| *** ijw has quit IRC | 00:26 | |
| *** ijw has joined #openstack-dev | 00:26 | |
| *** dolphm has joined #openstack-dev | 00:32 | |
| *** danwent has quit IRC | 00:34 | |
| *** e1mer has quit IRC | 00:36 | |
| *** e1mer has joined #openstack-dev | 00:37 | |
| *** e1mer has joined #openstack-dev | 00:37 | |
| *** danwent has joined #openstack-dev | 00:45 | |
| *** dtroyer is now known as dtroyer_zzz | 00:51 | |
| *** danwent has quit IRC | 01:19 | |
| *** dolphm has quit IRC | 01:22 | |
| *** sdake has quit IRC | 01:42 | |
| *** sdake has joined #openstack-dev | 01:42 | |
| *** william0 has joined #openstack-dev | 01:44 | |
| *** n0ano has quit IRC | 01:45 | |
| *** danwent has joined #openstack-dev | 01:46 | |
| *** nati_ueno has joined #openstack-dev | 01:54 | |
| *** pixelbeat has quit IRC | 02:08 | |
| *** littleidea has joined #openstack-dev | 02:11 | |
| *** yaguang has joined #openstack-dev | 02:18 | |
| *** Ryan_Lane1 has joined #openstack-dev | 02:18 | |
| *** nati_uen_ has joined #openstack-dev | 02:19 | |
| *** dtroyer_zzz is now known as dtroyer | 02:19 | |
| *** Ryan_Lane has quit IRC | 02:20 | |
| *** nati_ueno has quit IRC | 02:22 | |
| *** dolphm has joined #openstack-dev | 02:32 | |
| *** andrewsmedina has quit IRC | 02:33 | |
| *** dolphm has quit IRC | 02:37 | |
| *** shang has quit IRC | 02:50 | |
| *** littleidea has quit IRC | 03:07 | |
| *** shang has joined #openstack-dev | 03:07 | |
| *** edygarcia has joined #openstack-dev | 03:10 | |
| *** Ryan_Lane1 has quit IRC | 03:10 | |
| *** Ryan_Lane has joined #openstack-dev | 03:10 | |
| *** Ryan_Lane has quit IRC | 03:15 | |
| *** edygarcia has quit IRC | 03:22 | |
| notmyname | mtaylor: cool, thanks | 03:25 |
|---|---|---|
| *** tserong_ is now known as tserong | 03:26 | |
| *** Aaton_off is now known as Aaton | 03:26 | |
| *** Shrews has joined #openstack-dev | 03:27 | |
| *** littleidea has joined #openstack-dev | 03:27 | |
| *** Aaton is now known as Aaton_off | 03:27 | |
| *** dolphm has joined #openstack-dev | 03:30 | |
| *** dtroyer is now known as dtroyer_zzz | 03:30 | |
| *** dtroyer_zzz is now known as dtroyer | 03:36 | |
| *** sam_wan has joined #openstack-dev | 03:37 | |
| *** Mandell has joined #openstack-dev | 03:46 | |
| *** matwood_ has joined #openstack-dev | 03:52 | |
| *** nati_uen_ has quit IRC | 03:54 | |
| *** matwood has quit IRC | 03:55 | |
| *** dolphm has quit IRC | 03:55 | |
| *** matwood_ has quit IRC | 03:57 | |
| *** timello has joined #openstack-dev | 04:00 | |
| *** krtaylor has joined #openstack-dev | 04:01 | |
| *** timello has quit IRC | 04:03 | |
| *** timello has joined #openstack-dev | 04:03 | |
| *** littleidea has quit IRC | 04:15 | |
| *** dtroyer is now known as dtroyer_zzz | 04:22 | |
| *** Mandell has quit IRC | 04:25 | |
| *** arosen has quit IRC | 04:43 | |
| *** arosen has joined #openstack-dev | 05:03 | |
| *** william0 has quit IRC | 05:11 | |
| *** steveb_ has quit IRC | 05:17 | |
| *** ben___ has joined #openstack-dev | 05:18 | |
| *** dubsquared has quit IRC | 05:23 | |
| *** nati_ueno has joined #openstack-dev | 05:39 | |
| *** william0 has joined #openstack-dev | 05:43 | |
| *** matwood has joined #openstack-dev | 05:52 | |
| *** nati_ueno has quit IRC | 06:01 | |
| *** dachary has joined #openstack-dev | 06:05 | |
| *** anderstj has quit IRC | 06:12 | |
| *** dachary has quit IRC | 06:16 | |
| *** mindpixel has joined #openstack-dev | 06:26 | |
| *** mrunge has joined #openstack-dev | 06:30 | |
| *** garyk has joined #openstack-dev | 06:33 | |
| *** matwood has quit IRC | 06:33 | |
| garyk | ping gongys | 06:33 |
| *** dolphm has joined #openstack-dev | 06:51 | |
| *** mcolombo has quit IRC | 06:59 | |
| *** davidha has quit IRC | 07:00 | |
| *** mcolombo has joined #openstack-dev | 07:02 | |
| *** reidrac has joined #openstack-dev | 07:06 | |
| *** ben___ has quit IRC | 07:08 | |
| *** dachary has joined #openstack-dev | 07:13 | |
| *** erikzaadi has joined #openstack-dev | 07:15 | |
| *** erikzaadi has joined #openstack-dev | 07:16 | |
| *** littleidea has joined #openstack-dev | 07:19 | |
| *** littleidea has quit IRC | 07:30 | |
| *** asalkeld has joined #openstack-dev | 07:31 | |
| *** dolphm has quit IRC | 07:35 | |
| *** fc__ has joined #openstack-dev | 07:35 | |
| *** fc__ has quit IRC | 07:44 | |
| *** fc__ has joined #openstack-dev | 07:45 | |
| *** pixelbeat has joined #openstack-dev | 08:00 | |
| trapni | morning | 08:00 |
| trapni | is anyone willing to have a lookc at https://bugs.launchpad.net/nova/+bug/1016848 please? :-) | 08:01 |
| uvirtbot | Launchpad bug 1016848 in ubuntu "KVM instance stops communicating after some time" [Undecided,New] | 08:01 |
| *** rbasak has joined #openstack-dev | 08:02 | |
| *** darraghb has joined #openstack-dev | 08:09 | |
| *** davidha has joined #openstack-dev | 08:22 | |
| *** danpb has joined #openstack-dev | 08:25 | |
| *** Daisy has joined #openstack-dev | 08:27 | |
| *** Ryan_Lane has joined #openstack-dev | 08:32 | |
| *** rods has joined #openstack-dev | 08:36 | |
| *** dolphm has joined #openstack-dev | 08:45 | |
| *** dolphm has quit IRC | 08:50 | |
| *** sam_wan has quit IRC | 09:02 | |
| ttx | trapni: looking | 09:04 |
| *** Ryan_Lane has quit IRC | 09:05 | |
| ttx | trapni: does it look like: https://bugs.launchpad.net/ubuntu/+source/qemu-kvm/+bug/997978 | 09:07 |
| uvirtbot | Launchpad bug 997978 in qemu-kvm "KVM images lose connectivity with bridged network" [High,Confirmed] | 09:07 |
| *** zigo has joined #openstack-dev | 09:10 | |
| *** william0 has quit IRC | 09:15 | |
| *** maploin has joined #openstack-dev | 09:21 | |
| *** maploin has quit IRC | 09:22 | |
| *** maploin has joined #openstack-dev | 09:22 | |
| *** Daisy has quit IRC | 09:26 | |
| *** shang has quit IRC | 09:51 | |
| *** Ryan_Lane has joined #openstack-dev | 09:56 | |
| *** apevec has joined #openstack-dev | 10:03 | |
| *** danwent has quit IRC | 10:18 | |
| *** shang has joined #openstack-dev | 10:29 | |
| *** zigo has quit IRC | 10:41 | |
| *** salv-orlando has joined #openstack-dev | 10:57 | |
| *** davidha has quit IRC | 11:09 | |
| *** drewlander has joined #openstack-dev | 11:12 | |
| *** davidha has joined #openstack-dev | 11:13 | |
| *** Leander_ has joined #openstack-dev | 11:13 | |
| *** yaguang has quit IRC | 11:17 | |
| *** Leander_ has quit IRC | 11:18 | |
| *** markvoelker has joined #openstack-dev | 11:31 | |
| *** milner has joined #openstack-dev | 11:34 | |
| *** hugokuo has quit IRC | 11:37 | |
| ijw | Are there any docs or example code that corresponds to the current python-novaclient? I'm working backward from its CLI at the moment, which is a bit irritating; the api.rst doc seems to be completely wrong. | 11:40 |
| *** wiliam has joined #openstack-dev | 11:44 | |
| zykes- | ttx: ping | 11:55 |
| ijw | zykes-: optimist | 11:57 |
| zykes- | ijw: isn't he based on france ?;p | 11:58 |
| ijw | Maybe you should insist he moves there to save confusion? | 11:59 |
| ttx | zykes-: yes? | 11:59 |
| zykes- | ijw: you where saying ? | 11:59 |
| zykes- | ;p | 11:59 |
| ijw | dammit :p | 11:59 |
| zykes- | ttx: is there any news on hotel pricings for X design summit in the fall ? | 11:59 |
| ttx | zykes-: not yet | 12:00 |
| zykes- | any estimates? | 12:00 |
| ttx | zykes-: no idea. I was surprised they announced the location already :) Nothing was signed one week ago | 12:01 |
| zykes- | ;p | 12:01 |
| zykes- | which company is it that's located @ SD ? | 12:02 |
| *** salgado has joined #openstack-dev | 12:03 | |
| *** GheRivero_ has joined #openstack-dev | 12:04 | |
| *** mrunge has quit IRC | 12:05 | |
| *** agonella has joined #openstack-dev | 12:06 | |
| ijw | Maybe we should go with this theme. San Antonio, San Salvador... | 12:12 |
| zykes- | Is there a fee to signup ? | 12:13 |
| *** andrewsmedina has joined #openstack-dev | 12:13 | |
| *** reidrac has quit IRC | 12:27 | |
| *** reidrac has joined #openstack-dev | 12:29 | |
| *** andrewsmedina has quit IRC | 12:36 | |
| *** joesavak has joined #openstack-dev | 12:38 | |
| *** Leander_ has joined #openstack-dev | 12:38 | |
| Leander_ | hello all | 12:38 |
| Leander_ | i'm trying to run the tests to see if my code passes all the test | 12:39 |
| Leander_ | however i keep running in to this error: | 12:39 |
| Leander_ | FAIL: test_authors_up_to_date (nova.tests.test_misc.ProjectTestCase) | 12:39 |
| Leander_ | it keeps saying my email is not listed in the authors section | 12:40 |
| Leander_ | any way around it? | 12:40 |
| Leander_ | i also keep getting this error when i try to run the pep8 test: http://paste.openstack.org/show/18751/ | 12:41 |
| johngarbutt | looks like you don't have pep8 install | 12:42 |
| Leander_ | but i do | 12:42 |
| johngarbutt | in your virtual env | 12:42 |
| johngarbutt | try running with -N maybe | 12:42 |
| johngarbutt | that will run outside your virtual env | 12:42 |
| johngarbutt | (or recreate virtual env with -v -f ) | 12:43 |
| *** markmcclain has joined #openstack-dev | 12:43 | |
| Leander_ | the virtual env keeps failing at the mysql | 12:44 |
| Leander_ | it says i can't find the mysql.config | 12:44 |
| johngarbutt | ah OK, is that doing -v -f? | 12:44 |
| Leander_ | i usually run it with "-f" only | 12:45 |
| Leander_ | i'm currently running with "-v -f" | 12:45 |
| johngarbutt | It should make no difference I guess | 12:45 |
| johngarbutt | What you need is "apt-get install python-mysqldb" | 12:45 |
| johngarbutt | see this: http://stackoverflow.com/questions/5178292/pip-install-mysql-python-show-error | 12:45 |
| johngarbutt | I think that changed a few weeks back maybe | 12:46 |
| Leander_ | that package is already installed | 12:46 |
| johngarbutt | hmm, that fixed it for me | 12:46 |
| johngarbutt | have you got python-dev installed? | 12:46 |
| Leander_ | yes | 12:47 |
| johngarbutt | hmm, sorry, that was all I did to get it going, is this on Ubuntu 11.04? | 12:47 |
| Leander_ | i'm using ubuntu 12.04 | 12:47 |
| johngarbutt | sorry, my bad I think this is the one: | 12:47 |
| johngarbutt | sudo apt-get install libmysqlclient-dev | 12:47 |
| johngarbutt | or maybe without the -dev, can't 100% remember | 12:48 |
| Leander_ | it's with the -dev | 12:48 |
| johngarbutt | cool, did that work? | 12:49 |
| Leander_ | the virtual_env is still installing so far | 12:49 |
| Leander_ | guess i'll find out soon enough | 12:49 |
| Leander_ | yes | 12:50 |
| Leander_ | that did it =D | 12:51 |
| johngarbutt | cool :-) | 12:51 |
| Leander_ | while i'm at it | 12:51 |
| Leander_ | my first commit to openstack was rejected | 12:51 |
| johngarbutt | did they tell you why? | 12:51 |
| Leander_ | how to i submit a fix which resolves the problems associated with it? | 12:52 |
| johngarbutt | Oh right | 12:52 |
| *** andrewsmedina has joined #openstack-dev | 12:52 | |
| Leander_ | do i just commit angain= | 12:52 |
| johngarbutt | there is a wiki for that | 12:52 |
