| *** maurosr has joined #openstack-dev | 00:00 | |
| *** renier_ has joined #openstack-dev | 00:01 | |
| *** dolphm has quit IRC | 00:03 | |
| *** renier has quit IRC | 00:03 | |
| *** matwood has joined #openstack-dev | 00:03 | |
| *** anniec_ has joined #openstack-dev | 00:05 | |
| *** andrewsmedina has joined #openstack-dev | 00:06 | |
| andrewsmedina | danwent: hi, I cant reproduce this https://bugs.launchpad.net/quantum/+bug/1009215. Have you another issue for me? | 00:07 |
|---|---|---|
| uvirtbot | Launchpad bug 1009215 in quantum "During resuming instances get error "could not open /dev/net/tun: Operation not permitted"" [High,Confirmed] | 00:07 |
| *** anniec has quit IRC | 00:07 | |
| *** anniec_ is now known as anniec | 00:07 | |
| danwent | huh… that's odd, as arosen said he would reproduce. | 00:08 |
| danwent | let me forward you an email with another bit of work i'd love to see done. | 00:08 |
| arosen | hi andrewsmedina did you reboot the hypervisor the vm was running on after suspending it? | 00:08 |
| danwent | do I have your email? if not, can you send it to me? i'm dan at nicira.com | 00:08 |
| *** roge has quit IRC | 00:09 | |
| danwent | andrewsmedina: also, if we can repro, please update the bug saying that so that the filer can possibly respond with more info | 00:09 |
| danwent | and so we know that we've done our best to explore it | 00:09 |
| *** renier_ has quit IRC | 00:09 | |
| *** jiangyh has joined #openstack-dev | 00:11 | |
| andrewsmedina | danwent: andrewsmedina@gmail.com | 00:11 |
| *** colinmcnamara has quit IRC | 00:12 | |
| andrewsmedina | arosen: I will try again | 00:12 |
| *** renier has joined #openstack-dev | 00:13 | |
| arosen | andrewsmedina: cool, let me know if you need more info about it. | 00:14 |
| *** alanmac has quit IRC | 00:17 | |
| *** Mandell_ has quit IRC | 00:22 | |
| *** huats has quit IRC | 00:24 | |
| *** lcheng has quit IRC | 00:26 | |
| *** AlanClark has quit IRC | 00:26 | |
| *** matwood has quit IRC | 00:28 | |
| *** huats has joined #openstack-dev | 00:29 | |
| *** huats has quit IRC | 00:29 | |
| *** huats has joined #openstack-dev | 00:29 | |
| *** sunxin has joined #openstack-dev | 00:29 | |
| *** halfss_ has quit IRC | 00:31 | |
| *** ncode has quit IRC | 00:33 | |
| *** johnpur has left #openstack-dev | 00:34 | |
| *** renier_ has joined #openstack-dev | 00:35 | |
| *** maurosr has quit IRC | 00:36 | |
| *** rods has quit IRC | 00:36 | |
| *** renier has quit IRC | 00:36 | |
| *** avishay has quit IRC | 00:37 | |
| *** avishay has joined #openstack-dev | 00:38 | |
| *** lloydde_ has quit IRC | 00:38 | |
| andrewsmedina | arosen: I tried several times and I can't reproduce it =/ | 00:42 |
| *** s0mik has quit IRC | 00:44 | |
| *** avishay has quit IRC | 00:44 | |
| *** avishay has joined #openstack-dev | 00:45 | |
| *** novas0x2a|laptop has quit IRC | 00:46 | |
| *** ayoung has quit IRC | 00:49 | |
| *** renier_ has quit IRC | 00:49 | |
| arosen | andrewsmedina are you doing this on ubuntu? | 00:51 |
| andrewsmedina | arosen: yes | 00:52 |
| *** avishay has quit IRC | 00:52 | |
| andrewsmedina | arosen: I'm on ubuntu 12.04 and you? | 00:52 |
| arosen | andrewsmedina: yup, and you are using 2 nodes right? | 00:52 |
| *** issackelly has quit IRC | 00:52 | |
| *** mnewby has quit IRC | 00:53 | |
| arosen | it might work for one but I only tested this restarting the node running n-cpu. | 00:53 |
| arosen | andrewsmedina: so you are saying you could do nova suspend vm | 00:53 |
| andrewsmedina | arosen: I'm using one node | 00:53 |
| *** avishay has joined #openstack-dev | 00:53 | |
