*** ashaikh has quit IRC | 00:06 | |
*** flwang has joined #openstack-meeting-alt | 00:14 | |
*** sarob has joined #openstack-meeting-alt | 00:32 | |
*** jmaron has joined #openstack-meeting-alt | 00:34 | |
*** jmaron has quit IRC | 00:39 | |
*** dolphm has quit IRC | 00:39 | |
*** boris-42 has joined #openstack-meeting-alt | 01:03 | |
*** markwash has joined #openstack-meeting-alt | 01:05 | |
*** markwash has quit IRC | 01:06 | |
*** boris-42 has quit IRC | 01:17 | |
*** nati_ueno has joined #openstack-meeting-alt | 01:20 | |
*** sarob has quit IRC | 01:23 | |
*** sarob has joined #openstack-meeting-alt | 01:23 | |
*** nosnos has joined #openstack-meeting-alt | 01:24 | |
*** radix_ has joined #openstack-meeting-alt | 01:25 | |
*** sarob has quit IRC | 01:29 | |
*** jmaron has joined #openstack-meeting-alt | 01:34 | |
*** jmaron has quit IRC | 01:39 | |
*** brents has quit IRC | 01:53 | |
*** dolphm has joined #openstack-meeting-alt | 01:55 | |
*** nati_ueno has quit IRC | 01:59 | |
*** brents has joined #openstack-meeting-alt | 02:05 | |
*** alazarev has joined #openstack-meeting-alt | 02:17 | |
*** dolphm has quit IRC | 02:17 | |
*** yamahata has joined #openstack-meeting-alt | 02:21 | |
*** radix__ has joined #openstack-meeting-alt | 02:21 | |
*** radix_ has quit IRC | 02:23 | |
*** dolphm has joined #openstack-meeting-alt | 02:24 | |
*** kraman has joined #openstack-meeting-alt | 02:25 | |
*** alazarev has quit IRC | 02:28 | |
*** cody-somerville has joined #openstack-meeting-alt | 02:28 | |
*** alazarev has joined #openstack-meeting-alt | 02:30 | |
*** sarob has joined #openstack-meeting-alt | 02:34 | |
*** flwang has quit IRC | 02:35 | |
*** sarob has quit IRC | 02:39 | |
*** radix_ has joined #openstack-meeting-alt | 02:40 | |
*** flwang has joined #openstack-meeting-alt | 02:40 | |
*** radix__ has quit IRC | 02:40 | |
*** jmaron has joined #openstack-meeting-alt | 02:43 | |
*** flwang has quit IRC | 02:43 | |
*** cody-somerville has quit IRC | 02:44 | |
*** brents has quit IRC | 02:46 | |
*** jmaron has quit IRC | 02:47 | |
*** aignatov has joined #openstack-meeting-alt | 03:05 | |
*** kraman1 has joined #openstack-meeting-alt | 03:06 | |
*** kraman has quit IRC | 03:09 | |
*** nati_ueno has joined #openstack-meeting-alt | 03:09 | |
*** nati_ueno has quit IRC | 03:14 | |
*** eankutse has joined #openstack-meeting-alt | 03:22 | |
*** alazarev has quit IRC | 03:22 | |
*** markwash has joined #openstack-meeting-alt | 03:23 | |
*** alazarev has joined #openstack-meeting-alt | 03:23 | |
*** dolphm has quit IRC | 03:25 | |
*** markwash has quit IRC | 03:28 | |
*** alazarev has quit IRC | 03:29 | |
*** coolsvap has joined #openstack-meeting-alt | 03:42 | |
*** jmaron has joined #openstack-meeting-alt | 03:43 | |
*** ben_duyujie has joined #openstack-meeting-alt | 03:48 | |
*** jmaron has quit IRC | 03:48 | |
*** boris-42 has joined #openstack-meeting-alt | 03:49 | |
*** dolphm has joined #openstack-meeting-alt | 03:52 | |
*** flwang has joined #openstack-meeting-alt | 03:52 | |
*** boris-42 has quit IRC | 03:54 | |
*** eankutse has joined #openstack-meeting-alt | 03:56 | |
*** nati_ueno has joined #openstack-meeting-alt | 04:09 | |
*** markwash has joined #openstack-meeting-alt | 04:16 | |
*** markwash has quit IRC | 04:24 | |
*** eankutse has quit IRC | 04:31 | |
*** dolphm has quit IRC | 04:31 | |
*** jmaron has joined #openstack-meeting-alt | 04:44 | |
*** sergmelikyan has joined #openstack-meeting-alt | 04:47 | |
*** aignatov has quit IRC | 04:48 | |
*** jmaron has quit IRC | 04:49 | |
*** IlyaE has joined #openstack-meeting-alt | 04:49 | |
*** IlyaE has quit IRC | 04:59 | |
*** IlyaE has joined #openstack-meeting-alt | 05:03 | |
*** IlyaE has quit IRC | 05:08 | |
*** IlyaE has joined #openstack-meeting-alt | 05:10 | |
*** sarob has joined #openstack-meeting-alt | 05:11 | |
*** yamahata has quit IRC | 05:12 | |
*** kraman1 has quit IRC | 05:18 | |
*** kraman has joined #openstack-meeting-alt | 05:24 | |
*** kraman1 has joined #openstack-meeting-alt | 05:24 | |
*** kraman has quit IRC | 05:28 | |
*** alazarev has joined #openstack-meeting-alt | 05:32 | |
*** nosnos_ has joined #openstack-meeting-alt | 05:36 | |
*** nosnos has quit IRC | 05:36 | |
*** DennyZhang has joined #openstack-meeting-alt | 05:40 | |
*** nati_ueno has quit IRC | 05:41 | |
*** boris-42 has joined #openstack-meeting-alt | 05:44 | |
*** jmaron has joined #openstack-meeting-alt | 05:45 | |
*** jmaron has quit IRC | 05:50 | |
*** IlyaE has quit IRC | 05:52 | |
*** SergeyLukjanov has joined #openstack-meeting-alt | 05:54 | |
*** IlyaE has joined #openstack-meeting-alt | 05:55 | |
*** sarob has quit IRC | 05:57 | |
*** sarob has joined #openstack-meeting-alt | 05:57 | |
*** kraman has joined #openstack-meeting-alt | 05:59 | |
*** sergmelikyan has quit IRC | 06:00 | |
*** flwang has quit IRC | 06:02 | |
*** sarob has quit IRC | 06:02 | |
*** kraman1 has quit IRC | 06:03 | |
*** boris-42 has quit IRC | 06:03 | |
*** flwang has joined #openstack-meeting-alt | 06:03 | |
*** DennyZhang has quit IRC | 06:04 | |
*** boris-42 has joined #openstack-meeting-alt | 06:06 | |
*** alazarev has quit IRC | 06:07 | |
*** alazarev has joined #openstack-meeting-alt | 06:08 | |
*** IlyaE has quit IRC | 06:11 | |
*** sergmelikyan has joined #openstack-meeting-alt | 06:12 | |
*** IlyaE has joined #openstack-meeting-alt | 06:13 | |
*** markwash has joined #openstack-meeting-alt | 06:20 | |
*** alazarev has quit IRC | 06:26 | |
*** sarob has joined #openstack-meeting-alt | 06:36 | |
*** alazarev has joined #openstack-meeting-alt | 06:40 | |
*** SergeyLukjanov has quit IRC | 06:41 | |
*** jmaron has joined #openstack-meeting-alt | 06:46 | |
*** yamahata has joined #openstack-meeting-alt | 06:46 | |
*** jmaron has quit IRC | 06:46 | |
*** flwang has quit IRC | 06:53 | |
*** flwang has joined #openstack-meeting-alt | 06:56 | |
*** nosnos_ has quit IRC | 06:56 | |
*** nosnos has joined #openstack-meeting-alt | 06:56 | |
*** IlyaE has quit IRC | 07:01 | |
*** nati_ueno has joined #openstack-meeting-alt | 07:06 | |
*** nati_ueno has quit IRC | 07:08 | |
*** kraman has quit IRC | 07:16 | |
*** sarob has quit IRC | 07:19 | |
*** sarob has joined #openstack-meeting-alt | 07:19 | |
*** akuznetsov has quit IRC | 07:23 | |
*** sarob has quit IRC | 07:24 | |
*** jcoufal has joined #openstack-meeting-alt | 07:27 | |
*** boris-42 has quit IRC | 07:37 | |
*** IlyaE has joined #openstack-meeting-alt | 07:39 | |
*** yogesh has joined #openstack-meeting-alt | 07:43 | |
*** flaper87 has quit IRC | 07:48 | |
*** flaper87 has joined #openstack-meeting-alt | 07:48 | |
*** flwang has quit IRC | 07:55 | |
*** SergeyLukjanov has joined #openstack-meeting-alt | 08:02 | |
*** alazarev has quit IRC | 08:12 | |
*** IlyaE has quit IRC | 08:12 | |
*** jtomasek has joined #openstack-meeting-alt | 08:14 | |
*** kraman has joined #openstack-meeting-alt | 08:16 | |
*** akuznetsov has joined #openstack-meeting-alt | 08:17 | |
*** aignatov has joined #openstack-meeting-alt | 08:45 | |
*** flwang has joined #openstack-meeting-alt | 08:46 | |
*** kraman has quit IRC | 08:51 | |
*** ilyashakhat has joined #openstack-meeting-alt | 08:52 | |
*** nosnos has quit IRC | 08:59 | |
*** nosnos has joined #openstack-meeting-alt | 08:59 | |
*** nosnos has quit IRC | 09:03 | |
*** lsmola_ has joined #openstack-meeting-alt | 09:10 | |
*** ashaikh has joined #openstack-meeting-alt | 09:28 | |
*** derekh has joined #openstack-meeting-alt | 09:45 | |
*** kraman has joined #openstack-meeting-alt | 09:46 | |
*** yogesh has quit IRC | 09:48 | |
*** boris-42 has joined #openstack-meeting-alt | 09:49 | |