| johngarbutt | one sec | 12:52 |
| johngarbutt | http://wiki.openstack.org/HowToContribute | 12:52 |
| *** dachary has quit IRC | 12:52 | |
| johngarbutt | The developer sections tells you how to submit it to the code review system | 12:53 |
| johngarbutt | hope that helps | 12:53 |
| Leander_ | thx | 12:53 |
| johngarbutt | no prob | 12:53 |
| Leander_ | is this an error: "5 imports missing in this test environment" | 12:53 |
| Leander_ | i get this when i run_tests.sh -p | 12:53 |
| johngarbutt | Erm, think it told me it was 10 last time I did that on trunk | 12:54 |
| johngarbutt | I generally work in a separate branch | 12:54 |
| johngarbutt | so I just skip back to trunk to check that kind of stuff | 12:54 |
| johngarbutt | I guess someone should be fixing that error soon, patches welcome, obviously :-) | 12:54 |
| *** lts has joined #openstack-dev | 12:54 | |
| Leander_ | well i'm running on a separate branch | 12:54 |
| Leander_ | for my code | 12:54 |
| Leander_ | but i can't run the tests because of that | 12:55 |
| Leander_ | :/ | 12:55 |
| *** dachary has joined #openstack-dev | 12:55 | |
| johngarbutt | sure, I would just swtich back to master/trunk branch and see if you get the same result | 12:55 |
| johngarbutt | oh right, the tests will run if you leave off the "-p" bit | 12:55 |
| johngarbutt | I think the pep8 flag only runs pep8 | 12:55 |
| johngarbutt | normally pep8 runs after the unit tests, unless that has changed recently | 12:56 |
| Leander_ | noe | 12:56 |
| Leander_ | nope* | 12:56 |
| Leander_ | still getting the same error | 12:56 |
| Leander_ | :s | 12:56 |
| johngarbutt | but you have "--pep8" flag right? | 12:56 |
| Leander_ | nope, just "./run_tests.sh" | 12:56 |
| johngarbutt | oh... | 12:56 |
| johngarbutt | that must have changes | 12:56 |
| johngarbutt | changed^ | 12:56 |
| johngarbutt | maybe try "nosetests" that should get you going | 12:57 |
| *** andrewsmedina has quit IRC | 12:58 | |
| Leander_ | i don't get it | 13:00 |
| Leander_ | the test run on the main brach | 13:00 |
| Leander_ | but don't run on my branch | 13:00 |
| Leander_ | :/ | 13:00 |
| Leander_ | johngarbutt: ^ | 13:01 |
| *** segfault923 has joined #openstack-dev | 13:01 | |
| *** yaguang has joined #openstack-dev | 13:13 | |
| johngarbutt | hmm, that's rubbish | 13:16 |
| johngarbutt | what you changed? | 13:16 |
| Ryan_Lane | If I follow these directions (http://docs.openstack.org/diablo/openstack-compute/admin/content/setting-up-the-multi-host-mode.html) and scrub the network for my projects, what happens to my current instances? | 13:16 |
| Ryan_Lane | do they just lose their networking? | 13:16 |
| Leander_ | johngarbutt: i've added a new file for the libvirt diagnostics and modified connection.py in libvirt | 13:17 |
| Leander_ | that's all i did | 13:17 |
| *** dtroyer_zzz is now known as dtroyer | 13:18 | |
| johngarbutt | Leander_: does running the tests with nosetests work? | 13:18 |
| Leander_ | johngarbutt: "run_tests.sh nosetests" < like this? | 13:18 |
| *** GheRivero_ has quit IRC | 13:18 | |
| *** rkukura has left #openstack-dev | 13:19 | |
| *** ayoung has joined #openstack-dev | 13:19 | |
| johngarbutt | nope, just "nosetests" | 13:19 |
| johngarbutt | can you send on paste the error you see with the tests? | 13:20 |
| Leander_ | yes, hold on a sec, i'm currently reinstalling the virtualev | 13:20 |
| Leander_ | env* | 13:20 |
| Leander_ | forgot to remove the -f flag -.- | 13:20 |
| johngarbutt | @Ryan_Lane: I thought you could just restart all the nova-network nodes on your compute nodes, leaving the database alone, and things might just work | 13:21 |
| Ryan_Lane | well... | 13:21 |
| Ryan_Lane | the floating IPs only seem to exist on a single node | 13:21 |
| Ryan_Lane | the NAT rules, that is | 13:21 |
| johngarbutt | Leander_: hat it when that happens | 13:21 |
| Ryan_Lane | which is problematic | 13:21 |
| johngarbutt | yes that is correct I think | 13:21 |
| Ryan_Lane | how do you possibly set up routing for them, then? | 13:21 |
| johngarbutt | why is it a problem? | 13:21 |
| johngarbutt | oh i see | 13:22 |
| johngarbutt | not 100% sure... | 13:22 |
| Ryan_Lane | crap | 13:22 |
| Ryan_Lane | and now NAT rules have moved to a new networking node and I have no clue how to get them back | 13:23 |
| johngarbutt | oh I see | 13:23 |
| johngarbutt | I thought they should re-create if you go back to the old networking mode | 13:23 |
| johngarbutt | but not read that code, or played with it too cloesly | 13:23 |
| johngarbutt | recently anyway | 13:24 |
| Ryan_Lane | seems that isn't the case | 13:24 |
| Ryan_Lane | I disabled the new networking node | 13:24 |
| *** iccha has joined #openstack-dev | 13:24 | |
| johngarbutt | you need to update all the flags, and restart the old one right? | 13:24 |
| Ryan_Lane | oh, I need to turn off multi_host? | 13:24 |
| *** derekh has joined #openstack-dev | 13:24 | |
| johngarbutt | yep | 13:24 |
| johngarbutt | sorry, that should help at least | 13:24 |
| johngarbutt | multi-host basically makes it only take configuration for VMs on that host | 13:25 |
| johngarbutt | from what I remember | 13:25 |
| Ryan_Lane | well, it's funny, then, that if I do that on each network node, the NAT rules are correct, but I'll still have a SPOF for internal communication | 13:26 |
| johngarbutt | Do you not have floating ips on a separate network for public traffic, and just have your edge router allow those ips out, and let the rest just happne? | 13:26 |
| *** JStoker has quit IRC | 13:26 | |
| johngarbutt | SPOF? | 13:26 |
| Ryan_Lane | I'm not understanding how that would work | 13:26 |
| Ryan_Lane | the floating IPs are NAT rules | 13:26 |
| *** iccha has quit IRC | 13:27 | |
| *** maoy has joined #openstack-dev | 13:27 | |
| Ryan_Lane | the routers have to be able to send traffic to the public IPs, which it seems in this networking mode, can only be a single host for each IP | 13:27 |
| Ryan_Lane | there's got to be some way to route to the NAT forwarder | 13:27 |
| Ryan_Lane | and if the IP can be on random nodes, there's no way for the router to know which node to send the traffic to | 13:28 |
| johngarbutt | Would you not have your instances use the floating ip address as a second IP address, maybe that is what I am getting confused with | 13:28 |
| Ryan_Lane | that's not how nova works | 13:28 |
| Ryan_Lane | by default nova sets NAT rules for floating IPs | 13:28 |
| johngarbutt | right, hang on, got confused | 13:28 |
| Ryan_Lane | I don't think it even has a possibility of adding floating IPs on a second NIC | 13:28 |
| johngarbutt | I meant on the primary | 13:29 |
| johngarbutt | nic | 13:29 |
| Ryan_Lane | eh? | 13:29 |
| johngarbutt | give primary nic on instance two ip addresses | 13:29 |
| Ryan_Lane | nova can't do that | 13:29 |
| *** JStoker has joined #openstack-dev | 13:29 | |
| johngarbutt | sure, nova isn't doing that I guess | 13:29 |
| johngarbutt | hmm, there must be a way of making this work | 13:29 |
| Ryan_Lane | well, of course there is | 13:30 |
| Ryan_Lane | the rules are static | 13:30 |
| Ryan_Lane | have them on all nodes | 13:30 |
| Ryan_Lane | then have the router set next-hop to all of the nodes | 13:30 |
| Ryan_Lane | but that assumes all of the nodes have static rules | 13:30 |
| johngarbutt | can you not use ARP to sort that? | 13:30 |
| johngarbutt | maybe I am getting the wrong end of the stick here | 13:31 |
| johngarbutt | you seen this stuff: http://docs.openstack.org/trunk/openstack-compute/admin/content/existing-ha-networking-options.html | 13:31 |
| *** dachary has quit IRC | 13:33 | |
| Ryan_Lane | hm | 13:34 |
| BLZbubba | i'm having a weird problem here, launching a vm from a snapshot is like 100x slower than from an image | 13:35 |
| BLZbubba | is something different about the snapshot process that would cause this? | 13:36 |
| *** andrewbogott has joined #openstack-dev | 13:36 | |
| *** iccha has joined #openstack-dev | 13:38 | |
| *** belliott has joined #openstack-dev | 13:40 | |
| johngarbutt | BLZbubba: image launch might be from a cached image | 13:41 |
| *** sandywalsh has joined #openstack-dev | 13:41 | |
| *** matwood has joined #openstack-dev | 13:43 | |
| Leander_ | johngarbutt: nosetest runs, but gets stuck here: http://paste.openstack.org/show/18753/ | 13:44 |
| *** kbringard has joined #openstack-dev | 13:45 | |
| BLZbubba | johngarbutt: so it doesn't cache snapshots when you launch them? is it possible to turn a snapshot into an image? it looks like maybe just a glance property change | 13:45 |
| *** dprince has joined #openstack-dev | 13:45 | |
| Ryan_Lane | ugh, handling this via arp. | 13:46 |
| johngarbutt | BLZbubba: not sure, it might cache snapshots, but will not be quicker first time, and you might be choosing a different machine, depends on what you are doing I guess | 13:46 |
| Ryan_Lane | I'm going back to single network node until nova/quantum has sane routing | 13:46 |
| andrewbogott | sandywalsh, dhellman, maybe russelb: Can I get some reviews of 8585 today? I have lots of code piling up behind that one. | 13:46 |
| andrewbogott | That patch is GIANT but all the bits are interconnected, so it's not obvious how I would split it up. | 13:46 |
| sandywalsh | andrewbogott: on it now | 13:46 |
| andrewbogott | cool, thanks. | 13:46 |
| johngarbutt | Ryan_Lane: I think ARP is the way it works... is that a problem, you prefer static routes? | 13:46 |
| Ryan_Lane | I'd prefer that nova announced routes for its instances | 13:47 |
| andrewbogott | sandywalsh: Logging uses notifications and notications use logging, hence them being lumped together. I could probably break it into two stages, though, if need be. | 13:47 |
| Ryan_Lane | if it's doing routing, it should act like a router ;) | 13:47 |
| BLZbubba | johngarbutt: ok thanks, i just wanted to make sure i wasn't missing something obvious | 13:48 |
| johngarbutt | Ryan_Lane: right, I guess we could have some kind of router plug-in that could help integrate | 13:48 |
| Ryan_Lane | well, quantum eventually has plans for it | 13:48 |
| johngarbutt | BLZbubba: not AFAIK anyway! | 13:48 |
| johngarbutt | OK, that makes sense | 13:48 |
| Leander_ | johngarbutt: nosetest runs, but gets stuck here: http://paste.openstack.org/show/18753/ | 13:49 |
| sandywalsh | andrewbogott: 8585 has merge conflicts | 13:49 |
| johngarbutt | Leander_: right, you need to put yourself in the authors file, going through that getting started guide should help you | 13:49 |
| andrewbogott | sandywalsh: ok, I'll rebase and resolve in a minute. | 13:50 |
| andrewbogott | sandywalsh: Hm, actually, I just fetched and rebased w/out any conflicts. | 13:51 |
| sandywalsh | andrewbogott: ok, let me confirm on my side | 13:51 |
| Leander_ | johngarbutt: but i already added my name in the wiki and got approved in the cla group | 13:53 |