| arosen | Then reboot the the Hypervervisor the node was running on | 00:53 |
| andrewsmedina | arosen: I'm suspend a vm, reboot the host machine | 00:54 |
| *** sandywalsh has joined #openstack-dev | 00:54 | |
| arosen | and then call nova resume again? | 00:54 |
| *** avishay has quit IRC | 00:54 | |
| andrewsmedina | arosen: I'm resume the suspened vm and it works | 00:54 |
| arosen | and the vm resumes. | 00:54 |
| andrewsmedina | arosen: yes | 00:54 |
| arosen | andrewsmedina: give me a sec let me see if i can reproduce this with one node. | 00:55 |
| *** yuanz1 has joined #openstack-dev | 00:55 | |
| *** avishay has joined #openstack-dev | 00:55 | |
| *** harlowja has quit IRC | 00:55 | |
| andrewsmedina | arosen: ok, I will try install another node here. | 00:55 |
| *** renier has joined #openstack-dev | 00:56 | |
| *** dolphm has joined #openstack-dev | 00:58 | |
| arosen | andrewsmedina you are right I can not reproduce this with one node | 01:01 |
| arosen | I could suspend and resume | 01:01 |
| *** yuanz1 has left #openstack-dev | 01:02 | |
| arosen | When I did this though I had one node running with: ENABLED_SERVICES="g-api,g-reg,key,n-api,n-cpu,n-sch,n-vnc,mysql,rabbit,openstackx,q-svc,quantum,q-agt,n-cpu,q-dhcp" | 01:02 |
| *** zhuadl has quit IRC | 01:02 | |
| *** avishay has quit IRC | 01:03 | |
| arosen | and another node just running ENABLED_SERVICES=n-cpu,rabbit,g-api,quantum,q-agt,q-dhcp | 01:03 |
| arosen | andrewsmedina: i'm going to get some dinner but I'll be back later. | 01:03 |
| arosen | I'm arosen @ nicira .com btw | 01:03 |
| *** avishay has joined #openstack-dev | 01:05 | |
| *** Gordonz has quit IRC | 01:05 | |
| *** cloudvirt has joined #openstack-dev | 01:06 | |
| *** pixelbeat has quit IRC | 01:07 | |
| *** cloudvirt has quit IRC | 01:11 | |
| *** cloudvirt has joined #openstack-dev | 01:14 | |
| *** ncode has joined #openstack-dev | 01:15 | |
| *** edygarcia has quit IRC | 01:15 | |
| *** spiffxp has quit IRC | 01:15 | |
| *** samkottler has joined #openstack-dev | 01:19 | |
| *** ewindisch has quit IRC | 01:19 | |
| *** ewindisch has joined #openstack-dev | 01:20 | |
| *** ewindisch has quit IRC | 01:25 | |
| *** johnpostlethwait has quit IRC | 01:25 | |
| *** fsouza has quit IRC | 01:26 | |
| *** tgall_foo has joined #openstack-dev | 01:28 | |
| *** cloudvirt has quit IRC | 01:34 | |
| *** matiu has joined #openstack-dev | 01:37 | |
| *** jog0 has left #openstack-dev | 01:38 | |
| *** avishay has quit IRC | 01:38 | |
| *** fsouza has joined #openstack-dev | 01:40 | |
| *** cloudvirt has joined #openstack-dev | 01:42 | |
| *** PotHix has quit IRC | 01:42 | |
| *** mnewby has joined #openstack-dev | 01:49 | |
| *** Dr_Who has joined #openstack-dev | 01:52 | |
| *** Dr_Who has joined #openstack-dev | 01:52 | |
| *** adjohn has quit IRC | 01:57 | |
| *** zhuadl has joined #openstack-dev | 02:00 | |
| *** edygarcia has joined #openstack-dev | 02:04 | |
| *** cloudvirt has quit IRC | 02:10 | |
| *** cloudvirt has joined #openstack-dev | 02:13 | |
| *** cloudvirt has quit IRC | 02:15 | |
| *** winston-d has joined #openstack-dev | 02:20 | |
| *** tongli has joined #openstack-dev | 02:21 | |
| *** matwood has joined #openstack-dev | 02:21 | |
| *** tongli has quit IRC | 02:25 | |
| *** markmcclain has joined #openstack-dev | 02:25 | |
| *** anniec has quit IRC | 02:27 | |
| *** jiangyh has quit IRC | 02:30 | |
| *** dolphm has quit IRC | 02:36 | |
| *** danwent has quit IRC | 02:38 | |
| *** edygarcia has quit IRC | 02:39 | |
| *** renier has quit IRC | 02:46 | |
| *** renier has joined #openstack-dev | 02:46 | |