*** kraman has quit IRC | 09:51 | |
*** SergeyLukjanov has quit IRC | 09:51 | |
*** SergeyLukjanov has joined #openstack-meeting-alt | 09:54 | |
*** sergmelikyan has quit IRC | 10:02 | |
*** ikhudoshyn has quit IRC | 10:05 | |
*** ikhudoshyn has joined #openstack-meeting-alt | 10:07 | |
*** flwang has quit IRC | 10:08 | |
*** ashaikh has quit IRC | 10:11 | |
*** aignatov has quit IRC | 10:14 | |
*** akuznetsov has quit IRC | 10:22 | |
*** rsblendido has joined #openstack-meeting-alt | 10:38 | |
*** rossella_s has joined #openstack-meeting-alt | 10:38 | |
*** noslzzp has joined #openstack-meeting-alt | 10:40 | |
*** kraman has joined #openstack-meeting-alt | 10:47 | |
*** kraman has quit IRC | 10:52 | |
*** akuznetsov has joined #openstack-meeting-alt | 10:56 | |
*** sergmelikyan has joined #openstack-meeting-alt | 10:57 | |
*** SaD has joined #openstack-meeting-alt | 10:58 | |
*** aignatov has joined #openstack-meeting-alt | 11:02 | |
*** aignatov_ has joined #openstack-meeting-alt | 11:06 | |
*** aignatov has quit IRC | 11:07 | |
*** aignatov_ is now known as aignatov | 11:07 | |
*** jcoufal has quit IRC | 11:14 | |
*** sergmelikyan has quit IRC | 11:14 | |
*** ilyashakhat has quit IRC | 11:17 | |
*** ilyashakhat has joined #openstack-meeting-alt | 11:18 | |
*** pcm_ has joined #openstack-meeting-alt | 11:20 | |
*** pcm_ has quit IRC | 11:20 | |
*** pcm_ has joined #openstack-meeting-alt | 11:20 | |
*** SergeyLukjanov has quit IRC | 11:24 | |
*** yamahata_ has joined #openstack-meeting-alt | 11:26 | |
*** SergeyLukjanov has joined #openstack-meeting-alt | 11:27 | |
*** akuznetsov has quit IRC | 11:37 | |
*** ben_duyujie has quit IRC | 11:39 | |
*** kraman has joined #openstack-meeting-alt | 11:48 | |
*** akuznetsov has joined #openstack-meeting-alt | 11:52 | |
*** jcoufal has joined #openstack-meeting-alt | 12:01 | |
*** coolsvap has quit IRC | 12:09 | |
*** yamahata_ has quit IRC | 12:19 | |
*** kraman has quit IRC | 12:21 | |
*** IlyaE has joined #openstack-meeting-alt | 12:30 | |
*** noslzzp has quit IRC | 12:31 | |
*** IlyaE has quit IRC | 12:34 | |
*** IlyaE has joined #openstack-meeting-alt | 12:43 | |
*** IlyaE has quit IRC | 12:47 | |
*** SergeyLukjanov is now known as _SergeyLukjanov | 12:53 | |
*** _SergeyLukjanov has quit IRC | 12:53 | |
*** coolsvap has joined #openstack-meeting-alt | 12:55 | |
*** yamahata_ has joined #openstack-meeting-alt | 13:05 | |
*** pdmars has joined #openstack-meeting-alt | 13:14 | |
*** kraman has joined #openstack-meeting-alt | 13:16 | |
*** kraman has quit IRC | 13:21 | |
*** sergmelikyan has joined #openstack-meeting-alt | 13:29 | |
*** nati_ueno has joined #openstack-meeting-alt | 13:35 | |
*** nati_ueno has quit IRC | 13:37 | |
*** nati_ueno has joined #openstack-meeting-alt | 13:38 | |
*** enikanorov has quit IRC | 13:38 | |
*** enikanorov has joined #openstack-meeting-alt | 13:38 | |
*** SergeyLukjanov has joined #openstack-meeting-alt | 13:44 | |
*** yamahata_ has quit IRC | 13:47 | |
*** IlyaE has joined #openstack-meeting-alt | 13:48 | |
*** yogesh has joined #openstack-meeting-alt | 13:48 | |
*** yamahata_ has joined #openstack-meeting-alt | 13:49 | |
*** yogesh has quit IRC | 13:53 | |
*** dprince has joined #openstack-meeting-alt | 13:54 | |
*** noslzzp has joined #openstack-meeting-alt | 13:57 | |
*** sballe has quit IRC | 13:57 | |
*** kevinconway has joined #openstack-meeting-alt | 14:10 | |
*** julim has joined #openstack-meeting-alt | 14:12 | |
*** cody-somerville has joined #openstack-meeting-alt | 14:16 | |
*** aignatov has quit IRC | 14:17 | |
*** kraman has joined #openstack-meeting-alt | 14:17 | |
*** eankutse has joined #openstack-meeting-alt | 14:19 | |
*** eankutse has quit IRC | 14:20 | |
*** eankutse has joined #openstack-meeting-alt | 14:20 | |
*** kraman has quit IRC | 14:21 | |
*** julim has quit IRC | 14:23 | |
*** julim has joined #openstack-meeting-alt | 14:26 | |
*** zhaoqin has joined #openstack-meeting-alt | 14:27 | |
*** jmaron has joined #openstack-meeting-alt | 14:27 | |
*** lblanchard has joined #openstack-meeting-alt | 14:30 | |
*** SergeyLukjanov is now known as _SergeyLukjanov | 14:32 | |
*** _SergeyLukjanov has quit IRC | 14:32 | |
*** Barker has joined #openstack-meeting-alt | 14:33 | |
*** jprovazn has joined #openstack-meeting-alt | 14:38 | |
*** amytron has joined #openstack-meeting-alt | 14:38 | |
*** kraman has joined #openstack-meeting-alt | 14:53 | |
*** kraman has quit IRC | 14:57 | |
*** kraman has joined #openstack-meeting-alt | 14:57 | |
*** kraman has quit IRC | 15:02 | |
*** SergeyLukjanov has joined #openstack-meeting-alt | 15:02 | |
*** SergeyLukjanov has quit IRC | 15:04 | |
*** devkulkarni has quit IRC | 15:06 | |
*** SergeyLukjanov has joined #openstack-meeting-alt | 15:10 | |
*** aignatov has joined #openstack-meeting-alt | 15:16 | |
*** eankutse has quit IRC | 15:18 | |
*** eankutse has joined #openstack-meeting-alt | 15:19 | |
*** alazarev has joined #openstack-meeting-alt | 15:22 | |
*** tanisdl has joined #openstack-meeting-alt | 15:25 | |
*** eankutse has quit IRC | 15:27 | |
*** jmaron has quit IRC | 15:29 | |
*** kraman has joined #openstack-meeting-alt | 15:36 | |
*** devkulkarni has joined #openstack-meeting-alt | 15:38 | |
*** markwash has quit IRC | 15:40 | |
*** betsy has quit IRC | 15:41 | |
*** d0ugal has quit IRC | 15:51 | |
*** d0ugal has joined #openstack-meeting-alt | 15:52 | |
*** d0ugal has joined #openstack-meeting-alt | 15:52 | |
*** kraman has quit IRC | 15:53 | |
*** kraman has joined #openstack-meeting-alt | 15:58 | |
*** stanlagun has joined #openstack-meeting-alt | 16:00 | |
*** kgriffs_afk is now known as kgriffs | 16:00 | |
*** kraman has quit IRC | 16:02 | |
*** tsufiev has joined #openstack-meeting-alt | 16:04 | |
*** julim has quit IRC | 16:04 | |
*** julim has joined #openstack-meeting-alt | 16:05 | |
*** jergerber has joined #openstack-meeting-alt | 16:05 | |
*** alcabrera has joined #openstack-meeting-alt | 16:06 | |
*** ametts has joined #openstack-meeting-alt | 16:06 | |
kgriffs | who's here for the marconi mtg? | 16:06 |
---|---|---|
alcabrera | o/ | 16:06 |
*** IgorYozhikov has joined #openstack-meeting-alt | 16:07 | |
kgriffs | flaper87: you around? | 16:08 |
flaper87 | o/ | 16:08 |
flaper87 | Yes sir | 16:08 |
kgriffs | excellente | 16:08 |
kgriffs | Well, we have 3 of us | 16:08 |
alcabrera | yup | 16:08 |
kgriffs | where's malini? | 16:09 |
* ametts is lurking | 16:09 | |
kgriffs | ametts: welcome! | 16:09 |
ametts | (and multitasking) | 16:09 |
*** aignatov has quit IRC | 16:09 | |
alcabrera | malini is at the doctor, last I heard. She'll be around in about 30m. | 16:09 |
flaper87 | ametts: d00000000d | 16:09 |
flaper87 | alcabrera: welcoime :D | 16:09 |
alcabrera | flaper87: :D | 16:09 |
kgriffs | alcabrera: kk | 16:09 |
kgriffs | Let's get this party started. | 16:09 |
alcabrera | /begin marconi-fiesta | 16:09 |
kgriffs | #startmeeting marconi | 16:09 |
openstack | Meeting started Mon Nov 18 16:09:50 2013 UTC and is due to finish in 60 minutes. The chair is kgriffs. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:09 |
openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:09 |
*** openstack changes topic to " (Meeting topic: marconi)" | 16:09 | |
openstack | The meeting name has been set to 'marconi' | 16:09 |
kgriffs | #link https://wiki.openstack.org/wiki/Meetings/Marconi#Agenda | 16:09 |
kgriffs | #topic Hong Kong summit retrospective | 16:10 |
*** openstack changes topic to "Hong Kong summit retrospective (Meeting topic: marconi)" | 16:10 | |
kgriffs | ametts, flaper87: how did you think things went? | 16:10 |