| agonella | Has Anyone a architectural overview of openstack? I was able to find only this one: http://docs.openstack.org/essex/openstack-compute/admin/content/conceptual-architecture.html. Are there more? | 13:54 |
| andrewbogott | Leander_: There's progress towards automatically filling in the authors file, but in the meantime you need to literally add your name as a part of the first patch you submit. | 13:55 |
| Leander_ | andrewbogott: which file do i need to modify? | 13:55 |
| andrewbogott | It's called 'Authors' :) | 13:56 |
| andrewbogott | In the top dir of Nova. | 13:56 |
| Leander_ | ok, thx! | 13:56 |
| andrewbogott | Format should be obvious based on the existing contents of that file. | 13:57 |
| ttx | annegentle: around ? | 13:57 |
| *** adalbas has joined #openstack-dev | 14:00 | |
| *** chuckieb has joined #openstack-dev | 14:01 | |
| Leander_ | can anyone explain to me why this fails? | 14:02 |
| Leander_ | http://paste.openstack.org/show/18754/ | 14:02 |
| johngarbutt | Is that an array of instances that you are accessing using a string? | 14:07 |
| johngarbutt | oh, ignore me | 14:07 |
| andrewbogott | Leander_: By guess is that you have a single entry in 'instance' and you defined it as 'name' rather than as ['name'] | 14:07 |
| Leander_ | def get_diagnostics(self, instance): | 14:08 |
| Leander_ | dom = self._lookup_by_name(instance['name']) | 14:08 |
| Leander_ | return diagnostics.get_diagnostics(dom) | 14:08 |
| Leander_ | this is code for the method ^ | 14:08 |
| *** dolphm has joined #openstack-dev | 14:08 | |
| Leander_ | andrewbogott: see above | 14:08 |
| johngarbutt | is it passed an instance_ref or an instance_uuid? | 14:09 |
| *** Ruetobas has joined #openstack-dev | 14:09 | |
| johngarbutt | looks like it is passed the instance name | 14:09 |
| *** dolphm_ has joined #openstack-dev | 14:10 | |
| johngarbutt | at least in your test | 14:10 |
| andrewbogott | The code is expecting a full-blown object or dict, not just a name. | 14:10 |
| *** nunosantos_ has joined #openstack-dev | 14:10 | |
| *** mattstep has quit IRC | 14:12 | |
| *** dolphm has quit IRC | 14:12 | |
| *** nunosantos has joined #openstack-dev | 14:12 | |
| Leander_ | andrewbogott: i've seen the code being used in similar circumstances | 14:13 |
| Leander_ | when i ran the code manually everything worked | 14:13 |
| Leander_ | i've run the code with devstack | 14:13 |
| Leander_ | and everything is working just fine | 14:13 |
| Leander_ | i have no idea why the test is failing | 14:14 |
| Leander_ | :S | 14:14 |
| *** japage has joined #openstack-dev | 14:14 | |
| Leander_ | or should i simply move the line "self._lookup_by_name(instance['name'])" into diagnostics.get_diagnostics ? | 14:14 |
| johngarbutt | the test is wrong | 14:14 |
| johngarbutt | you pass in the name, and not the instance_ref | 14:14 |
| *** mattstep has joined #openstack-dev | 14:14 | |
| johngarbutt | into get_diagnostics | 14:15 |
| *** chuckieb has quit IRC | 14:15 | |
| Leander_ | johngarbutt, wouldn't the code fail then when run? | 14:16 |
| jaypipes | davidkranz: thx for the recent review of the parallel testing branch. So, over the weekend I was able to track down a few more bugs in the code and now have fixed many of the issues that had cropped up. Strangely, one of the tests (ImagesTest) seems to be unable to clean up the user/tenant it creates in its setUpClass method, and I'm currently looking into why that is. | 14:16 |
| Leander_ | i've run it with devstack and it worked just fine | 14:16 |
| jaypipes | davidkranz: I'll push the code up again once more -- I modified things so that the users and tenants created for the test cases are named the same as the test case classes. This makes diagnosing issues MUCH easier! :) | 14:17 |
| jaypipes | davidkranz: also added a run_tests.sh script that uses the openstack.nose_plugin that jkoelker wrote to get "pretty" results. | 14:17 |
| johngarbutt | given your code nova/tests/test_virt_drivers.py should have self.connection.get_diagnostics(instance_ref) not self.connection.get_diagnostics(instance_ref['name']) | 14:18 |
| *** dachary has joined #openstack-dev | 14:19 | |
| davidkranz | jaypipes: Great. I will take a look. Are you planning to submit this version, or still just a draft? | 14:19 |
| *** edygarcia has joined #openstack-dev | 14:19 | |
| Leander_ | johngarbutt: how come this didn't fail to execute with devstack? :S | 14:19 |
| davidkranz | jaypipes: I am going to try to get vishy to look at https://bugs.launchpad.net/nova/+bug/1016633 today. I think that will really get in our way. | 14:20 |
| uvirtbot | Launchpad bug 1016633 in nova "Bad performance problem with nova.virt.firewall" [Undecided,New] | 14:20 |
| *** rkukura has joined #openstack-dev | 14:20 | |
| *** dubsquared has joined #openstack-dev | 14:20 | |
| jaypipes | davidkranz: it's still a Work in Progress. | 14:20 |
| davidkranz | jaypipes: k | 14:20 |
| jaypipes | davidkranz: k (on vishy integration :) | 14:20 |
| jaypipes | davidkranz: also, that soft reboot test consistently takes FOREVER. :( Didn't we have a patchset to change that test? | 14:21 |
| *** zigo has joined #openstack-dev | 14:21 | |
| davidkranz | jaypipes: Yes, but Daryl objected so I removed the commenting out of that case. We could revisit though. | 14:21 |
| davidkranz | jaypipes: I think the spec for softboot is confused, if it exists at all, and it is not clear what we should really be testing. | 14:23 |
| jaypipes | davidkranz: yeah, I'd be cool with modding that dang test | 14:23 |
| *** dachary has quit IRC | 14:23 | |
| davidkranz | jaypipes: Which is unfortunate since it is the default for reboot! | 14:23 |
| jaypipes | davidkranz: I think it would be best to email the main ML on that one... | 14:23 |
| jaypipes | davidkranz: you emailed the QA list, but I think it would be best to do it to the main list. | 14:24 |
| davidkranz | jaypipes: OK, I'll do that. | 14:24 |
| jaypipes | davidkranz: I know folks like maoy would chime in on that :) | 14:24 |
| jaypipes | OK, time to feed the Pug. She's begging. | 14:24 |
| Leander_ | johngarbutt: how come this didn't fail to execute with devstack? Is there a possibility that there is a problem with the test? | 14:28 |
| maoy | jay pipes, davidkranz: soft reboot depends on the guest responding to ACPI | 14:29 |
| maoy | jaypipes: ^^ | 14:30 |
| maoy | jaypipes, davidkranz: there are also parameters to convert soft reboot to hard reboot if it times out in X seconds. by default it's disabled. | 14:31 |
| *** mnaser has joined #openstack-dev | 14:32 | |
| *** davidha has quit IRC | 14:33 | |
| jaypipes | maoy: yeah, we figured that much out :) It's just that the timeout is painful in Tempest ;) | 14:33 |
| jaypipes | maoy: we need to add some stuff to tempest to do some inspection of the image used in testing (to see if it supports soft reboots) but the only way to do that is to agree to have glance store some metadata about it.. | 14:34 |
| davidkranz | maoy: Are you sure it is disabled? Because I see soft boot "succeeding" after about 2.5 minutes. | 14:34 |
| *** ncode has joined #openstack-dev | 14:35 | |
| johngarbutt | Leander_: when you ran devstack, are you sure you configured it to use your code? | 14:35 |
| Leander_ | johngarbutt: yes, i did | 14:36 |
| davidkranz | maoy: # libvirt_wait_soft_reboot_seconds=120 | 14:36 |
| davidkranz | #### (IntOpt) Number of seconds to wait for instance to shut down after | 14:36 |
| davidkranz | #### soft reboot request is made. We fall back to hard reboot if | 14:36 |
| davidkranz | #### instance does not shutdown within this window. | 14:36 |
| Leander_ | i ran nova diagnostics <instance> and my code was returned | 14:36 |
| maoy | davidkranz: reboot_timeout is default to 0. but it could be something else that made it work after 2.5 min. (actually are you sure it's rebooted? sometimes I get success but nothing is rebooted) | 14:36 |
| davidkranz | maoy: ^^^ | 14:36 |
| Leander_ | johngarbutt: i ran nova diagnostics <instance> and my code was returned | 14:37 |
| johngarbutt | Leander_: what was your localrc file? did you push your code to your own github account? | 14:38 |
| trapni | ttx: many thanks. that pretty much exactly looks like the bug I am having, too. :-( | 14:38 |
| maoy | davidkranz: yeah. that's another parameter, but kinda different. it doesn't try hard reboot after this times out. there seems to be also a bug with that code | 14:39 |
| *** edygarcia has quit IRC | 14:39 | |
| Leander_ | johngarbutt: the localrc file remaind untouched since the installation. I simply added the code to the nova repository downloaded by devstack and restarted it so that it loaded the newly added code. I have no github account of my own | 14:39 |
| *** edygarcia has joined #openstack-dev | 14:39 | |
| davidkranz | maoy, jaypipes : So this is kind of a mess. None of this behavior is documented in the API docs at docs.openstack.org. | 14:40 |
| davidkranz | maoy: How do you think we should proceed? | 14:41 |
| maoy | davidkranz, jaypipes: https://review.openstack.org/#/c/8731/ it's actually merged. | 14:41 |
| *** mnaser has quit IRC | 14:42 | |
| davidkranz | maoy: I will file a doc bug. | 14:42 |
| maoy | fixing the doc seems to be the first step. | 14:43 |
| *** alaski has joined #openstack-dev | 14:43 | |
| *** jdg has joined #openstack-dev | 14:43 | |
| maoy | jaypipes: i'm not sure what to do with tempest. perhaps to set libvirt_wait_soft_reboot_seconds more aggressive is an easy fix? | 14:43 |
| johngarbutt | Leander_: devstack may have just pulled new code down if you have the wrong setting. What did you restart? also, did you change any tests? | 14:44 |
| *** mnaser has joined #openstack-dev | 14:44 | |
| Leander_ | johngarbutt: i did not modify any test, and by restarting i mean reboot the machine and run stack.sh again | 14:44 |
| davidkranz | maoy: The problem is that that is a nova flag not controlled by tempest. Also, we really want to test that soft reboot works, not just that it times out to a hard reboot. | 14:46 |
| Leander_ | johngarbutt: also i've made at least 3 clean installations of devstack, and the code always workedn | 14:47 |
| jaypipes | maoy: k, doc fix it is then. thx for your help! | 14:47 |
| ttx | trapni: ok, so looks like a KVM/Ubuntu bug, will mark as duplicate and let you comment on that other thread | 14:47 |
| davidkranz | I filed a tempest but about this last week https://bugs.launchpad.net/tempest/+bug/1014647 | 14:47 |
| uvirtbot | Launchpad bug 1014647 in tempest "Soft reboot has multiple issues" [Undecided,New] | 14:47 |
| trapni | ttx: yep. many thanks, but.... | 14:49 |
| maoy | davidkranz: by default it actually won't trigger a hard reboot | 14:49 |
| *** rnirmal has joined #openstack-dev | 14:49 | |
| maoy | davidkranz: it just times out.. | 14:49 |
