================================= #opnfv-meeting: BGS weekly meetin ================================= Meeting started by frankbrockners at 16:06:21 UTC. The full logs are available at http://ircbot.wl.linuxfoundation.org/meetings/opnfv-meeting/2015/opnfv-meeting.2015-04-07-16.06.log.html . Meeting summary --------------- * in TSC meeting discussed: move bgs meeting to IRC (pbandzi, 16:07:30) * Frank asks: any opinions about this proposal? (pbandzi, 16:08:10) * Jonas is OK with IRC meeting (pbandzi, 16:08:50) * Szilard OK too (pbandzi, 16:09:08) * reason to move to IRC only - too many audio meetings are running on goto, and anyway all the notes are taking on IRC (pbandzi, 16:10:33) * suggestion move to IRC only and use audio if neccessary (pbandzi, 16:11:26) * BGS status workplan review (pbandzi, 16:12:09) * LF lab is still in process of configuring and cabling so it will be ready at the end or at the begging of next week (pbandzi, 16:12:43) * in TSC suggeston was to use Erricson equipment as a workaround till LF lab will be ready (pbandzi, 16:13:32) * Erricson HW is already setup and connected to jenkins. (pbandzi, 16:14:11) * Jonas indicated that Ericson lab can be accessible publicly (pbandzi, 16:14:34) * deploy.sh discussion (pbandzi, 16:15:57) * Szilard already workin on aut deploy script (pbandzi, 16:16:59) * one script needed to bring up Fuel VM. inside VM another script which launch main deployment procedure (pbandzi, 16:17:47) * Szilard describes status of deploy: not sure if he will be ready till the 9. of april (pbandzi, 16:18:37) * not yet done script which start VM itself. (pbandzi, 16:19:17) * testing is done on ericson equipment (pbandzi, 16:19:40) * Frank point to mail from Tim (pbandzi, 16:20:58) * Tim has already has one script that creates VM and installs foreman inside and kicks off installation of opnfv (pbandzi, 16:22:57) * doing it through ansible more info can be also found on wiki (pbandzi, 16:24:34) * Michal suggest Fuel also has API which can be used in script (pbandzi, 16:25:20) * Tim asks if we have specific HW with specific given MAC addresses (pbandzi, 16:26:06) * yaml file can be therefore configured staticaly (pbandzi, 16:26:23) * Szilard made python script which find MAC addrsses automaticaly (pbandzi, 16:26:55) * Frank asks can we have single deploy script? which include both bring up VM and installs OPNFV too (pbandzi, 16:27:41) * Szilard: doesn't matter to have one script or two... (pbandzi, 16:29:02) * Jonas: suggests we should have single deploy scripts which might be different inside but should have same method inside (pbandzi, 16:29:54) * to have same script arguments (pbandzi, 16:30:46) * AGREED: single deployment script regardless installer method (implementation can be different) (pbandzi, 16:31:36) * LF HW not available (pbandzi, 16:32:00) * use interim solution -> ericson HW (pbandzi, 16:32:50) * Jonas decsribes Ericson HW (pbandzi, 16:33:18) * it might be needed to add more storage (pbandzi, 16:33:34) * 7servers for foreman, and 7 servers for Fuel - all on same VLAN (pbandzi, 16:34:03) * according to blueprint they sshould be (pbandzi, 16:34:21) * Tim and Dan not yet tried to access Ericsson lab (pbandzi, 16:35:48) * they are gonna try to access ... Jonas will give them access (pbandzi, 16:36:25) * ACTION: Jonas will send mail to Tim and Dan with access details (pbandzi, 16:37:10) * Frank suggests for next 24 hours to try access Ericson and have on Thursday something to report for R1 (pbandzi, 16:38:34) * Jonas affraids about remote managing of HW (pbandzi, 16:39:31) * and suggest to deploy on nested VM. (pbandzi, 16:40:26) * Dan: MAC addresses are set staticaly in script and then forwrd to foreman... and they don't use iso (pbandzi, 16:42:26) * Dan: our iso is essentialy the cenos with foreman installe (pbandzi, 16:43:08) * Jumphost will be on VM located on "jumphost" (pbandzi, 16:45:49) * Jonas: for autodeployment can we use some foreman API to deploy on metal? (pbandzi, 16:47:01) * search for foreman on web and foreman.org has all the documentation (pbandzi, 16:47:43) * Dan: assumed foreman node will be rebuild each time (pbandzi, 16:48:29) * Jonas: in fuel no IPMI implemented - cannot manage bare metal. (pbandzi, 16:49:21) * Dan; ipMI functionality present currently on Foreman (pbandzi, 16:49:55) * Dan: happy to share how they do this functionality (pbandzi, 16:50:21) * ACTION: Dan send some reference about IPMI implementation (pbandzi, 16:51:20) * ACTION: Dan and Jonas summarize the status what works and what not on erricson lab (pbandzi, 16:52:34) * any other burning issue? (pbandzi, 16:57:50) Meeting ended at 16:58:23 UTC. People present (lines said) --------------------------- * pbandzi (55) * collabot (4) * frankbrockners (2) * morgan_orange (1) Generated by `MeetBot`_ 0.1.4