alcabrera | I'm going to drop a few relevant links here. | 16:10 |
flaper87 | I'm very happy with how things went | 16:10 |
alcabrera | #link https://etherpad.openstack.org/p/IcehouseMarconiFeaturesAPIChanges | 16:10 |
alcabrera | #link https://etherpad.openstack.org/p/IcehouseMarconiNotificationService | 16:11 |
alcabrera | #link https://etherpad.openstack.org/p/IcehouseMarconiAMQPSupport | 16:11 |
flaper87 | People attended to our sessions | 16:11 |
alcabrera | #link https://etherpad.openstack.org/p/IcehouseMarconiInOut | 16:11 |
flaper87 | we've got feedback from the community overall | 16:11 |
ametts | Decent attendance: 30-50 people at the design sessions, and I counted about 65 at the afternoon session. | 16:11 |
flaper87 | After our talk people gave feedback as well | 16:11 |
*** brents has joined #openstack-meeting-alt | 16:11 | |
ametts | We probably would have had more folks there, except it was Friday afternoon and people were starting to head out already. | 16:12 |
flaper87 | I'm also happy with how kgriffs session went | 16:12 |
*** eankutse has joined #openstack-meeting-alt | 16:12 | |
*** eankutse has quit IRC | 16:12 | |
kgriffs | I also thought it went well | 16:12 |
kgriffs | In spite of the jet lag. :p | 16:12 |
*** eankutse has joined #openstack-meeting-alt | 16:12 | |
alcabrera | Was there any particular feature | 16:12 |
alcabrera | that people were asking for? | 16:13 |
*** SergeyLukjanov has quit IRC | 16:13 | |
kgriffs | Lots of people interested in notifications | 16:13 |
kgriffs | one person was asking about security | 16:13 |
*** jmontemayor has joined #openstack-meeting-alt | 16:13 | |
kgriffs | as in, how do I ensure that one client can't get messages meant for a different one | 16:13 |
kgriffs | Another theme was "queue provisioning as a service" | 16:14 |
alcabrera | hmmm | 16:14 |
*** zane has joined #openstack-meeting-alt | 16:14 | |
flaper87 | yeah | 16:14 |
kgriffs | I think at some point it may make sense to spin up a project to provision single-tenant AMQP instances or somthing, ala Trove | 16:14 |
alcabrera | so notifications is definitely a hot-topic - cool. | 16:14 |
flaper87 | I'm not 100% sure about it but I haven't put many thoughts on that | 16:14 |
kgriffs | right | 16:15 |
kgriffs | something to keep in the back of our minds - not a high priority right now | 16:15 |
kgriffs | HP suggested several things | 16:15 |
kgriffs | like, binary payloads combined with storing the content type in the metadata | 16:16 |
kgriffs | (ala Swift) | 16:16 |
*** jmaron has joined #openstack-meeting-alt | 16:16 | |
kgriffs | Sounds like they don't have bandwidth to contribute code at the moment, but maybe that will change if they hire more people | 16:16 |
kgriffs | flaper87, ametts: other things you would call out? | 16:16 |
alcabrera | I'm looking forward to that! +1 for contributor bandwidth | 16:16 |
flaper87 | that's cool. I found Hung suggestions very valuable, unfortunatelly many of them didn't fit into current MArconi's goals | 16:17 |
flaper87 | but we should definitely keep an eye on those | 16:17 |
kgriffs | +1 | 16:17 |
kgriffs | he had some good "lessons learned" to share based on customer feedback | 16:17 |
kgriffs | sounded like most customers are internal ones, fwiw | 16:17 |
ametts | Sounds like there was good support for an early 1.1 version in the fourth session. | 16:18 |
kgriffs | (internal == HP) | 16:18 |
kgriffs | yep | 16:18 |
kgriffs | I was happy that Everett attended | 16:18 |
kgriffs | flaper87: He maintains the Rackspace Java SDK | 16:18 |
kgriffs | - Jclouds guy | 16:19 |
kgriffs | https://github.com/everett-toews | 16:19 |
kgriffs | flaper87: you should talk about client design with him some time | 16:20 |
flaper87 | ow, didn't know that! Awesome! | 16:20 |
kgriffs | which also reminds me, a couple people asked about JMS support | 16:20 |
flaper87 | I'll definitely do that | 16:20 |
kgriffs | flaper87: maybe you can ask him about JMS as well | 16:20 |
flaper87 | +1 | 16:20 |
kgriffs | rock on | 16:20 |
flaper87 | I've already some ideas re JMS / AMQP / STOMP / MQTT | 16:20 |
flaper87 | but we need to dig more into that | 16:21 |
kgriffs | kk | 16:21 |
flaper87 | and sounds like Jth summit thing | 16:21 |
flaper87 | anyway, I'll ping him offline | 16:21 |
kgriffs | otherwise, I still need to finish digesting my notes from the summit and updating the wiki/blueprints accordingly. TBH, I've been busy just catching up with email and preparing for the Solum mini-design-summit happening in SFO tomorrow and wed. | 16:22 |
kgriffs | #action flaper87 to get in touch with Everett Toews and talk shop | 16:22 |
*** katyafervent has joined #openstack-meeting-alt | 16:22 | |
kgriffs | speaking of which, we should start thinking about how Marconi could tie into a PaaS | 16:23 |
*** dprince has quit IRC | 16:23 | |
*** boris-42 has quit IRC | 16:23 | |
flaper87 | +1 | 16:23 |
alcabrera | +1 | 16:23 |
kgriffs | ok, anything else on the topic of the summit? | 16:23 |
*** alazarev has quit IRC | 16:23 | |
flaper87 | I have to take a deeper look at Solum, I whish I had more bandwidth to contirbute to it | 16:23 |
kgriffs | flaper87: bandwidth is hard to come by. I hear ya! | 16:24 |
kgriffs | I need to upgrade to 10g | 16:24 |
flaper87 | lol | 16:24 |
alcabrera | :P | 16:24 |
kgriffs | fwiw I am planning to starting contributing on a regular basis to Oslo | 16:24 |
kgriffs | I also may be doing some webob work | 16:25 |
kgriffs | we'll see | 16:25 |
kgriffs | so much to do, so little time. :p | 16:25 |
kgriffs | aaanyway | 16:25 |
kgriffs | I would like to see more Marconi folks contributing to other OS projects | 16:25 |
kgriffs | helps us stay in sync with the community, and we may end up getting people to reciprocate and contribute to Marconi | 16:26 |
kgriffs | I know flaper87 already does that. | 16:26 |
kgriffs | (kudos) | 16:26 |
flaper87 | :D | 16:26 |
alcabrera | I see the value of it. Just got to make the time. :) | 16:26 |
kgriffs | ok, let's move on | 16:27 |
kgriffs | #topic review actions from last mtg | 16:27 |
*** openstack changes topic to "review actions from last mtg (Meeting topic: marconi)" | 16:27 | |
kgriffs | http://eavesdrop.openstack.org/meetings/marconi/2013/marconi.2013-10-28-16.04.html | 16:27 |
kgriffs | #link http://eavesdrop.openstack.org/meetings/marconi/2013/marconi.2013-10-28-16.04.html | 16:27 |
kgriffs | I'll go first | 16:27 |
kgriffs | T-shirts: mission accomplished! | 16:27 |
flaper87 | kgriffs: +1 | 16:28 |
alcabrera | w00t | 16:28 |
kgriffs | we handed out all but about 4 of the 75 I brought with me | 16:28 |
* ametts is glad kgriffs didn't get arrested by Hong Kong Customs agents | 16:28 | |
kgriffs | lol | 16:28 |
flaper87 | lol | 16:28 |
kgriffs | they were like, "dude, do you have any drugs in there?". | 16:28 |
kgriffs | "no" | 16:28 |
kgriffs | "carry on" | 16:28 |
ametts | (I'm not sure how to encode bail on an expense report) | 16:28 |
kgriffs | LOL | 16:29 |
alcabrera | lol | 16:29 |
kgriffs | ametts: you have the contact info for the screen printing shop if you'd like to order more | 16:29 |
ametts | Yes I do. Thanks. | 16:29 |
kgriffs | They are great for meetups and recruiting fairs | 16:30 |
*** alazarev has joined #openstack-meeting-alt | 16:30 | |
kgriffs | flaper87: I will forward the graphics on to you | 16:30 |
*** alazarev has quit IRC | 16:30 | |
kgriffs | moving on | 16:30 |
flaper87 | kgriffs: awesome, thanks! | 16:30 |
kgriffs | " flaper87 to firm up cross-transport-api-spec" | 16:30 |
flaper87 | kgriffs: done, I've been also chatting with Cindy about that | 16:31 |
flaper87 | we've been writing mores stuff there | 16:31 |