| trapni | ttx: I am really very disappointed, I mean, kind of helpless, in how to actually fix/work around that. I mean, I've done so many to actually get OpenStack accepted in our company, but are overshadowed by incidents like this. Not that I'm complaining... just feeling helpless :-) | 14:50 |
| davidkranz | maoy: Oh. So it times out and sets the state back to ACTIVE, but it didn't really reboot? | 14:50 |
| trapni | ttx: please tell me if I can provide any debugging infos. I'd like to do as much as I can to get this thing solved faster :-) | 14:50 |
| maoy | davidkranz: i think so. unless you play with the reboot_timeout parameter too. vishy is the best person to confirm. | 14:50 |
| ttx | trapni: I understand. Myself I'd like to know if it's Ubuntu-specific or an upstream issue | 14:51 |
| davidkranz | maoy: Thanks for your help. This is even messier than I thought! | 14:51 |
| trapni | ttx: although, I'd like to state, that this happens the most on I/O intensive KVM instances. (not saying it is caused by /usr/bin/kvm) | 14:51 |
| ttx | trapni: always been reported on Precise so far | 14:51 |
| johngarbutt | Leander_:OK, just wanted to make sure you didn't loose your changes in restarting devstack. not sure how it works at run-time then. | 14:51 |
| trapni | ttx: I am having a dedicated nova-network node which never caused any issues, IMHO ... FYI | 14:52 |
| trapni | ttx: I wanted to state that, since on the nova-network gateway you've quite a few bridges, too. | 14:53 |
| Leander_ | johngarbutt: well i'm quite puzzled myself, i've already submitted a mail in the mailing list. Maybe someone there will shed some light on the subject | 14:53 |
| ttx | trapni: you should comment on that bug that you have the same issue, and maybe offer Serge some help in debugging it | 14:53 |
| trapni | ok | 14:53 |
| ttx | trapni: Serge Hallyn is <hallyn> on IRC | 14:53 |
| trapni | thanks | 14:54 |
| *** datsun180b has joined #openstack-dev | 14:54 | |
| *** anderstj has joined #openstack-dev | 15:01 | |
| *** cmagina has quit IRC | 15:01 | |
| *** danwent has joined #openstack-dev | 15:01 | |
| danpb | if someone in the core team +2's a patch in gerrit and I then -1 it, will it still get merged given that my -1 is not from a core reviewer ? | 15:06 |
| *** mindpixel has quit IRC | 15:06 | |
| *** reidrac has quit IRC | 15:06 | |
| *** wiliam has quit IRC | 15:12 | |
| *** thingee has quit IRC | 15:12 | |
| *** anderstj has quit IRC | 15:13 | |
| mtaylor | danpb: yes | 15:13 |
| mtaylor | danpb: well, it doesn't get merged until there is a +1 Approval vote - but -1 votes do not block thaqt | 15:13 |
| *** Mandell has joined #openstack-dev | 15:15 | |
| danpb | mtaylor: ah so a +1 Approval vote is an extra manual approval, separate from a +2 Code Review vote ? | 15:21 |
| *** e1mer has quit IRC | 15:21 | |
| * danpb had thought the +1 Approval was something automatically added once there was a +2 worth of Code Review votes from the core review team | 15:22 | |
| sandywalsh | can I get one more novaclient review here pulleaze? https://review.openstack.org/#/c/8672/ | 15:22 |
| *** davidha_who_took has joined #openstack-dev | 15:24 | |
| *** markmc has joined #openstack-dev | 15:25 | |
| mtaylor | danpb: yes | 15:26 |
| danpb | mtaylor: thanks for the explanations | 15:26 |
| mtaylor | danpb: it's a manual step, because those with +2 ability sometimes wanted to be able to vote +2 but still not trigger a merge that instant | 15:26 |
| mtaylor | danpb: sure thing! | 15:26 |
| *** winston-d has joined #openstack-dev | 15:28 | |
| apevec | sandywalsh, shouldn't keyring usage be optional ? | 15:29 |
| sandywalsh | apevec: it will silently fail if not available (when I remove the print statements) ... suggestions for other places to store token safely? | 15:30 |
| apevec | I mean "import keyring" | 15:30 |
| mnaser | Is there a way right now to get deleted instances from the OpenStack Nova API? | 15:37 |
| mnaser | opts = {'deleted': True} ----------- nova_api.servers.list(search_opts=opts) | 15:37 |
| mnaser | Returns only active ones | 15:37 |
| maoy | bcwaldon: this one got dinged by jenkins again.. could you help to re-approve? thanks.. https://review.openstack.org/#/c/8664/ | 15:38 |
| sandywalsh | apevec: oh, good point | 15:41 |
| andrewbogott | markmc, jkoelker: Could I get some more eyes on 8585 perchance? | 15:43 |
| *** agonella has left #openstack-dev | 15:43 | |
| *** Dr_Who has joined #openstack-dev | 15:45 | |
| *** Dr_Who has quit IRC | 15:45 | |
| *** Dr_Who has joined #openstack-dev | 15:45 | |
| mnaser | dprince: have you had time to take a look at this? https://bugs.launchpad.net/nova/+bug/1007116 | 15:47 |
| uvirtbot | Launchpad bug 1007116 in nova "nova should support showing 'DELETED' servers" [Low,In progress] | 15:47 |
| *** rgoodwin_ has joined #openstack-dev | 15:47 | |
| *** rgoodwin has quit IRC | 15:47 | |
| *** rgoodwin_ is now known as rgoodwin | 15:47 | |
| dprince | mnaser: not yet. I was looking into it and then got distracted. | 15:49 |
| mnaser | dprince: i can take care of it today as I need that functionality, the other thing is I was wondering if there was a way to show deleted images/etc to follow the same steps in here | 15:49 |
| mnaser | dprince: It'd be nice to be able to pull list of deleted servers as well as deleted time | 15:50 |
| *** johngarbutt has quit IRC | 15:50 | |
| dprince | mnaser: I think deleted images are already shown. At least that is what I mentioned in the ticket. | 15:50 |
| ttx | vishy: would love to have your input on https://lists.launchpad.net/openstack/msg13489.html | 15:50 |
| mnaser | dprince: yes but they don't seem to be shown in nova image-list (or at least a way to pull them out) | 15:51 |
| *** johngarbutt has joined #openstack-dev | 15:51 | |
| dprince | mnaser: feel free to grab the ticket though. | 15:51 |
| *** zykes- has quit IRC | 15:51 | |
| *** johngarbutt has quit IRC | 15:51 | |
| *** johngarbutt has joined #openstack-dev | 15:51 | |
| *** davidha_who_took is now known as davidha | 15:51 | |
| dprince | mnaser: That wasn't my intent here with /servers either. I was only suggesting we show a deleted /server on a GET. | 15:52 |
| mnaser | dprince: ah I see | 15:52 |
| mnaser | trying to figure out the best way to approach this sort of scenario in order to find deleted servers wtih their time, oh well | 15:52 |
| dprince | mnaser: The main point here was consistency I guess. Accross /flavors, /images, and /servers. There really is no contract that deleted servers will stay in the database forever anyway. A provider may delete them. | 15:52 |
| mnaser | dprince: I see, I guess in that way I'll have to listen in on the notification events and track those myself | 15:53 |
| sandywalsh | andrewbogott: I'll look at the other one after lunch | 15:55 |
| andrewbogott | sandywalsh: The other one's merged already; 8585 is all that's left at the moment. | 15:56 |
| andrewbogott | Once that one's in, though, there'll be the corresponding patch that rips logging and notifications out of nova. | 15:56 |
| BLZbubba | johngarbutt: it sure looks like it doesn't cache snapshots, i guess i should convert to images for any that I will want to launch regularly | 15:57 |
| *** jkyle has quit IRC | 15:59 | |
| *** jkyle has joined #openstack-dev | 15:59 | |
| johngarbutt | BLZbubba: didn't know that, interesting... | 15:59 |
| *** Mandell has quit IRC | 16:00 | |
| *** kindaopsdevy has joined #openstack-dev | 16:01 | |
| jaypipes | Anybody seeing this with latest commit? https://bugs.launchpad.net/nova/+bug/1017572 | 16:01 |
| uvirtbot | Launchpad bug 1017572 in nova "rootwrap iptables-save broken in recent build" [Undecided,New] | 16:01 |
| jaypipes | sandywalsh, dprince: if you look at https://bugs.launchpad.net/nova/+bug/1017572, any ideas? It's currently got me completely bottlenecked, as devstack/tempest won't run anything properly... | 16:03 |
| uvirtbot | Launchpad bug 1017572 in nova "rootwrap iptables-save broken in recent build" [Undecided,New] | 16:03 |
| mnaser | jaypipes: 2012-06-25 11:51:41 DEBUG nova.utils [-] ['sudo', '/usr/local/bin/nova-rootwrap', 'iptables-save', '-t', 'filter'] failed. Retrying. from (pid=22147) execute /opt/stack/nova/nova/utils.py:202 -- can the devstack used sudo? | 16:04 |
| mnaser | jaypipes: oh my bad, it can, incorrect configration file in stacktrace, sorry | 16:05 |
| jaypipes | :) no worries | 16:05 |
| *** lloydde has joined #openstack-dev | 16:06 | |
| jaypipes | mnaser: yeah, it looks like nova-rootwrap is expecting the first arg to be a config file now? is that right ttx? | 16:06 |
| mnaser | configfile = sys.argv.pop(0) | 16:06 |
| jaypipes | looks like the rootwrap commit needs to be reverted IMHO... commit 8fe25a36171f13d2c9413b3693e982169bdb32ff | 16:07 |
| jaypipes | how the hell did that pass gating? :( | 16:07 |
| mnaser | jaypipes: update your nova.conf? | 16:08 |
| mnaser | i'm assuming the new default should be what is listed in there: root_helper=sudo nova-rootwrap /etc/nova/rootwrap.conf | 16:08 |
| jaypipes | mnaser: so perhaps a devstack change is needed... | 16:08 |
| *** thingee has joined #openstack-dev | 16:09 | |
| mnaser | jaypipes: does devstack default to another differnent setting for the wrapper? not sure | 16:09 |
| *** Leander_ has quit IRC | 16:09 | |
| jaypipes | mnaser: not sure it sets anything at all :) | 16:09 |
| mnaser | which means nova should be setting the default that new setting if the root wrapper changed | 16:09 |
| jaypipes | mnaser: ah, yes, it does: | 16:09 |
| z | jaypipes: congrats on the move to AT&T | 16:10 |
| jaypipes | add_nova_opt "root_helper=sudo $NOVA_ROOTWRAP" | 16:10 |
| *** sniperd has joined #openstack-dev | 16:10 | |
| jaypipes | z: thx mate | 16:10 |
| mnaser | there, that's probably why | 16:10 |
| jaypipes | z: where are you now? | 16:10 |
| z | jaypipes: LivingSocial | 16:10 |
| jaypipes | ah, nice :) | 16:10 |
| jaypipes | congrats to yoU! | 16:10 |
| jaypipes | mnaser: yep, looks like it... | 16:10 |
| z | jaypipes: thanks :-) its been really good fun so far, now if I could only molest Nova networking into doing what I want :P (long story) | 16:11 |
| mnaser | wonder why devstack wants to explicitly set it, weird | 16:11 |
| jaypipes | mnaser: thx for your help! | 16:12 |
| jaypipes | z: hehe\ | 16:12 |
| mnaser | jaypipes: np i guess :p | 16:12 |
| *** littleidea has joined #openstack-dev | 16:17 | |
| *** n0ano has joined #openstack-dev | 16:18 | |
| *** andrewsben_zz is now known as andrewsben | 16:22 | |
| *** jkelly has joined #openstack-dev | 16:25 | |
| *** danpb has quit IRC | 16:25 | |
| *** heckj has joined #openstack-dev | 16:28 | |
| *** vanchester has joined #openstack-dev | 16:33 | |
| *** maploin has quit IRC | 16:36 | |
| *** anderstj has joined #openstack-dev | 16:36 | |
| *** mdomsch has joined #openstack-dev | 16:38 | |
