detect HP blades
14:33:02 <lmcdasm> you joke - but we might wanna if we are going to have a multi-user lab in MTl and maintain it with both /// and external people comig in
14:33:22 <[1]JonasB> Ok time to move forward?
14:33:26 <Szila317> ok
14:33:37 <[1]JonasB> #topic documentation
14:33:51 <stefan_berg> Szila317: Sure. As I said, I have done some rethinking around the DEA, DHA and adapter including the changes to make on the ISO.
14:34:11 <stefan_berg> Testing this with some new protype code.
14:34:18 <lmcdasm> #info - suggests that we have the roundtable meeting on Monday about this
14:34:18 <stefan_berg> Which is a more clean design.
14:34:26 <[1]JonasB> #Info Documentation 100% done - will be up for review in Gerrit tonight
14:34:35 <lmcdasm> we need to have some "talky time" all of us cause there are many ways to do this and I would like the brainstorm sessoin
14:34:43 <lmcdasm> nice work Jonas!
14:34:52 <Szila317> Stefan, can we discuss that on chat, or can you share it with me, so that I can understand and learn, that would be good
14:35:33 <[1]JonasB> Any more topics?
14:35:49 <lmcdasm> #info LF?
14:35:52 <stefan_berg> Szila317: I will have runnable code during the day. Also some new API specs on internal wiki
14:36:26 <Szila317> Stefan, great! wonderful :) thanks
14:36:46 <[1]JonasB> lmcdasm: My plan was to handover your set-up desc to LF lab
14:36:56 <lmcdasm> sneaky - sounds good
14:37:10 <[1]JonasB> And dont focus more on that before we have autodeploy going in the lab
14:37:35 <[1]JonasB> And dont focus more on that before we have autodeploy going in the la
14:37:38 <Szila317> ok
14:37:47 <[1]JonasB> #topic LF lab
14:37:59 <lmcdasm> ok
14:38:20 <[1]JonasB> #info Plan is to handover Daniels set-up desc to LF lab
14:38:27 <[1]JonasB> Anything more?
14:38:52 <[1]JonasB> Great work every one
14:39:04 <[1]JonasB> 1
14:39:06 <Szila317> Tacka dig!
14:39:14 <[1]JonasB> :-)
14:39:16 <[1]JonasB> 2
14:39:23 <[1]JonasB> 3
14:39:43 <[1]JonasB> Se you tomorrow!
14:39:47 <Szila317> bye
14:39:51 <[1]JonasB> #endmeeting