flaper87 | and We'll start coding on it ASAP | 16:31 |
kgriffs | ok, cool | 16:31 |
*** rnirmal has joined #openstack-meeting-alt | 16:31 | |
*** alazarev has joined #openstack-meeting-alt | 16:31 | |
*** alazarev has quit IRC | 16:31 | |
flaper87 | once the whole idea is well defined, we'll also move the info to the wiki | 16:32 |
flaper87 | Architecture related info | 16:32 |
kgriffs | flaper87: you may also want to chat with the Nova guys about this | 16:32 |
kgriffs | since they will be using JSON schema as well, sounds like | 16:32 |
flaper87 | yeah | 16:32 |
alcabrera | jsonschema - picking up steam across the board. gtk. | 16:32 |
flaper87 | we are also using it in Glance, FWIW. But I'll definitely keep an eye on Nova work | 16:32 |
kgriffs | I envision a sort of RESTful WSME module growing out of this that multiple projects can share. | 16:32 |
*** ben_duyujie has joined #openstack-meeting-alt | 16:33 | |
kgriffs | alcabrera: oh yeah, you should have been at the Pecan/WSME nova session | 16:33 |
alcabrera | oh? | 16:33 |
*** betsy has joined #openstack-meeting-alt | 16:33 | |
kgriffs | let me summarize | 16:33 |
kgriffs | "you guys should be using Pecan/WSME for your next API" | 16:34 |
kgriffs | "OK. Who else is using it?" | 16:34 |
kgriffs | "Just ceilometer, but we have commitments from a bunch of other projects" | 16:34 |
kgriffs | "Blah blah blah..." | 16:34 |
kgriffs | "So how do extensions work with WSME" | 16:34 |
kgriffs | "They don't really. WSME definitions are static. But that is OK because you can just define a generic string table and use that. And OS projects should stop supporting arbitrary extensions anyway." | 16:35 |
*** ben_duyujie has quit IRC | 16:35 | |
*** yidclare has joined #openstack-meeting-alt | 16:35 | |
kgriffs | "That's not going to happen any time soon" | 16:35 |
kgriffs | "But we need to enforce portability of OS apps and scripts" | 16:36 |
kgriffs | [Back-and-forth, no consensus reached on that point] | 16:36 |
alcabrera | doesn't sound like that went well. | 16:36 |
kgriffs | "So, going back to WSME and extensions, did I hear right that the definitions are static." | 16:37 |
kgriffs | "That's correct." | 16:37 |
kgriffs | "OK, then that isn't going to work." | 16:37 |
kgriffs | [blah blah blah...] | 16:37 |
kgriffs | "OK, let's go with Pecan and JSON schema" | 16:37 |
kgriffs | I forgot to mention this little gem earlier... | 16:38 |
alcabrera | heh. :) | 16:38 |
kgriffs | "Marconi gets a pass on using Falcon since they are a data plane api" | 16:38 |
alcabrera | oh yeah! | 16:38 |
alcabrera | I heard that one from flaper87 | 16:38 |
alcabrera | very cool | 16:38 |
alcabrera | Lemme just note that down... | 16:39 |
alcabrera | #info Marconi gets a pass on using Falcon since they are a data plane api | 16:39 |
kgriffs | but, I still promised to do a POC pecan driver and try to help improve Pecan performance, which is starting to look like improving webob | 16:39 |
alcabrera | isn't a spin off of webob used in Swift? | 16:39 |
alcabrera | swiftob? | 16:39 |
kgriffs | p.s. - flaper87 and I only spoke a couple sentences the entire time | 16:39 |
kgriffs | swob | 16:40 |
flaper87 | :D | 16:40 |
kgriffs | not sure how related it is | 16:40 |
kgriffs | we may be able to "steal" some code or at least ideas from there | 16:40 |
kgriffs | ok, so other action item | 16:41 |
kgriffs | alcabrera to research ways we can use heat to deploy marconi itself, as well as provision queues, whatever | 16:41 |
alcabrera | started | 16:41 |
alcabrera | I did some reading on Heat this morning to familiarize myself with what it is, what it can be used for. | 16:41 |
alcabrera | The whole orchestration idea. | 16:41 |
notmyname | alcabrera: swob = a swift replacement to webob that doesn't break on every minro version bump | 16:41 |
kgriffs | alcabrera: have you already reached out to keith bray? | 16:41 |
alcabrera | kgriffs: I have not. | 16:41 |
kgriffs | notmyname: does it have the same interface | 16:42 |
kgriffs | alcabrera: ok, I will introduce you | 16:42 |
alcabrera | thanks! | 16:42 |
*** sacharya has joined #openstack-meeting-alt | 16:42 | |
notmyname | kgriffs: it's intended to be a drop-in replacement (ie that's what it first did). I assume it still is, but we don't test that path. https://github.com/openstack/swift/blob/master/swift/common/swob.py#L17 | 16:43 |
kgriffs | notmyname: ah, interesting. I wonder how much in Pecan would break if you just swapped it in place for webob | 16:44 |
notmyname | transit time to the office. if there are other questions, just ping me in one of those dozen or so of channels I'm in | 16:45 |
kgriffs | notmyname: will do, thanks! | 16:45 |
alcabrera | I'll keep researching heat-side and I'll speak with Keith Bray for next time. | 16:46 |
kgriffs | sounds like a plan | 16:46 |
kgriffs | #action alcabrera to keep researching heat, and engage with the team | 16:46 |
kgriffs | #action kgriffs to play with swob | 16:47 |
kgriffs | #topic Sharding Updates (cpp-cabrera) | 16:47 |
*** openstack changes topic to "Sharding Updates (cpp-cabrera) (Meeting topic: marconi)" | 16:47 | |
alcabrera | alright | 16:47 |
alcabrera | lots of great news here | 16:48 |
alcabrera | so - we have sharding being tested extensively over at Rackspace. | 16:48 |
alcabrera | Here's a report from a test run last week using 3 mongodb shards: | 16:48 |
alcabrera | #link http://198.61.239.147:8000/log/20131112-1806/report.html | 16:48 |
alcabrera | While the report tells us little about the scaling behavior, it *is* scaling, as we were maxing out a single storage instance at about 2000 rps previously. :) | 16:49 |
alcabrera | So - sharding works! | 16:49 |
kgriffs | so we peak at just under 4000 shards | 16:49 |
alcabrera | well, not 4000 shards... 4000 rps. :) | 16:50 |
alcabrera | but yes. | 16:50 |
kgriffs | one other data point - I noticed that the standard deviation (so to speak) is a lot smaller under heavy load for shards vs. no shards | 16:50 |
kgriffs | so, it gives more consistent performance to users | 16:50 |
kgriffs | which is always a good thing | 16:50 |
kgriffs | (just ask anyone using AWS) | 16:50 |
alcabrera | yup | 16:50 |
flaper87 | hahaha | 16:50 |
alcabrera | we're better able to hide GC latency this way. :D | 16:51 |
kgriffs | I believe the reason is that having more shards adds more "jitter" | 16:51 |
kgriffs | as in, more chances for requests to get through in a timely manner without stalling | 16:51 |
kgriffs | you know, chaos theory. ;) | 16:52 |
alcabrera | This'll be true, as long as long as few heavy hitters end up colocated - the "busy queues are colocated" problem | 16:52 |
alcabrera | coined as of this moment. :P | 16:52 |
*** alazarev has joined #openstack-meeting-alt | 16:52 | |
alcabrera | So other sharding news - | 16:53 |
alcabrera | The patches! | 16:53 |
alcabrera | We have patches awaiting review, and I've been updating as I go. | 16:53 |
alcabrera | Thanks for the review, flaper87! | 16:53 |
flaper87 | alcabrera: +1 | 16:53 |
alcabrera | The entry patch to sharding is: | 16:53 |
alcabrera | #link https://review.openstack.org/#/c/54605/ | 16:54 |
flaper87 | BTW, we've been lacking of reviews lately. We should get back to do a number of reviews per day | 16:54 |
alcabrera | The rest follows from there. | 16:54 |
alcabrera | flaper87: +1 | 16:54 |
alcabrera | I think the summit threw us all off our routine. | 16:54 |
alcabrera | :D | 16:54 |
alcabrera | Hopefully, we can get sharding merged soon. I'm requesting some help with reviews on that. | 16:55 |
alcabrera | so that's that for sharding. Any thoughts or questions on this? | 16:55 |