| BLZbubba | if a compute node dies, assuming NFS mounted /var/lib/nova/instances, what does it take to bring the vm's up on another node? i tried doing this recently but there was a lot of libvirt goo that was stored in non-NFS areas | 16:40 |
| *** issackelly has joined #openstack-dev | 16:43 | |
| mtaylor | bcwaldon, jk0: quick question re: novaclient versioning | 16:46 |
| jk0 | sup | 16:46 |
| mtaylor | jk0: so, to get the post_version patch landed, it means that new versions to pypi will be based on tags (Which will make it easier to release, yay) | 16:46 |
| mtaylor | jk0: the outstanding question is - what should the next version be (I'm voting 2.6.1) | 16:47 |
| jk0 | I've no preferences as far as that's concerned | 16:47 |
| mtaylor | jk0: and is there a revision after the folsom-1 tag I can tag in the repo so that we're starting off on the right foot? | 16:47 |
| mtaylor | jk0: like, if I tagged curent tip as 2.6.1? | 16:47 |
| jk0 | that would be fine | 16:48 |
| mtaylor | jk0: awesome. thanks! | 16:48 |
| jk0 | nothing sticks out at me on the log | 16:49 |
| jk0 | np | 16:49 |
| mtaylor | yeah, it all seems fairly normal, and most things seem to be following tip right now anyway | 16:49 |
| *** yaguang has quit IRC | 16:50 | |
| BLZbubba | ok is it just me or is ubuntu server bad with nfs? in particular, if it can't mount nfs from fstab it just stops processing upstart altogether with no messages | 16:50 |
| *** salgado is now known as salgado-lunch | 16:51 | |
| *** jsavak has joined #openstack-dev | 16:55 | |
| *** pixelbeat has quit IRC | 16:56 | |
| *** zykes- has joined #openstack-dev | 16:57 | |
| *** issackelly has quit IRC | 16:58 | |
| *** joesavak has quit IRC | 16:58 | |
| mtaylor | jaypipes, jkoelker: https://review.openstack.org/#/c/8760/ if you get a second | 16:59 |
| *** issackelly has joined #openstack-dev | 16:59 | |
| cloudfly | so can we change horizon to display tenant instead of project? | 16:59 |
| cloudfly | this is really kind of annoying | 17:00 |
| *** erikzaadi has quit IRC | 17:01 | |
| *** darraghb has quit IRC | 17:02 | |
| *** torgomatic has joined #openstack-dev | 17:03 | |
| *** torgomatic has quit IRC | 17:05 | |
| *** torgomatic has joined #openstack-dev | 17:06 | |
| cloudfly | yeah seriously tempted to refactor the entire thing and submit | 17:07 |
| jaypipes | cloudfly: lol, I'd love that, but at the last design summit, people wanted to keep project and scrap tenant :( | 17:07 |
| *** sleepsonzzz is now known as sleepsonthefloor | 17:08 | |
| ijw | Is there any difference between the two or are they synonyms? | 17:08 |
| jaypipes | ijw: same. | 17:09 |
| ijw | Thought so. | 17:09 |
| cloudfly | well they are idiots | 17:11 |
| cloudfly | users need their tenant id to auth to keystone | 17:11 |
| cloudfly | so they come into horizon looking for it | 17:11 |
| cloudfly | and can't find it | 17:11 |
| cloudfly | even though its staring them in the face | 17:11 |
| *** cryptographrix has joined #openstack-dev | 17:12 | |
| cloudfly | it's silly | 17:12 |
| *** Slower has joined #openstack-dev | 17:12 | |
| cloudfly | it's meant to be less confusing but its more confusing | 17:12 |
| *** nati_ueno has joined #openstack-dev | 17:12 | |
| *** wiliam has joined #openstack-dev | 17:13 | |
| jsavak | heckj are we changing tenant to project in v3? Looked to be still tenant in the api doc | 17:13 |
| dolphm_ | cloudfly: it's going to get more confusing with 'domains' | 17:13 |
| cloudfly | i withhold judgement on that until i see the end result. | 17:14 |
| dolphm_ | v3 currently has 'domains' and 'tenants', where 'domains' are really 'tenants' and 'tenants' are really 'projects' | 17:14 |
| cloudfly | oof | 17:14 |
| cloudfly | sounds perfectly insidious | 17:14 |
| heckj | jsavak: I haven't done a global search and replace and proposed it yet. I was thinking of keeping domains and projects, not trying to re-use tenants because it's so damned confused. | 17:14 |
| *** harlowja has joined #openstack-dev | 17:14 | |
| jsavak | ok | 17:15 |
| cloudfly | ultimately the problem is keystone auth is like a day 1 or 2 problem for users | 17:15 |
| dolphm_ | heckj: domains+projects is a bit better than domains+tenants | 17:15 |
| cloudfly | and they need their tenant info to do it | 17:15 |
| cloudfly | and probably domain in the future | 17:15 |
| cloudfly | and so far all horizon offers as a clue is a downloadable rc file | 17:15 |
| heckj | looks like a SED hit in the near future | 17:16 |
| cloudfly | i mean i don't mind calling them projects but keystone and horizon have to agree | 17:16 |
| *** PotHix has joined #openstack-dev | 17:16 | |
| cloudfly | we can't have projects picking their own names for shared resources. | 17:16 |
| cloudfly | it leaves users utterly bewildered | 17:17 |
| *** ncode has quit IRC | 17:20 | |
| *** Hitesh has joined #openstack-dev | 17:21 | |
| Hitesh | danwent: Hi | 17:21 |
| *** mnaser has quit IRC | 17:23 | |
| heckj | dolphm_: could you look at https://review.openstack.org/#/c/8155/, https://review.openstack.org/#/c/8833/, and https://review.openstack.org/#/c/8909/ when you have a few today? | 17:23 |
| dolphm_ | heckj: absolutely | 17:24 |
| *** dachary has joined #openstack-dev | 17:24 | |
| heckj | dolphm_: thxusir! | 17:25 |
| *** mnaser has joined #openstack-dev | 17:25 | |
| dolphm_ | heckj: is service name really optional?? | 17:26 |
| heckj | dolphm_: huh? (what context?) | 17:26 |
| heckj | probably not, but I'm not sure what you're focused on | 17:27 |
| dolphm_ | heckj: https://review.openstack.org/#/c/8909/2/keystone/common/models.py | 17:30 |
| dolphm_ | heckj: Service model shows name as optional | 17:30 |
| jaypipes | cloudfly: I actually prefer Account to Tenant... see my comment on that same bug report... | 17:30 |
| *** garyk has quit IRC | 17:30 | |
| jaypipes | cloudfly: though I would suggest that calling people idiots ain't the best way to get your point across. | 17:31 |
| heckj | dolphm_: in the current code, it looks like it is optional - although we probably want to assert that it isn't. I'll update | 17:31 |
| cloudfly | heh sorry | 17:31 |
| jaypipes | don't need to say sorry to me.. | 17:31 |
| cloudfly | early monday | 17:31 |
| jaypipes | no worries. | 17:32 |
| *** salgado-lunch is now known as salgado | 17:32 | |
| *** johngarbutt has quit IRC | 17:32 | |
| *** Hitesh has quit IRC | 17:32 | |
| cloudfly | honestly i don't really care what its called so long as the name is the same across the board | 17:34 |
| cloudfly | as it is it's very confusing | 17:35 |
| *** johngarbutt has joined #openstack-dev | 17:37 | |
| heckj | dolphm_: https://review.openstack.org/#/c/8909/ updated | 17:37 |
| winston-d | smoser, ping | 17:44 |
| *** dachary has quit IRC | 17:45 | |
| *** dachary has joined #openstack-dev | 17:45 | |
| *** winston-d has quit IRC | 17:47 | |
| *** Mandell has joined #openstack-dev | 17:52 | |
| *** kindaopsdevy has quit IRC | 17:53 | |
| *** kindaopsdevy has joined #openstack-dev | 17:54 | |
| vishy | davidkranz: did you cut anything out of that log file? | 17:58 |
| davidkranz | vishy: Yes, but I still have the whole file. | 17:58 |
| bcwaldon | mtaylor: sir | 17:58 |
| *** jakedahn_zz is now known as jakedahn | 17:58 | |
| vishy | davidkranz: were there any obvious things going slowly in between? As in was it iptables calls? | 17:59 |
| davidkranz | vishy: It was hard to tell because there are so many things in the log happening at once. | 17:59 |
| *** kindaopsdevy_ has joined #openstack-dev | 18:00 | |
| vishy | davidkranz: so I'm honestly not all that concerned with not having speedup launching on a single box | 18:00 |
| *** sstent has quit IRC | 18:00 | |
| *** sstent has joined #openstack-dev | 18:00 | |
| vishy | davidkranz: there has been discussion about making everything synchronous on the compute host | 18:00 |
| vishy | davidkranz: but I am curious why it is going so slow, my guess is multiple calls to iptables | 18:01 |
| *** Mandell has quit IRC | 18:01 | |
| davidkranz | vishy: Yeah, you can see that they are not (call/return) pairs. | 18:01 |
| *** Mandell has joined #openstack-dev | 18:02 | |
| *** epim has joined #openstack-dev | 18:02 | |
| *** cryptographrix has quit IRC | 18:02 | |
| *** Ryan_Lane has quit IRC | 18:02 | |
| davidkranz | vishy: So you are saying that it is OK to be very slow to launch servers even on a large-multicore machine? | 18:02 |
| vishy | davidkranz: simultaneously yes | 18:03 |
| mtaylor | bcwaldon: sup? | 18:03 |
| vishy | davidkranz: i do see one potential for speedup | 18:03 |
| *** kindaopsdevy has quit IRC | 18:03 | |
| *** kindaopsdevy_ is now known as kindaopsdevy | 18:03 | |
| vishy | davidkranz: It would be interesting to see if it helps, I suspect it won't make much difference | 18:03 |
| bcwaldon | mtaylor: wrt novaclient versioning | 18:03 |
| bcwaldon | mtaylor: I'm too lazy to parse through the previous convo you had with jk0 | 18:04 |
| davidkranz | vishy: If you have a patch I can try it (after the current devstack total network lossage is fixed) | 18:04 |
| mtaylor | bcwaldon: oh, jk0 gave me good feedback | 18:04 |
| bcwaldon | mtaylor: what did you decide | 18:04 |
| *** cryptographrix has joined #openstack-dev | 18:04 | |
| mtaylor | bcwaldon: tag current tip 2.6.1 | 18:04 |
| vishy | davidkranz: actually i see one very quick change that might speedup slightly | 18:04 |
| bcwaldon | mtaylor: what will next release be? | 18:04 |
| vishy | davidkranz: and I see another change that will be a huge speedup | 18:04 |
| davidkranz | vishy: Cool. | 18:04 |
| *** johngarbutt has left #openstack-dev | 18:04 | |
| mtaylor | bcwaldon: it'll be whatever you want it to be :) | 18:04 |
| bcwaldon | mtaylor: version 3.0 | 18:05 |
| bcwaldon | mtaylor: I dont even know | 18:05 |
| bcwaldon | mtaylor: 'current tip' being whatever master is at this moment? | 18:05 |
| mtaylor | bcwaldon: yeah | 18:05 |
| bcwaldon | mtaylor: I'm inclined to say we should start fresh right now with v3.0 | 18:05 |
| bcwaldon | mtaylor: since we're changing the rules of versioning | 18:05 |
| mtaylor | bcwaldon: really? did we break api compat somewhere? or just clean slate? | 18:06 |
| bcwaldon | mtaylor: but I would be okay with v2.6.1 | 18:06 |
| *** Mandell has quit IRC | 18:06 | |
| bcwaldon | mtaylor: more of a clean slate thing | 18:06 |
| mtaylor | bcwaldon: makes sense | 18:06 |
| bcwaldon | mtaylor: our versioning made no sense up until this point | 18:06 |
| bcwaldon | mtaylor: it was basicall 'add 1 whenever you want to release' | 18:06 |
| mtaylor | bcwaldon: well... once we land that patch and I get one thing landed on jenkins ... releasing a 3.0 will be a really easy decision :) | 18:06 |
| bcwaldon | ok | 18:07 |
| bcwaldon | well, 3.0.0 | 18:07 |
| mtaylor | bcwaldon: after this initial set get up, i'm going to leave tag/version decisions fully in your hands | 18:07 |
| mtaylor | and butt my fat head out of ut | 18:07 |
| bcwaldon | mtaylor: whaaaaa | 18:07 |
| mtaylor | it | 18:07 |
| mtaylor | :) | 18:07 |
| bcwaldon | well fine | 18:07 |
| vishy | davidkranz: here is the easy version: http://paste.openstack.org/show/18764/ | 18:08 |
| * mtaylor wants to enable folks and stuff | 18:08 | |
| vishy | see if it helps at all | 18:08 |
| bcwaldon | mtaylor: a fine goal to have | 18:08 |
| mtaylor | (leaves more time for me to drink the beers) | 18:08 |
| vishy | davidkranz: The hard version will take a bit of refactoring and memoization so I will only try it if that one doesn't work. | 18:08 |
| maoy | vishy: appreciate much if you could help approve this: https://review.openstack.org/#/c/8254/ | 18:09 |
| davidkranz | vishy: I will give it a try as soon as I can. Thanks. | 18:09 |
| vishy | maoy: done | 18:09 |
| maoy | vishy: thx. | 18:10 |
| davidkranz | jaypipes: Is there a work around for the currently failing devstack network issue? I get the same error as you and am stuck. | 18:10 |
| *** apevec has quit IRC | 18:10 | |
| jaypipes | davidkranz: yep, one sec | 18:11 |
| vishy | Vek: ping | 18:11 |
| vishy | davidkranz, jaypipes: link? I will help | 18:11 |
| jaypipes | davidha: http://paste.openstack.org/show/18765/ | 18:11 |
| jaypipes | vishy: it's a devstack typo... not a biggie. going to propose that patch shortly. | 18:12 |
| *** mdomsch has quit IRC | 18:17 | |
| *** cryptographrix has quit IRC | 18:18 | |
| *** cryptographrix has joined #openstack-dev | 18:20 | |
| *** kindaopsdevy_ has joined #openstack-dev | 18:22 | |
| jaypipes | davidkranz: turns out my devstack install was a few commits behind... just need to git checkout master && git pull. | 18:22 |
| jaypipes | vishy: ^^ sorry for the confusion... | 18:22 |
| *** kindaopsdevy has quit IRC | 18:23 | |
| *** kindaopsdevy_ is now known as kindaopsdevy | 18:23 | |
| *** roge has joined #openstack-dev | 18:24 | |
| *** adjohn has joined #openstack-dev | 18:25 | |
| bcwaldon | eglynn__: https://review.openstack.org/#/c/8868/ | 18:27 |
| bcwaldon | eglynn__: are you stable-maint? | 18:27 |
| eglynn__ | bcwaldon: alas, no ... I don't have plus-one-ability on the stable/* branches | 18:28 |
| bcwaldon | eglynn__: well dang | 18:28 |
| eglynn__ | bcwaldon: perhaps Chuck Short is your man? | 18:29 |
| markmc | bcwaldon, rule for stable is if someone from stable-maint proposes, just one +2 is needed | 18:29 |
| * markmc just approved | 18:29 | |
| markmc | bcwaldon, http://wiki.openstack.org/StableBranch#Gerrit | 18:29 |
| bcwaldon | markmc: I learned something new today... :) | 18:29 |
| markmc | bcwaldon, arbitrary rules ftw :) | 18:30 |
| bcwaldon | markmc: I've got some of those in Glance, too | 18:30 |
| markmc | bcwaldon, keeps folks on their toes | 18:30 |
| bcwaldon | markmc: that's my intention | 18:30 |
| zykes- | 3.0 of what ? | 18:31 |
| sandywalsh | andrewbogott: heh 8610 was already merged | 18:31 |
| sandywalsh | whoops | 18:31 |
| *** jdg has quit IRC | 18:33 | |
| *** eglynn__ is now known as eglynn | 18:36 | |
| cloudfly | hrmm does anyone know where a shift from tenant to project inside keystone is documented? | 18:38 |
| cloudfly | is it a bug ? | 18:38 |
| cloudfly | not seeing a blueprint anywhere | 18:38 |
| dolphm_ | cloudfly: afaik, it's only been discussed at the last two summits | 18:38 |
| dolphm_ | cloudfly: if we're going to make a change, it'd be rolled into https://docs.google.com/document/d/1_TkawQIa52eSBfS4pv_nx1SJeoBghIlGVZsRJJynKAM/edit#heading=h.gp0kxtkcktct | 18:39 |
| cloudfly | okay so a bunch of people are telling me its planned for folsom | 18:40 |
| cloudfly | but i can find no evidence of that | 18:40 |
| cloudfly | i mean that's a pretty big change to blindside people with | 18:41 |
| *** Aaton_off is now known as Aaton | 18:41 | |
| *** danwent has quit IRC | 18:43 | |
| dolphm_ | cloudfly: if it's being promoted from "under discussion" to "planned", it'll be in the next v3 draft :) | 18:43 |
| *** mnaser has quit IRC | 18:44 | |
| *** Ryan_Lane has joined #openstack-dev | 18:44 | |
| *** reed has joined #openstack-dev | 18:45 | |
| *** mnaser has joined #openstack-dev | 18:45 | |
| jaypipes | cloudfly: FWIW, Glance uses the term tenant everywhere, and not project. | 18:47 |
| cloudfly | dolphm_ sounds like a very bad idea | 18:47 |
| cloudfly | great | 18:47 |
| * cloudfly face palms | 18:47 | |
| *** vanchester has quit IRC | 18:48 | |
| dolphm_ | cloudfly: what's a bad idea? making the change, putting it in the v3 draft..? | 18:48 |
| cloudfly | without discussing it and getting it approved in advance? yes. | 18:48 |
| cloudfly | especially since we're nearly half way to folsom | 18:49 |
| dolphm_ | cloudfly: the v3 draft is explicitly intended for community review & feedback | 18:49 |
| *** dubsquared has quit IRC | 18:49 | |
| cloudfly | and its not published yet | 18:49 |
| cloudfly | and it's intended as a folsom change | 18:49 |
| dolphm_ | cloudfly: it's been posted to the mailing list at least twice (first and second drafts) | 18:50 |
| cloudfly | yeah but the launcpad link from the blueprint is wrong | 18:50 |
| mtaylor | bcwaldon: you always have good feedbacks and stuff - what are your thoughts on this idea: https://review.openstack.org/#/c/8912/ (would use for all projects if we like it) | 18:50 |
| mtaylor | jkoelker: ^^ | 18:50 |
| dolphm_ | cloudfly: link? | 18:50 |
| cloudfly | https://blueprints.launchpad.net/keystone/+spec/implement-v3-core-api | 18:51 |
| cloudfly | and the api specs all show tenant in use for v3 to date | 18:51 |
| cloudfly | going back to jaypipes point glance refers to tenants as well | 18:52 |
| cloudfly | it's a big change to start removing the term from use | 18:52 |
| dolphm_ | cloudfly: looks like something was renamed... i added a link to the latest draft | 18:52 |
| dolphm_ | cloudfly: yep | 18:52 |
| dolphm_ | cloudfly: getting out from under all the confusing/overloaded legacy terminology in an elegant manner is going to be difficult | 18:53 |
| jaypipes | mtaylor: FWIW, the code looked good but I abstained frmo the review because I'm not a git guru... | 18:54 |
| *** zul has quit IRC | 18:54 | |
| mtaylor | jaypipes: then just trust me and +2 anyway! ;) | 18:54 |
| jaypipes | hehe | 18:54 |
| *** zul has joined #openstack-dev | 18:54 | |
| *** jog0 has joined #openstack-dev | 18:54 | |
| cloudfly | honestly it sounds to me like folsom is going to end up worse off than essex with tenants half way removed. | 18:54 |
| *** littleidea has quit IRC | 18:55 | |
| bcwaldon | mtaylor: you've got feedback | 18:55 |
| *** littleidea has joined #openstack-dev | 18:56 | |
| *** littleidea has quit IRC | 19:00 | |
| *** sleepsonthefloor is now known as sleepsonzzz | 19:01 | |
| *** heckj has quit IRC | 19:01 | |
| *** littleidea has joined #openstack-dev | 19:01 | |
| mtaylor | bcwaldon: baller. thanks | 19:01 |
| mtaylor | bcwaldon: and because I care, I wrote you this patch: https://review.openstack.org/8949 | 19:02 |
| bcwaldon | mtaylor: awww, for me!? | 19:02 |
| *** rkukura has quit IRC | 19:02 | |
| bcwaldon | mtaylor: test-options vs test-optional? | 19:02 |
| mtaylor | bcwaldon: just to stay consistent with the conjugation of the other words | 19:04 |
| mtaylor | bcwaldon: test-requires + tests-optional seemed weird. optional seems like it would match required more | 19:04 |
| bcwaldon | mtaylor: test-requires vs pip-requires doesnt make sense either | 19:05 |
| bcwaldon | mtaylor: but I guess thats another discussion | 19:05 |
| mtaylor | bcwaldon: ++ | 19:05 |
| mtaylor | bcwaldon: I fixed the other thing - thanks, that was a holdover from an old iteration | 19:05 |
| vishy | mtaylor: any idea why the libvirt skips show up as OK instead of SKIP when they are skipped? | 19:07 |
| mtaylor | vishy: no. we should have fixed the bit where they were erroneously printing FAIL | 19:07 |
| mtaylor | vishy: but that would be unrelated. I'll poked and see what I can find | 19:07 |
| mtaylor | jkoelker: unless you have ideas ^^ | 19:08 |
| vishy | mtaylor: it is calling 85 raise nose.SkipTest(self.message) | 19:08 |
| jkoelker | python 2.6 or python2.7 | 19:08 |
| jkoelker | or on both? | 19:08 |
| *** littleidea has quit IRC | 19:09 | |
| *** littleidea has joined #openstack-dev | 19:09 | |
| vishy | jkoelker: the libvirt tests have been being skipped by the builders for a while | 19:09 |
| vishy | jkoelker: but they don't show up as SKIP in the log afaik | 19:09 |
| jkoelker | right, on both python 2.6 and python 2.7? | 19:10 |
| jkoelker | bbl, have to run to a meeting | 19:11 |
| *** kbringard has quit IRC | 19:12 | |
| *** andrewsmedina has joined #openstack-dev | 19:12 | |
| *** kbringard has joined #openstack-dev | 19:12 | |
| *** mrunge has joined #openstack-dev | 19:14 | |
| *** jakedahn is now known as jakedahn_zz | 19:14 | |
| kbringard | vishy: I have a quick ? re: multinic on diablo-stable… do you have a few moments? | 19:15 |
| vishy | kbringard: yes but I'm not sure if i will be able to help | 19:15 |
| kbringard | no worries if not | 19:15 |
| kbringard | so, using vlan manager… my understanding is that if I create 2 networks with nova-manage and use —project to assign them both to the same tenant, then each VM launched in that tenant network will get a NIC on each network assigned to it | 19:16 |
| kbringard | correct? | 19:16 |
| kbringard | if so, do you know off the top of your head if it's possible to select which VMs get connections on which networks at instantiation time? | 19:17 |
| vishy | kbringard: that is the idea but I'm not sure that it works | 19:17 |
| kbringard | ah, OK, good to know. I'll browse the API code and see if that sheds any light | 19:17 |
| kbringard | thanks :-) | 19:17 |
| *** jakedahn_zz is now known as jakedahn | 19:17 | |
| vishy | kbringard: there is an extension where you can specify a network id on boot but I have no idea if it works in diablo | 19:17 |
| kbringard | ah, OK, cool, I'll look around a bit more, thanks for pointing me in the right direction | 19:19 |
| *** novas0x2a|laptop has joined #openstack-dev | 19:20 | |
| *** zigo has quit IRC | 19:22 | |
| *** apevec has joined #openstack-dev | 19:23 | |
| zykes- | what was the 3.0 stuff? | 19:25 |
| *** garyk has joined #openstack-dev | 19:26 | |
| *** jpeeler has joined #openstack-dev | 19:27 | |
| davidkranz | jaypipes: I am not sure what you meant about your devstack install. I just pulled the latest and still get the "Incorrect configuration" problem :( | 19:29 |
| *** danwent has joined #openstack-dev | 19:30 | |
| andrewbogott | thanks jkoelker. | 19:30 |
| davidkranz | jaypipes: Actually, never mind. I think I know what may have happened. | 19:31 |
| *** spiffxp has joined #openstack-dev | 19:32 | |
| ijw | kbringard: With Quantum, http://docs.openstack.org/trunk/openstack-network/admin/content/specifying-vm-network-connectivity-on-boot.html works - haven't tried it without, don't know which bit of code is responsible for processing those options | 19:32 |