kgriffs | +1 wrt reviews | 16:56 |
kgriffs | flaper87, alcabrera: I would love it if you guys could also peak at solum once in a while and do some +1/-1's there | 16:56 |
kgriffs | I want to make sure it is aligned with other projects | 16:56 |
kgriffs | #link https://review.openstack.org/#/q/status:open+project:stackforge/solum,n,z | 16:56 |
flaper87 | sure thing! | 16:56 |
alcabrera | wow, solum is busy. | 16:57 |
alcabrera | A page of reviews pending! | 16:57 |
kgriffs | I will be doing lots of reviews for Marconi this week | 16:57 |
kgriffs | alcabrera: looks like you have -1 here to address | 16:57 |
kgriffs | https://review.openstack.org/#/c/54605/ | 16:57 |
kgriffs | flaper87: I was kind of waiting for you to get to a point where you are ready to +2 alcabrera's patches before I did my final review | 16:58 |
alcabrera | kgriffs: yup. I'm trying to DRY up the sharding implementation. It's proving to be a little tricky. | 16:58 |
alcabrera | I've addressed all feedback on that one save for the DRYness. | 16:58 |
kgriffs | I went over the patches quite a bit, esp. when I filled in a week or so ago for alcabrera while he was out of the office | 16:58 |
kgriffs | ok | 16:59 |
kgriffs | maybe we leave the DRYness for a follow-up patch? | 16:59 |
flaper87 | kgriffs: sounds good! I'll take another look at that patch | 16:59 |
alcabrera | it wouldn't hurt to save it for later, IMO. | 16:59 |
kgriffs | flaper87: thanks. I just feel like I am too close to the code and need to let it "cool off" before I look again | 16:59 |
kgriffs | flaper87: thoughts on saving DRY refactoring? | 17:00 |
kgriffs | (for the near future) | 17:00 |
flaper87 | If DRY changes are not big, I'd prefer seeing them in this patch, otherwise we can let them land in a separate patch | 17:00 |
* kgriffs has his eye on the time | 17:01 | |
alcabrera | oops - I actually didn't upload the patch where I addressed the feedbaclk. Let me do that now... >.> | 17:01 |
kgriffs | flaper87: agreed | 17:01 |
alcabrera | *feedback | 17:01 |
kgriffs | alcabrera: how much churn would the DRY work create? | 17:01 |
alcabrera | kgriffs: It would change the implementation for every forwarded method. | 17:01 |
*** SergeyLukjanov has joined #openstack-meeting-alt | 17:01 | |
*** jmontemayor has quit IRC | 17:01 | |
kgriffs | ah, *that* | 17:01 |
*** Barker has quit IRC | 17:01 | |
kgriffs | I remember thinking it could be made more DRY | 17:02 |
kgriffs | I may have even commented on it at one point | 17:02 |
kgriffs | but that's beside the point | 17:02 |
kgriffs | hmmm | 17:02 |
alcabrera | The pattern currently is: 1) lookup a target, 2) if it exists, call a method on a particular controller, 3) if it doesn't return a value or raise an exception. | 17:02 |
alcabrera | 1) ... 2a) ... 2b) | 17:02 |
* flaper87 just read alcabrera answer to his comment about that | 17:02 | |
*** akuznetsov has quit IRC | 17:04 | |
*** aignatov has joined #openstack-meeting-alt | 17:04 | |
kgriffs | ok, we are out of time | 17:05 |
alcabrera | should we move on to the next topic and address sharding further in #openstack-marconi? | 17:05 |
alcabrera | yep, that would do it. :P | 17:05 |
flaper87 | hahaha | 17:05 |
kgriffs | #action alcabrera to work on sharding feedback and get it approved | 17:05 |
kgriffs | #endmeeting | 17:05 |
*** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings" | 17:05 | |
openstack | Meeting ended Mon Nov 18 17:05:53 2013 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 17:05 |
openstack | Minutes: http://eavesdrop.openstack.org/meetings/marconi/2013/marconi.2013-11-18-16.09.html | 17:05 |
openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/marconi/2013/marconi.2013-11-18-16.09.txt | 17:05 |
openstack | Log: http://eavesdrop.openstack.org/meetings/marconi/2013/marconi.2013-11-18-16.09.log.html | 17:05 |
kgriffs | thanks guys! | 17:06 |
kgriffs | alcabrera: Can you add the minutes to the wiki? | 17:06 |
kgriffs | I will update the agenda | 17:06 |
alcabrera | kgriffs: will do | 17:06 |
kgriffs | thanks! | 17:06 |
*** enikanorov_ has joined #openstack-meeting-alt | 17:06 | |
*** amitgandhi has joined #openstack-meeting-alt | 17:07 | |
*** jmontemayor has joined #openstack-meeting-alt | 17:08 | |
*** alcabrera has left #openstack-meeting-alt | 17:08 | |
*** ametts has quit IRC | 17:09 | |
*** amitgandhi has left #openstack-meeting-alt | 17:09 | |
*** cpallares has joined #openstack-meeting-alt | 17:12 | |
*** cpallares has left #openstack-meeting-alt | 17:13 | |
*** Barker has joined #openstack-meeting-alt | 17:14 | |
*** julim has quit IRC | 17:15 | |
*** ashaikh has joined #openstack-meeting-alt | 17:18 | |
*** julim has joined #openstack-meeting-alt | 17:19 | |
*** jtomasek has quit IRC | 17:21 | |
*** enikanorov_ has quit IRC | 17:25 | |
*** enikanorov_ has joined #openstack-meeting-alt | 17:25 | |
*** jmaron has left #openstack-meeting-alt | 17:28 | |
*** jcoufal has quit IRC | 17:31 | |
*** dprince has joined #openstack-meeting-alt | 17:33 | |
*** flaper87 is now known as flaper87|afk | 17:37 | |
*** twoputt_ has joined #openstack-meeting-alt | 17:43 | |
*** twoputt has joined #openstack-meeting-alt | 17:43 | |
*** yjiang5 has joined #openstack-meeting-alt | 17:44 | |
*** aignatov has quit IRC | 17:48 | |
*** markwash has joined #openstack-meeting-alt | 17:51 | |
*** aignatov has joined #openstack-meeting-alt | 17:57 | |
*** harlowja has joined #openstack-meeting-alt | 17:59 | |
*** derekh has quit IRC | 18:00 | |
*** AlanClark has joined #openstack-meeting-alt | 18:07 | |
*** sarob has joined #openstack-meeting-alt | 18:07 | |
*** yogesh has joined #openstack-meeting-alt | 18:20 | |
*** amcrn has joined #openstack-meeting-alt | 18:24 | |
*** rossella_s has quit IRC | 18:25 | |
*** rsblendido has quit IRC | 18:25 | |
*** coolsvap has quit IRC | 18:28 | |
*** akuznetsov has joined #openstack-meeting-alt | 18:30 | |
*** ashaikh has quit IRC | 18:37 | |
*** eankutse has quit IRC | 18:38 | |
*** ashaikh has joined #openstack-meeting-alt | 18:39 | |
*** lsmola_ has quit IRC | 18:42 | |
*** jprovazn has quit IRC | 18:43 | |
*** rook has joined #openstack-meeting-alt | 18:49 | |
*** Brian___ has joined #openstack-meeting-alt | 18:50 | |
*** baoli has joined #openstack-meeting-alt | 18:55 | |
*** irenab has joined #openstack-meeting-alt | 18:57 | |
*** rnirmal has quit IRC | 18:58 | |
*** AlanClark has quit IRC | 18:59 | |
*** aclark_ has joined #openstack-meeting-alt | 18:59 | |
*** itzikb has joined #openstack-meeting-alt | 19:00 | |
*** sadasu has joined #openstack-meeting-alt | 19:00 | |
*** yidclare has quit IRC | 19:00 | |
*** vipul is now known as vipul-away | 19:01 | |
*** sarob has quit IRC | 19:01 | |
*** sarob has joined #openstack-meeting-alt | 19:01 | |
baoli | Hello everyone | 19:03 |
yjiang5 | hi | 19:03 |
sadasu | Hello! Welcome to our 1st lci passthrough meeting | 19:03 |
sadasu | pci* | 19:03 |
itzikb | hi | 19:04 |
irenab | hi | 19:04 |
*** ijw has joined #openstack-meeting-alt | 19:04 | |
yjiang5 | sadasu, baoli ,who will chair this meeting? | 19:05 |
sadasu | I put baoli as the chair | 19:06 |
*** sarob has quit IRC | 19:06 | |
yjiang5 | sadasu: cool | 19:06 |
baoli | @yjiang5, I was thinking to start our first meeting with discussion on pci whitelist and pci alias | 19:06 |
baoli | I had a discussion with yongli during the summit | 19:06 |
*** IlyaE has quit IRC | 19:07 | |
yjiang5 | baoli: sure. I know Yongli is working on this and finish part of the coding. | 19:07 |
irenab | @baoli: any specific networking requirements with regards to this? | 19:07 |
irenab | or we just go with white list approach first? | 19:08 |