| kbringard | ijw: cool, thanks, that's good to know | 19:32 |
| ijw | (Essex and Folsom, I've done it under) | 19:33 |
| kbringard | ijw: but multi-nic should work to give every VM a nic in each network assigned to the tenant, in diblo, correct? | 19:33 |
| kbringard | diablo* | 19:33 |
| ijw | dildo* | 19:33 |
| ijw | (at least, that's what my dirty mind told me) | 19:33 |
| *** krtaylor has quit IRC | 19:34 | |
| ijw | Yes, that's what happens by default, though again I've only tried it with Quantum. Quantum's supposed to respect nova-network behaviour though so I assume that it's default behaviour it's emulating | 19:34 |
| ijw | Pretty straightforward to test if you have an env where you can create the second network. | 19:34 |
| kbringard | ijw: totally, that's about all diablo is good for these days ;-) | 19:35 |
| kbringard | yea, I'm going to test it and implement it, but wanted to ask if someone knew what the expected behavior is | 19:36 |
| kbringard | cheers, I appreciate the input | 19:36 |
| *** mnaser has quit IRC | 19:36 | |
| *** mrunge has quit IRC | 19:36 | |
| *** apevec has quit IRC | 19:37 | |
| *** RoboticCheese has joined #openstack-dev | 19:38 | |
| *** mnaser has joined #openstack-dev | 19:39 | |
| *** drewlander has quit IRC | 19:42 | |
| *** anderstj has quit IRC | 19:42 | |
| *** novas0x2a|laptop has quit IRC | 19:43 | |
| *** jdg has joined #openstack-dev | 19:43 | |
| *** ewindisch has joined #openstack-dev | 19:48 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:48 | |
| *** kindaopsdevy_ has joined #openstack-dev | 19:48 | |
| *** apevec has joined #openstack-dev | 19:50 | |
| *** kindaopsdevy has quit IRC | 19:52 | |
| *** kindaopsdevy_ is now known as kindaopsdevy | 19:52 | |
| *** sieutruc has joined #openstack-dev | 19:54 | |
| *** dubsquared has joined #openstack-dev | 19:55 | |
| *** sieutruc has quit IRC | 19:56 | |
| *** sieutruc has joined #openstack-dev | 19:56 | |
| * andrewbogott desperately craves one more openstack-common reviewer for https://review.openstack.org/#/c/8585/ | 19:57 | |
| *** littleidea has quit IRC | 19:59 | |
| *** kindaopsdevy has quit IRC | 20:00 | |
| *** kindaopsdevy has joined #openstack-dev | 20:01 | |
| *** pixelbeat has joined #openstack-dev | 20:09 | |
| *** mnaser has quit IRC | 20:11 | |
| *** mnaser has joined #openstack-dev | 20:12 | |
| *** japage has quit IRC | 20:13 | |
| *** wiliam has quit IRC | 20:16 | |
| *** markvoelker has quit IRC | 20:21 | |
| *** sleepsonzzz is now known as sleepsonthefloor | 20:22 | |
| *** s0mik has joined #openstack-dev | 20:25 | |
| ttx | jaypipes: yes | 20:28 |
| *** sleepsonthefloor is now known as sleepsonzzz | 20:28 | |
| *** alaski has quit IRC | 20:28 | |
| ttx | jaypipes: you might need to update your root_helper | 20:29 |
| ttx | jaypipes: oh, I see it's solved now, good. | 20:30 |
| * ttx expects a bit of transition trouble until the distros catch up anyway | 20:32 | |
| *** dolphm_ has quit IRC | 20:35 | |
| *** sleepsonzzz is now known as sleepsonthefloor | 20:36 | |
| ttx | annegentle: ping | 20:37 |
| annegentle | pong | 20:38 |
| ttx | annegentle: started working on documentation around rootwrap. Pinged you earlier to see where I should put the WIP... settled on http://wiki.openstack.org/Nova/Rootwrap | 20:38 |
| annegentle | ttx: ah, thanks for re-pinging, had not seen this. Reading now. | 20:39 |
| ttx | annegentle: will let you know when I'm done with it, but wanted to know if the format was useful at all | 20:39 |
| ttx | or if I should just do it another way | 20:40 |
| *** krtaylor has joined #openstack-dev | 20:40 | |
| *** dtroyer is now known as dtroyer_zzz | 20:40 | |
| annegentle | ttx: I think the docs.openstack.org would take the "Architecture" and "Rootwrap for users" sections, those are extremely useful, thanks. All of it can live on the wiki too since there are multiple audiences. | 20:41 |
| maoy | vishy: if you got some cycles.. could you kick jenkins to rerun this: https://review.openstack.org/#/c/8664/ | 20:41 |
| vishy | davidkranz: so no luck on the little change :( | 20:41 |
| annegentle | ttx: it's definitely one of those concepts that affects many | 20:41 |
| *** matwood has quit IRC | 20:42 | |
| ttx | annegentle: exactly, that's why I wrote the "audience-oriented" sections | 20:42 |
| annegentle | ttx: is root_helper in essex? | 20:42 |
| ttx | annegentle: yes | 20:42 |
| davidkranz | vishy: No, the instances never get out of BUILD | 20:42 |
| vishy | davidkranz: any error? I was hoping we didn't have to do an external lock on those but apparently we do. | 20:42 |
| annegentle | ttx: ok, found it. I'll add to the docs, in the config chapter for Compute I believe. | 20:42 |
| ttx | annegentle: but used to be just root_helper=sudo nova-rootwrap | 20:42 |
| ttx | annegentle: this covers the stuff that just went in | 20:43 |
| davidkranz | vishy: I didn't see any error. | 20:43 |
| annegentle | ttx: yeah essex docs has sudo nova-rootwrap | 20:43 |
| *** matwood has joined #openstack-dev | 20:43 | |
| vishy | davidkranz: the other fix I was thinking was to actually check to make sure something has changed before doing an iptables.apply() | 20:43 |
| ttx | annegentle: I'll let you know when I'm done with the docs and you can take them. Probably tomorrow or wednesday. | 20:43 |
| vishy | davidkranz: which might help significantly | 20:43 |
| annegentle | ttx: perfect, I'll log a "doc bug as task" today | 20:43 |
| davidkranz | vishy: I think that would make a lot of sense in this case. The slow call would not happen, like batching. | 20:43 |
| ttx | just wanted to check if I was not completely off-track with the format. | 20:44 |
| vishy | davidkranz: let me try a quick version to see if it helps | 20:45 |
| davidkranz | vishy: k | 20:45 |
| *** dprince has quit IRC | 20:46 | |
| *** matwood has quit IRC | 20:46 | |
| annegentle | ttx: in Essex, did the rootwrap.conf file exist at all? Is there a sample file available? (I can't click any of the files under "Reference") | 20:46 |
| *** littleidea has joined #openstack-dev | 20:47 | |
| andrewbogott | dhellmann, If you're a core committer for common (I think you are), could you grant 8585 a checkmark? | 20:48 |
| * andrewbogott fears he is turning into a nag | 20:49 | |
| mtaylor | dhellmann: if you're doing common reviews... review me too... | 20:49 |
| mtaylor | andrewbogott: I think it's part of the game :) | 20:49 |
| *** ncode has joined #openstack-dev | 20:50 | |
| *** ncode has quit IRC | 20:50 | |
| *** ncode has joined #openstack-dev | 20:50 | |
| *** markvoelker has joined #openstack-dev | 20:51 | |
| *** Aaton is now known as Aaton_off | 20:52 | |
| *** littleidea has quit IRC | 20:53 | |
| *** littleidea has joined #openstack-dev | 20:53 | |
| *** ncode has quit IRC | 20:54 | |
| *** ncode has joined #openstack-dev | 20:54 | |
| dhellmann | andrewbogott and mtaylor, I'll do common reviews shortly. I want to get these ceilometer jobs running cleanly first | 20:55 |
| *** anderstj has joined #openstack-dev | 20:55 | |
| *** roge has quit IRC | 20:56 | |
| *** epim has quit IRC | 20:56 | |
| *** edygarcia has quit IRC | 20:58 | |
| *** mestery_ is now known as mestery | 20:58 | |
| *** edygarcia has joined #openstack-dev | 20:58 | |
| *** dolphm has joined #openstack-dev | 21:00 | |
| *** rkukura has joined #openstack-dev | 21:02 | |
| *** derekh has quit IRC | 21:02 | |
| *** edygarcia has quit IRC | 21:03 | |
| *** davidha has quit IRC | 21:03 | |
| vishy | davidkranz: I think the provider rules might be a red herring | 21:03 |
| vishy | davidkranz: it is probably just blocking because there are a whole bunch of iptables rules that get remade every single time. | 21:03 |
| *** edygarcia has joined #openstack-dev | 21:04 | |
| vishy | davidkranz: can you see if this (untested) patch makes a difference though? http://paste.openstack.org/show/18777/ | 21:04 |
| *** epim has joined #openstack-dev | 21:04 | |
| andrewbogott | mtaylor: I guess it won't do any good to nag Jenkins | 21:04 |
| *** markmc has quit IRC | 21:04 | |
| davidkranz | vishy: OK, I'll do it now. | 21:04 |
| mtaylor | andrewbogott: not as much as you might hope :) | 21:05 |
| *** lts has quit IRC | 21:06 | |
| *** davidha has joined #openstack-dev | 21:07 | |
| *** eglynn has quit IRC | 21:08 | |
| *** steveb_ has joined #openstack-dev | 21:08 | |
| andrewbogott | What does it mean that Jenkins isn't automatically assigned patches in openstack-common? Does it not know how to run tests there? | 21:08 |
| davidkranz | vishy: I am not a git expert and get an error doing 'git apply your-patch-file'. How do I apply this patch? | 21:10 |
| vishy | davidkranz: applying to current nova master? | 21:11 |
| vishy | davidkranz: don't apply it on top of my other patch... | 21:11 |
| davidkranz | vishy: Yes, I get 'fatal: corrupt patch at line 79'. I already reverted the other patch. | 21:12 |
| davidkranz | Line 79 is the end of the file. | 21:12 |
| vishy | davidkranz: might have pastied it wrong | 21:12 |
| vishy | let me try again | 21:12 |
| davidkranz | vishy: I got it. git didn't like that there was no newline at the end of the file. | 21:13 |
| vishy | davidkranz: try here: https://raw.github.com/gist/2991284/5c708d267dd053748b5735dc7d7fc9a314150596/gistfile1.txt | 21:13 |
| vishy | oh cool then | 21:13 |
| *** sleepsonthefloor is now known as sleepsonzzz | 21:16 | |
| *** littleidea has quit IRC | 21:17 | |
| *** littleidea has joined #openstack-dev | 21:18 | |
| *** rods has quit IRC | 21:18 | |
| *** davidha has quit IRC | 21:24 | |
| *** jsavak has quit IRC | 21:27 | |
| davidkranz | vishy: So this seems to have some effect. It is now taking about 40s to do 4. I think it was more than a minute before. | 21:27 |
| davidkranz | vishy: I have to run now but will revert and make sure of that tomorrow. Thanks. | 21:27 |
| vishy | davidkranz: ok cool | 21:28 |
| vishy | davidkranz: let me know and I will clean it up and propose it if it helps. | 21:28 |
| davidkranz | vishy: Will do. | 21:28 |
| *** dubsquared has quit IRC | 21:29 | |
| *** dubsquared has joined #openstack-dev | 21:33 | |
| *** sleepsonzzz is now known as sleepsonthefloor | 21:40 | |
| *** ewindisch has quit IRC | 21:40 | |
| *** littleidea has quit IRC | 21:45 | |
| *** markvoelker has quit IRC | 21:47 | |
| *** segfault923 has quit IRC | 21:48 | |
| *** ewindisch has joined #openstack-dev | 21:49 | |
| *** littleidea has joined #openstack-dev | 21:50 | |
| *** ewindisch has quit IRC | 21:50 | |
| *** rkukura has left #openstack-dev | 21:58 | |
| sdague | vishy: can you re-approve - https://review.openstack.org/#/c/8778/. That area is changing fast enough that it got hit by a new test added that used the old namespace | 21:58 |
| sdague | I just rebased, should be good now | 21:58 |
| *** dolphm has quit IRC | 21:59 | |
| *** ywu has joined #openstack-dev | 21:59 | |
| *** roge has joined #openstack-dev | 22:00 | |
| *** andrewsben is now known as andrewsben_is_a_ | 22:01 | |
| *** andrewbogott has quit IRC | 22:01 | |
| clarkb | jkoelker: ping | 22:02 |
| *** ayoung has quit IRC | 22:02 | |
| clarkb | jkoelker: https://review.openstack.org/#/c/8966/ passes pep8 when I run tox. Where/when is it failing? | 22:02 |
| vishy | sdague: https://review.openstack.org/#/c/8778/ | 22:03 |
| jkoelker | ah, it failed locally for me, I probably just need to regen my tox env | 22:03 |
| *** eglynn has joined #openstack-dev | 22:03 | |
| clarkb | k | 22:04 |
| *** salgado is now known as salgado-afk | 22:04 | |
| jkoelker | clarkb: I'm regenerating it now, if it passed I'll push it through | 22:04 |
| *** Dr_Who has quit IRC | 22:04 | |
| clarkb | jkoelker: in the results you pasted there are two spaces after return. In the copy of foxinsocks.py that I have there is only one | 22:05 |
| jkoelker | i may have had a back checkout | 22:05 |
| sdague | vishy: that's fixed in patch3 that's up there | 22:05 |
| sdague | did you happen to grab the old patch? | 22:06 |
| clarkb | jkoelker: also once the tox.ini stuff is pushed https://review.openstack.org/#/c/8957/ can land | 22:06 |
| *** ncode has quit IRC | 22:06 | |
| vishy | sdague: ah gotcha | 22:06 |
| vishy | sdague: missed that was from patch set 2 | 22:07 |
| vishy | :) | 22:07 |
| sdague | vishy: :) | 22:07 |
| jkoelker | clarkb: pushed through, gitweb was showing what you have as well | 22:07 |
| jkoelker | my repo is probably just jacked | 22:07 |
| sdague | yeh, the libvirt driver is active enough this patch has to land fast after a rebase, otherwise it inevitably conflicts | 22:08 |
| *** krtaylor has quit IRC | 22:08 | |
| clarkb | jkoelker: great. Thank you | 22:09 |
| *** ncode has joined #openstack-dev | 22:13 | |
| *** zykes- has quit IRC | 22:17 | |
| *** anderstj has quit IRC | 22:20 | |
| *** apevec has quit IRC | 22:20 | |
| *** kbringard has quit IRC | 22:23 | |
| *** ewindisch has joined #openstack-dev | 22:25 | |
| *** ncode has quit IRC | 22:25 | |
| *** ewindisch has quit IRC | 22:25 | |
| *** ncode has joined #openstack-dev | 22:25 | |
| *** ncode has quit IRC | 22:25 | |
| *** ncode has joined #openstack-dev | 22:25 | |
| *** ncode has quit IRC | 22:27 | |
| *** ncode has joined #openstack-dev | 22:28 | |
| *** ncode has quit IRC | 22:28 | |
| *** ncode has joined #openstack-dev | 22:28 | |
| *** dachary has quit IRC | 22:33 | |
| *** s0mik has quit IRC | 22:51 | |
| *** s0mik has joined #openstack-dev | 22:53 | |
| *** datsun180b has quit IRC | 22:57 | |
| *** edygarcia has quit IRC | 22:57 | |
| vishy | comstud: ping | 22:59 |
| *** sleepsonthefloor is now known as sleepsonzzz | 22:59 | |
| comstud | pong | 22:59 |
| bcwaldon | comstud: ping | 22:59 |
| comstud | PONG | 22:59 |
| bcwaldon | vishy: ping | 22:59 |
| vishy | comstud: so there are a couple of extensions which also slow things down. I was going to try and add another one but I didn't want to add more calls | 22:59 |
| vishy | bcwaldon: plop | 22:59 |
| bcwaldon | aw | 22:59 |
| comstud | vishy: I have a patch for that too | 22:59 |
| *** sleepsonzzz is now known as sleepsonthefloor | 22:59 | |
| comstud | vishy: :) | 22:59 |
| comstud | vishy: But i'm not sure you'll like it.. for trunk | 23:00 |
| vishy | comstud: I think we need to refactor the extension stuff…ooh really? | 23:00 |
| comstud | yes | 23:00 |
| vishy | comstud: for the output extensions? | 23:00 |
| comstud | yes | 23:00 |
| comstud | the extra DB calls they do | 23:00 |
| comstud | i hide the results from the core API.. | 23:00 |
| comstud | in 'req' | 23:00 |
| comstud | and re-use them in the extensions | 23:00 |
| comstud | I'll throw it up if you want to look it over | 23:00 |
| *** segfault923 has joined #openstack-dev | 23:00 | |
| vishy | comstud: me waits to see how they work. I would like to turn the createserverext extension that adds security groups to the output to work that way | 23:01 |
| comstud | I'm not sure about the approach | 23:01 |
| comstud | the extensions really should be refactored differently | 23:01 |
| comstud | to support this | 23:01 |
| comstud | but it was a quick hack.. | 23:01 |
| comstud | i'm working on getting some of our very recent local patches into trunk | 23:01 |
| comstud | that improve perf. | 23:01 |
| vishy | comstud: well, I think it makes sense to put the data from the request somewhere that the output extensions can access it | 23:01 |
| vishy | comstud: it is silly to re-request status and secgroup etc. | 23:02 |
| comstud | yes. | 23:02 |
| comstud | agree! | 23:02 |
| comstud | this cuts 'nova list' time down a lot | 23:02 |
| comstud | on lots of instances | 23:02 |
| comstud | with the extensions loaded | 23:02 |
| comstud | we need further audit on the joins we're doing, also | 23:02 |
| vishy | comstud: perhaps we should just cache the results in the db layer? | 23:02 |
| comstud | based on what I found with sec groups | 23:02 |
| vishy | comstud: agreed :) | 23:03 |
| comstud | i'm in the middle of something else here i'm finishing up.. | 23:04 |
| *** matwood has joined #openstack-dev | 23:04 | |
| comstud | then I'll throw up that patch | 23:05 |
| comstud | at least for you to see | 23:05 |
| vishy | comstud: cool, no worries | 23:05 |
| comstud | i'm rebasing my cells branch | 23:05 |
| comstud | based on that email | 23:05 |
| *** littleidea has quit IRC | 23:05 | |
| comstud | lol | 23:05 |
| *** Aaton_off is now known as Aaton | 23:06 | |
| *** littleidea has joined #openstack-dev | 23:07 | |
| *** pixelbeat has quit IRC | 23:07 | |
| *** andrewsben_is_a_ is now known as andrewsben | 23:09 | |
| *** dolphm has joined #openstack-dev | 23:10 | |
| *** dtroyer_zzz is now known as dtroyer | 23:11 | |
| *** lloydde has quit IRC | 23:14 | |
| *** dolphm has quit IRC | 23:14 | |
| *** mestery_ has joined #openstack-dev | 23:16 | |
| *** tgall_foo has quit IRC | 23:17 | |
| *** thingee has quit IRC | 23:17 | |
| *** mestery has quit IRC | 23:17 | |
| *** markmcclain has quit IRC | 23:20 | |
| *** salv-orlando has left #openstack-dev | 23:20 | |
| *** e1mer has joined #openstack-dev | 23:23 | |
| *** e1mer has joined #openstack-dev | 23:23 | |
| *** epim has quit IRC | 23:25 | |
| *** sleepsonthefloor is now known as sleepsonzzz | 23:27 | |
| *** dubsquared has quit IRC | 23:27 | |
| comstud | vishy: https://review.openstack.org/#/c/8976/ | 23:27 |
| *** dabo has quit IRC | 23:28 | |
| comstud | vishy: curious your feedback on that | 23:28 |
| *** dtroyer is now known as dtroyer_zzz | 23:30 | |
| *** dabo has joined #openstack-dev | 23:30 | |
| vishy | comstud: I can't really think of a better approach in the short term | 23:31 |
| *** markmcclain has joined #openstack-dev | 23:31 | |
| *** vuntz has quit IRC | 23:31 | |
| vishy | comstud: Vek might have some ideas though | 23:31 |
| *** garyk has quit IRC | 23:31 | |
| *** vuntz has joined #openstack-dev | 23:31 | |
| comstud | vishy: Nod | 23:31 |
| *** lcheng has quit IRC | 23:31 | |
| comstud | I haven't shown this to him yet | 23:32 |
| *** taziden has quit IRC | 23:32 | |
| comstud | or, explicitly pointed him to it anyway | 23:32 |
| *** lcheng has joined #openstack-dev | 23:32 | |
| comstud | this feels hacky... but works. | 23:32 |
| vishy | comstud: I would really like to get rid of createserverext which currently really just seems to be there to add security groups to the response | 23:32 |
| comstud | been running for a couple weeks :) | 23:32 |
| comstud | nod | 23:32 |
| vishy | comstud: so this would give me a way to do it. Maybe ping Vek and see if he has any ideas | 23:33 |
| *** taziden has joined #openstack-dev | 23:33 | |
| *** cryptographrix has quit IRC | 23:33 | |
| *** matwood has quit IRC | 23:33 | |
| vishy | comstud: I'd really like to get it fixed soon. | 23:33 |
| comstud | Sure | 23:34 |
| *** sieutruc has quit IRC | 23:34 | |
| comstud | doing so now | 23:34 |
| *** _0x44 has quit IRC | 23:34 | |
| *** reed has quit IRC | 23:34 | |
| *** sieutruc has joined #openstack-dev | 23:34 | |
| *** dabo has quit IRC | 23:34 | |
| *** dabo has joined #openstack-dev | 23:35 | |
| *** _0x44 has joined #openstack-dev | 23:36 | |
| *** jakedahn is now known as jakedahn_zz | 23:36 | |
| bcwaldon | mtaylor: I have a bone to pick with you | 23:36 |
| *** issackelly has quit IRC | 23:37 | |
| *** reed has joined #openstack-dev | 23:37 | |
| comstud | vishy: He just disappeared.. might have left for home | 23:37 |
| comstud | vishy: Anyway.. he should see I pointed him at it | 23:37 |
| *** issackelly has joined #openstack-dev | 23:37 | |
| *** jakedahn_zz is now known as jakedahn | 23:37 | |
| vishy | comstud: doh. I noticed he doesn't respond on irc much, so email/gerrit is probably the best bet. | 23:38 |
| comstud | i'm at castle | 23:38 |
| comstud | just saw him | 23:38 |
| comstud | :) | 23:38 |
| comstud | he pays more attention to our internal server | 23:38 |
| *** cryptographrix has joined #openstack-dev | 23:39 | |
| *** tgall_foo has joined #openstack-dev | 23:39 | |
| *** pandemicsyn has quit IRC | 23:39 | |
| comstud | we have another pep8 failure in trunk | 23:39 |
| comstud | nova/tests/test_xenapi.py:46:1: NOVA N306: imports not in alphabetical order (nova.virt.xenapi.volume_utils, nova.tests.api.openstack.fakes) | 23:39 |
| *** sieutruc has quit IRC | 23:40 | |
| *** littleidea has quit IRC | 23:41 | |
| comstud | vishy: I guess he saw it | 23:41 |
| *** littleidea has joined #openstack-dev | 23:42 | |
| vishy | comstud: yeah gate doesn't use hacking.py | 23:42 |
| *** garyk has joined #openstack-dev | 23:43 | |
| *** pandemicsyn has joined #openstack-dev | 23:43 | |
| *** ChanServ sets mode: +v pandemicsyn | 23:43 | |
| comstud | ahh ok | 23:43 |
| *** BLZbubba has quit IRC | 23:45 | |
| *** BLZbubba has joined #openstack-dev | 23:46 | |
| comstud | gotta run for now | 23:51 |
| comstud | laters | 23:51 |
| * comstud & | 23:51 | |
| mtaylor | bcwaldon: what do I this time? | 23:54 |
| bcwaldon | mtaylor: I want a better way to respond to comments on a review | 23:54 |
| bcwaldon | mtaylor: not everything is a 'Review'! | 23:54 |
| bcwaldon | mtaylor: replying inline is a PITA | 23:55 |
| mtaylor | bcwaldon: yeah. I agree with you | 23:55 |
| bcwaldon | mtaylor: fix it! | 23:55 |
| bcwaldon | mtaylor: fix it fix it | 23:55 |
| mtaylor | bcwaldon: we have a bug open about even | 23:55 |
| bcwaldon | mtaylor: whoa! | 23:55 |
| mtaylor | bcwaldon: so it has not escaped notice! | 23:55 |
| bcwaldon | mtaylor: well thats good to hear | 23:55 |
| bcwaldon | mtaylor: any progress on fixing? | 23:55 |
| *** andrewsben is now known as andrewsben_is_a_ | 23:57 | |
| *** jkelly has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!