baoli | @irenab yes. that's how we group the PCI devices | 19:08 |
ijw | So we'd whitelist into groups and then attach data to each group, I presume? | 19:08 |
yjiang5 | baoli: what do you mean exactly of the white list? I think white list is mostly to select the devices that can be assigned to guest, right? | 19:08 |
*** aclark_ has quit IRC | 19:08 | |
baoli | Ok, first Let me recap what we have discussed f | 19:08 |
*** eankutse has joined #openstack-meeting-alt | 19:09 | |
baoli | A white list is define by each and every compute node | 19:09 |
yjiang5 | baoli: great. | 19:09 |
baoli | As of now, you can define a white list as something like pci_passthrough_whitelist='[{"vendor_id":"1137","product_id":"0071"}] | 19:10 |
*** vipul-away is now known as vipul | 19:10 | |
irenab | alias defined by controller and has a meeting of provider network in case of NIC device type for PCI passthrough. Am I right? | 19:10 |
irenab | I mean meaning not meeting, sorry | 19:11 |
baoli | @irenab, sorry, let me finish the recap. I will indicate the end of the recap | 19:11 |
baoli | So basically, it's a list of (<vendor_id, product_id>) now. But it's not sufficient to group the devices based on their common charactoristics (either by networking requirements, or others) | 19:13 |
*** sgordon has joined #openstack-meeting-alt | 19:13 | |
baoli | So Yongli thinks to enhance it as a list of tuples <filter, group_name>, with <filter> as <vendor_id, product_id, physical_function_id>, and <group_name> as a string | 19:15 |
*** akuznetsov has quit IRC | 19:15 | |
baoli | On top of that, on the controller node, one can define PCI alias for grouping | 19:16 |
ijw | PCI alias being what, just a name? | 19:17 |
*** eankutse has quit IRC | 19:17 | |
baoli | @ijw, essentially, yes. But it represents a group of PCI devices that share the same charactoristics as designated by the cloud admin | 19:18 |
yjiang5 | ijw: IIUC, alias is an alias to PCI requests. | 19:18 |
ijw | That's fine for some use cases but not ideal for all of them, I think. But did you want to finish the recap before I get into the details? | 19:18 |
*** derekh has joined #openstack-meeting-alt | 19:19 | |
baoli | @ijw, sure. Essentially, the compute node reports all the PCI devices by the means of PCI whitelist. PCI requests can be made by the means of PCI alias. end of recap | 19:20 |
ijw | OK, so the compute host is reporting a list of devices (or possibly a count, I know we've discussed both before) and the control node contains the additional data. That is generally the wrong way around, though. | 19:21 |
baoli | Can you elaborate why it's wrong? | 19:21 |
baoli | it's basically how it's working now. | 19:22 |
ijw | For network cards needing additional data - be that the switch port they're attached to, if you're dealing with them individually, or the provider network they're attached to if you're dealing with them in groups - it's usually the compute node that needs that information at port update time. | 19:22 |
yjiang5 | ijw: here the compute node means nova, or neurtron agent? | 19:22 |
ijw | Also, for NIC ports plugged into a switch where you expect to reprogram the switch when you put the port into a VM, the allocation pool can't contain the data. | 19:23 |
ijw | nova, in practice. Nova would call neutron's update-port, neutron would program the switch port accordingly and, presumably, inform nova of the encap to put on the port. | 19:23 |
ijw | (if one is required) | 19:23 |
ijw | Also, I'll raise the point I was making several months ago, that in fact you don't really need to report every device back to the control node, only the number of devices of a particular type, which would be the number of devices per alias. The scheduler never needs to know what the PCI address is, it just needs to know if sufficient resource is free. | 19:25 |
*** jtomasek has joined #openstack-meeting-alt | 19:25 | |
yjiang5 | ijw: For your scheduler information, that's been done as is now. The so called pci_stats. +1 for your discussion and I will give my input after you finished it. | 19:25 |
ijw | Done now. | 19:25 |
*** garyduan has quit IRC | 19:26 | |
*** garyduan has joined #openstack-meeting-alt | 19:26 | |
yjiang5 | I think white list is the mechanism to select the device to be assigned. I'm not sure if it's best to specify the pci white list on it. I'd elaborate my idea. | 19:26 |
ijw | yjiang5: I'm fine with the grouping concept of the whitelist, that sounds eminently sensible, I just take issue with where auxiliary data should be attached. | 19:27 |
*** demorris has joined #openstack-meeting-alt | 19:28 | |
ijw | yjiang5: I don't think you can attach it to the alias because it can differ per device within the alias. | 19:28 |
yjiang5 | IMO, PCi device in fact two type of property, one is generic property like vendor_id, device_id, BDF. , another is usage specific property like switch port, the group name, anything that is defined. | 19:28 |
*** sarob has joined #openstack-meeting-alt | 19:28 | |
ijw | Yea | 19:28 |
baoli | ijw: additional networking info is defined by means of port profile, which is entered in the nova boot or nova interface-attach commands | 19:29 |
yjiang5 | currently the generic property is saved in the table as a separated column, while the usage specific one is defined in the extra_info. | 19:29 |
baoli | ijw, I agree with your comments about scheduling | 19:29 |
ijw | baoli: how can the tenant calling 'nova boot' know anything about the physical connection of the network interface? | 19:30 |
irenab | @ijw: tetant should not know | 19:30 |
yjiang5 | I'd keep the white list to be based on generic property, like BDF, vendor_id etc (we need add the BDF to white list). Then we need another mechanism to specify the extra attributes to the devices, and that extra attributes will be saved in DB as extra_info. | 19:30 |
ijw | yjiang5: that extra information can remain in memory on the compute host. | 19:31 |
ijw | Providing it's not required for scheduling, at least. | 19:31 |
ijw | It doesn't need to be persistently stored. | 19:31 |
ijw | (and that's true *even* if it's needed for scheduling, but talk to boris42 about that bit, he's messing about with the scheduler) | 19:31 |
yjiang5 | ijw: No, I think it is required in DB, for example, the group name can be one of them, and should be used by scheduler to select device. | 19:32 |
baoli | ijw: good question. But someone needs to know when trying to instantiate a VM that requires pci-passthrough and special networking requirement | 19:32 |
yjiang5 | irenab: a curios question, how will neutron get such information, like network switch ? | 19:32 |
baoli | One way to do that is to associate that with the network a VM will be attaching to. | 19:33 |
ijw | yjiang5: what at present needs to go into the DB is what's required for scheduling and nothing more - which, at present, is the alias. | 19:33 |
ijw | baoli: you need to specify when you want a PCI device, but in my opinion you should simply specify 'this should be a PCI device' to the —nic command rather than anything else, and the net-id should be respected as normal. | 19:34 |
yjiang5 | ijw: no, alias is just an alias to pci request, like group_name is xxx, vendor_id is xxx etc. And the scheduler will match the pci request with pci_stats in DB. | 19:34 |
ijw | yjiang5: the switch port was intended to be additional information that was stored in the compute node's config file. | 19:34 |
ijw | yjiang5: I could, quite reasonably, have two different NIC types with different vendor_id and want to use them the same. | 19:35 |
irenab | ijw: by switch port you mean the physical switch attached to Server? | 19:35 |
ijw | irenab: yup | 19:35 |
baoli | ijw, did you get a chance to go through an email thread we had before the summit? | 19:35 |
yjiang5 | ijw: alias can be ORed , so it's ok. It's just a rule. | 19:35 |
ijw | But why would you specify alias on the control node rather than the compute node? | 19:36 |
*** brents has quit IRC | 19:36 | |
*** ashaikh has quit IRC | 19:37 | |
baoli | Actually, with the enhanced whitelist defintion, no alias needs to be specified on the controller node. In a matter of fact, the <group_name> is the PCI alias name | 19:37 |
yjiang5 | ijw: to seperate the request with the property. | 19:37 |
*** yidclare has joined #openstack-meeting-alt | 19:37 | |
irenab | and group name is a provider network name? | 19:38 |
irenab | or switch port? | 19:38 |
ijw | Seems to me that if you're using a provider network name, or a switch port, that should be specifiable per-device or per-group on the compute node. So lines on the compute node should look something like: | 19:38 |
yjiang5 | baoli: I think that's in fact a quite simplified solution, not sure if it will meet all requirement. If yes, that's ok for me. | 19:38 |
sadasu | ire nab: IIUC: a switch port | 19:38 |
ijw | pci_passthrough_whitelist='{ "my_group_name": {"vendor_id":"1137","product_id":"0071", extra_data: { "provider_network": "fred"}] | 19:39 |
baoli | @irenab, if need be, yes you can use a pci alias to represent a provider network | 19:39 |
ijw | Then I could specify multiple lines with my_group_name, and the selection criteria could be very general or completely specific to slot and function. | 19:40 |
irenab | just need to make sure scheduler take in to account the network connectivity | 19:40 |
baoli | ijw, I don't think that you want to put network specific info in the whitelist | 19:40 |
*** jcooley_ has joined #openstack-meeting-alt | 19:41 | |
ijw | I want to put it somewhere where (a) it resides on the compute node, since the compute node should carry all its own information and (b) you can bind it to a specific port on a network card, specifically | 19:41 |
baoli | The association of networking info with VM is made during boot or interface operations | 19:41 |
ijw | No, it really isn't | 19:41 |
ijw | I associate it with a network at that point, yes, absolutely. But if I have to reprogram devices in the network to make that association I need to have the information around telling me what it is I have to reprogram | 19:42 |
*** ashaikh has joined #openstack-meeting-alt | 19:43 | |
baoli | That info will be maintained in the neutron port database | 19:43 |
sadasu | @baoli: the network information in the whitelist serves a diff purpose from the networking info given when we bring up a VM | 19:43 |
ijw | Where will it get that from? | 19:43 |
baoli | It can get it from the port-create api, for one example | 19:44 |
irenab | @baoli: can you please make a step abck and explain how the scheduling part is solved? | 19:45 |
baoli | The point is, when someone wants to create a port with special pci requirement, it should be able to indicate that and the special requirement he intends to assigen to that port | 19:45 |
ijw | That's an interesting idea, though not specifically what I was getting at. | 19:46 |
baoli | @irenab, can we discuss scheduling in more detail in the next meeting? | 19:46 |
irenab | @baoli: sure | 19:46 |
baoli | We should come to consensus about what exactly PCI whitelist and PCI alias are | 19:47 |
*** eankutse has joined #openstack-meeting-alt | 19:47 | |
ijw | So we agree that there's a filter on which PCI devices are included for passthrough, and we're grouping PCI devices into groups that we're calling 'aliases', I think | 19:48 |
sadasu | from the current discussion it appears that we need only one and not both. | 19:48 |
*** IlyaE has joined #openstack-meeting-alt | 19:48 | |
ijw | We seem to disagree on how the whitelist and aliases actually interact with each other. | 19:48 |
*** aignatov has quit IRC | 19:48 | |
yjiang5 | I'm not fully agree with the alias. | 19:48 |
irenab | As far as I understand alias is used for VM request | 19:49 |
yjiang5 | at least in current implementation, alias is just an alias to pci request, which is rules of pci property. | 19:49 |
ijw | Meaning it just turns a friendly name into a vendor/product set? | 19:49 |
irenab | white list is used to group devices available for VMs on certain compute node, right? | 19:49 |
*** sarob has quit IRC | 19:50 | |
ijw | I think at the moment whitelist just determines what's made available, rather than grouping them, doesn't it? | 19:50 |
*** sarob has joined #openstack-meeting-alt | 19:50 | |
baoli | @yjang5, PCI alias is part of the pci request, that indicate the requested PCI device should be alloacated from PCI devices that are defined by the alias. Is this right? | 19:50 |
*** kraman has joined #openstack-meeting-alt | 19:51 | |
yjiang5 | ijw: yes, at least now, whitelist is just determine what's available, not groupiing. I'm not against to use it as grouping, but that's in fact an extension to white list. After all, white list is white list, right? | 19:51 |
yjiang5 | baoli: alias is not part of request, it will be translated to pci request. | 19:51 |
yjiang5 | you can think alias as 'name' for pci request, which can be long and boring :) | 19:51 |
ijw | I would argue that you sometimes want to alias different device types together (for instance perhaps I don't care who my NIC vendor is just so long as I get a nic) | 19:52 |
*** SergeyLukjanov has quit IRC | 19:52 | |
*** sarob_ has joined #openstack-meeting-alt | 19:52 | |
*** sarob has quit IRC | 19:52 | |
yjiang5 | ijw: that's how you define the pci request. You want a pci request to cover the devices you want to group. | 19:52 |
*** IlyaE has quit IRC | 19:53 | |
ijw | OK | 19:53 |
ijw | I was thinking in terms of passing the friendly alias name between control and compute nodes, really, but I guess it works either way | 19:53 |
*** twoputt_ has quit IRC | 19:53 | |
*** twoputt has quit IRC | 19:53 | |
yjiang5 | baoli: I'm not against use whitelist for groupname, but just want to make sure every one is on the same page. | 19:54 |
yjiang5 | baoli: one ways step back, have you discussed with yongli of the exact format of the extended white list that covers the group? | 19:55 |
*** IlyaE has joined #openstack-meeting-alt | 19:56 | |
baoli | yjiang5, use of whitelist for groupname is actually a side effect of putting a "name" as part of the whitelist. Originally, I thought we could still have both whitelist and alias: with whitelist indicating all the PCI devices available on the compute node, and the alias to group pci devices | 19:56 |
baoli | yjang5, I had a long discussion with Yongli during the summit on this. | 19:57 |
irenab | @baoli: so how the VM boot request will look like? | 19:57 |
baoli | Unfortunately, it's 2am his time now, and he can't attend to this discusstion | 19:57 |
yjiang5 | irenab: your question is important , the vm boot request format :) | 19:58 |
baoli | @irenab, it should be similar to what we have discussed in the email thread | 19:58 |
yjiang5 | baoli: IMHO, group just mean device with similar property, right? Why you emphasis it ? | 19:59 |
irenab | @baoli: so there is still a flavor containing extra_spec with PCI device request, right? | 19:59 |
*** aclark_ has joined #openstack-meeting-alt | 19:59 | |
yjiang5 | baoli: Possibly I missed important information. | 19:59 |
*** rook has quit IRC | 20:00 | |
sadasu | Lets keep the vm boot request format as the topic for next week's discussion. | 20:00 |
baoli | so folks, time is up for this meeting. Let's continue with email. I am also planning to write up something based on our discussion so far, for everyoen to comment. We really need to come to consensus on what white list and alias are | 20:00 |
yjiang5 | baoli: sadasu , are you in US timezone? | 20:01 |
irenab | @baoli: +1 | 20:01 |
*** mestery_ has joined #openstack-meeting-alt | 20:01 | |
yjiang5 | if yes, we can discuss in IRC. | 20:01 |
sadasu | yjiang5 : yes | 20:01 |
yjiang5 | which channel you usually stays at? in #openstack-dev? | 20:02 |
*** sarob_ has quit IRC | 20:02 | |
*** devkulkarni1 has joined #openstack-meeting-alt | 20:02 | |
*** sarob has joined #openstack-meeting-alt | 20:02 | |
irenab | please keep us posted with details of your further discussion to comment | 20:02 |
*** mestery has quit IRC | 20:02 | |
yjiang5 | irenab: sure | 20:02 |
baoli | yjang5, I have a different meeting to attend now. | 20:03 |
yjiang5 | irenab: If we have any conclusion, will post to the list | 20:03 |
sadasu | yjiang5: lets meet in openstack-dev | 20:03 |
yjiang5 | ok | 20:03 |
irenab | Thank you for the meeting | 20:03 |
itzikb | thanks | 20:03 |
sadasu | thanks everyone | 20:03 |
yjiang5 | thanks. | 20:04 |
*** irenab has left #openstack-meeting-alt | 20:04 | |
*** devkulkarni has quit IRC | 20:05 | |
*** jcooley_ has quit IRC | 20:05 | |
*** devkulkarni1 has quit IRC | 20:06 | |
*** sarob has quit IRC | 20:07 | |
*** sadasu has quit IRC | 20:08 | |
*** mestery_ is now known as mestery | 20:08 | |
*** devkulkarni has joined #openstack-meeting-alt | 20:13 | |
*** rnirmal has joined #openstack-meeting-alt | 20:18 | |
*** alazarev has quit IRC | 20:22 | |
*** yogesh has quit IRC | 20:24 | |
*** IlyaE has quit IRC | 20:27 | |
*** yogesh_ has joined #openstack-meeting-alt | 20:28 | |
*** dprince has quit IRC | 20:30 | |
*** alazarev has joined #openstack-meeting-alt | 20:32 | |
*** itzikb has quit IRC | 20:35 | |
*** alazarev has quit IRC | 20:36 | |
*** derekh has quit IRC | 20:36 | |
*** IlyaE has joined #openstack-meeting-alt | 20:42 | |
*** IlyaE has quit IRC | 20:42 | |
*** IlyaE has joined #openstack-meeting-alt | 20:44 | |
*** sarob has joined #openstack-meeting-alt | 20:44 | |
*** Barker has quit IRC | 20:45 | |
*** jcooley_ has joined #openstack-meeting-alt | 20:45 | |
*** jcooley_ has quit IRC | 20:46 | |
*** rook has joined #openstack-meeting-alt | 20:54 | |
*** kgriffs is now known as kgriffs_afk | 20:55 | |
*** jtomasek has quit IRC | 20:55 | |
*** jtomasek has joined #openstack-meeting-alt | 20:56 | |
*** kgriffs_afk is now known as kgriffs | 20:58 | |
*** alazarev has joined #openstack-meeting-alt | 20:58 | |
*** ashaikh_ has joined #openstack-meeting-alt | 20:58 | |
*** denis_makogon_ has joined #openstack-meeting-alt | 20:59 | |
*** ashaikh has quit IRC | 21:00 | |
*** ashaikh_ is now known as ashaikh | 21:00 | |
*** jtomasek has quit IRC | 21:01 | |
*** geoffarnold has joined #openstack-meeting-alt | 21:01 | |
*** amotoki has joined #openstack-meeting-alt | 21:02 | |
*** yogesh_ has quit IRC | 21:02 | |
*** denis_makogon has quit IRC | 21:03 | |
*** denis_makogon_ is now known as denis_makogon | 21:03 | |
*** dmakogon_ has joined #openstack-meeting-alt | 21:03 | |
*** jtomasek has joined #openstack-meeting-alt | 21:04 | |
*** yogesh has joined #openstack-meeting-alt | 21:05 | |
*** noslzzp has quit IRC | 21:06 | |
*** aclark_ has quit IRC | 21:07 | |
*** IlyaE has quit IRC | 21:07 | |
*** lblanchard has quit IRC | 21:09 | |
*** IlyaE has joined #openstack-meeting-alt | 21:10 | |
*** IlyaE has quit IRC | 21:14 | |
*** julim has quit IRC | 21:19 | |
*** aignatov has joined #openstack-meeting-alt | 21:24 | |
*** aignatov has quit IRC | 21:25 | |
*** enikanorov_ has quit IRC | 21:26 | |
*** ilyashakhat has quit IRC | 21:26 | |
*** Brian___ has quit IRC | 21:26 | |
*** yogesh has quit IRC | 21:26 | |
*** ilyashakhat has joined #openstack-meeting-alt | 21:26 | |
*** enikanorov_ has joined #openstack-meeting-alt | 21:27 | |
*** yogesh has joined #openstack-meeting-alt | 21:27 | |
*** yogesh has quit IRC | 21:31 | |
*** jcooley_ has joined #openstack-meeting-alt | 21:43 | |
*** devkulkarni has quit IRC | 21:52 | |
*** devkulkarni1 has joined #openstack-meeting-alt | 21:52 | |
*** yogesh has joined #openstack-meeting-alt | 21:54 | |
*** baoli has quit IRC | 21:54 | |
*** vipul is now known as vipul-away | 21:59 | |
*** nati_uen_ has joined #openstack-meeting-alt | 22:00 | |
*** geoffarnold has quit IRC | 22:02 | |
*** matrohon has joined #openstack-meeting-alt | 22:03 | |
*** nati_ueno has quit IRC | 22:04 | |
*** pdmars has quit IRC | 22:08 | |
*** devkulkarni1 has quit IRC | 22:12 | |
*** vipul-away is now known as vipul | 22:13 | |
*** sgordon has quit IRC | 22:21 | |
*** sarob_ has joined #openstack-meeting-alt | 22:22 | |
*** sarob has quit IRC | 22:25 | |
*** sarob_ has quit IRC | 22:27 | |
*** RajeshMohan has quit IRC | 22:27 | |
*** RajeshMohan has joined #openstack-meeting-alt | 22:28 | |
*** kraman has quit IRC | 22:30 | |
*** yogesh has quit IRC | 22:34 | |
*** jtomasek has quit IRC | 22:35 | |
*** jcooley_ has quit IRC | 22:38 | |
*** lblanchard has joined #openstack-meeting-alt | 22:39 | |
*** yogesh has joined #openstack-meeting-alt | 22:40 | |
*** yogesh has quit IRC | 22:44 | |
*** yogesh has joined #openstack-meeting-alt | 22:45 | |
*** betsy has quit IRC | 22:48 | |
*** yogesh has quit IRC | 22:49 | |
*** julim has joined #openstack-meeting-alt | 22:49 | |
*** yogesh has joined #openstack-meeting-alt | 22:49 | |
*** eankutse has quit IRC | 22:53 | |
*** yogesh has quit IRC | 22:53 | |
*** sacharya has quit IRC | 22:55 | |
*** kraman has joined #openstack-meeting-alt | 22:57 | |
*** yogesh has joined #openstack-meeting-alt | 22:58 | |
*** pcm_ has quit IRC | 23:00 | |
*** julim has quit IRC | 23:01 | |
*** yogesh has quit IRC | 23:02 | |
*** devkulkarni has joined #openstack-meeting-alt | 23:02 | |
*** yogesh has joined #openstack-meeting-alt | 23:03 | |
*** RajeshMohan has quit IRC | 23:03 | |
*** RajeshMohan has joined #openstack-meeting-alt | 23:04 | |
*** amotoki has quit IRC | 23:04 | |
*** demorris has quit IRC | 23:06 | |
*** ashaikh has quit IRC | 23:08 | |
*** yogesh has quit IRC | 23:08 | |
*** ashaikh has joined #openstack-meeting-alt | 23:11 | |
*** stanlagun has quit IRC | 23:11 | |
*** beekneemech is now known as bnemec | 23:14 | |
*** matrohon has quit IRC | 23:16 | |
*** bdpayne has joined #openstack-meeting-alt | 23:16 | |
*** julim has joined #openstack-meeting-alt | 23:17 | |
*** amytron has quit IRC | 23:20 | |
*** yamahata_ has quit IRC | 23:21 | |
*** kgriffs is now known as kgriffs_afk | 23:24 | |
*** denis_makogon has quit IRC | 23:26 | |
*** vipul is now known as vipul-away | 23:28 | |
*** vipul-away is now known as vipul | 23:32 | |
*** alazarev has quit IRC | 23:40 | |
*** boris-42 has joined #openstack-meeting-alt | 23:41 | |
*** alazarev has joined #openstack-meeting-alt | 23:43 | |
*** vipul is now known as vipul-away | 23:44 | |
*** vipul-away is now known as vipul | 23:44 | |
*** boris-42 has quit IRC | 23:47 | |
*** RajeshMohan has quit IRC | 23:48 | |
*** RajeshMohan has joined #openstack-meeting-alt | 23:48 | |
*** RajeshMohan has quit IRC | 23:50 | |
*** RajeshMohan has joined #openstack-meeting-alt | 23:51 | |
*** vipul is now known as vipul-away | 23:54 | |
*** vipul-away is now known as vipul | 23:54 | |
*** vipul is now known as vipul-away | 23:55 | |
*** vipul-away is now known as vipul | 23:55 | |
*** sacharya has joined #openstack-meeting-alt | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!