| *** troytoman-away is now known as troytoman | 02:49 | |
| *** colinmcnamara has joined #openstack-dev | 02:53 | |
| *** mjfork has quit IRC | 02:54 | |
| *** jdurgin has quit IRC | 02:55 | |
| *** littleidea_ has joined #openstack-dev | 03:03 | |
| *** Gordonz has joined #openstack-dev | 03:05 | |
| smoser | mikal, around ? | 03:06 |
| smoser | https://review.openstack.org/#/c/10934/ looks pretty good. thank you for your work there. | 03:07 |
| *** littleidea has quit IRC | 03:07 | |
| *** littleidea_ is now known as littleidea | 03:07 | |
| *** renier_ has joined #openstack-dev | 03:08 | |
| *** renier has quit IRC | 03:08 | |
| arosen | hi andrewsmedina were you able to reproduce it? | 03:09 |
| *** Gordonz has quit IRC | 03:10 | |
| mikal | smoser: yep, just eating lunch... | 03:13 |
| mikal | smoser: all I need to do now is deploy it and see if nova still works | 03:13 |
| smoser | i do think that your security concerns are relevant. | 03:14 |
| smoser | (commenting in bug now) | 03:14 |
| smoser | and i'd really like to use mtools in the "must use vfat" scenario | 03:14 |
| smoser | and i'd like to get the injected files path off of / | 03:15 |
| smoser | i know that would break things, but if not configured for config-drive-v1, then put them under '/files' or something. | 03:15 |
| mikal | Which bug BTW | 03:15 |
| mikal | ? | 03:15 |
| mikal | I think the mailing list decided yesterday that the vfat mount for config drive v2 is ok because its a fresh filesystem | 03:16 |
| mikal | The image from glance support is now gone | 03:16 |
| mikal | I'd like file injection to be completely replaced by config drive in grizzly | 03:16 |
| smoser | not bug | 03:17 |
| smoser | sorry | 03:17 |
| smoser | gerrit review | 03:17 |
| mikal | Ahh cool | 03:17 |
| *** mnewby has quit IRC | 03:24 | |
| smoser | i was confused by your comment in the code. | 03:25 |
| smoser | i thought you were saying it was not safe. | 03:25 |
| smoser | but you were saying it was. | 03:25 |
| smoser | anyway, just pushed 'review' | 03:25 |
| *** fsouza has quit IRC | 03:25 | |
| smoser | mikal, thank you for your patience with me. | 03:25 |
| *** edygarcia has joined #openstack-dev | 03:27 | |
| mikal | smoser: is no problem | 03:28 |
| *** edygarcia has quit IRC | 03:28 | |
| mikal | smoser: I'm sure I'll find a way to take revenge sometime... | 03:28 |
| smoser | i think we shoudl break config-drive-v1 by default | 03:29 |
| smoser | i really dont think anyone uses it. | 03:29 |
| smoser | and having to determine which files were "injected" that live at / from the guest is a pita. | 03:29 |
| mikal | Its broken now, in the sense that you can't fetch an image any more and the drive format is different | 03:29 |
| smoser | ie, did the user "inject" metadata.js ? /ec2/2009-04-04/user-data ? | 03:29 |
| mikal | So that being in a subdirectory of its own solves that problem? | 03:30 |
| smoser | well, even less people used the fetch-an-image than the drive format. | 03:30 |
| smoser | and the drive format is configurable | 03:30 |
| smoser | well, yes. | 03:30 |
| smoser | i think it solves the problem. | 03:30 |
| smoser | do you not? | 03:30 |
| mikal | Although the vfat drive image code is more horrible than the iso version | 03:30 |
| smoser | i know that "injected files" go in /injected | 03:30 |
| mikal | I think it does | 03:30 |
| mikal | But I'm not a distro person, so perhaps I'm confused | 03:30 |
| smoser | and i look under there. | 03:30 |
| smoser | permissions are unavailable and as a result injection this way requires some other bit of knowledge. | 03:31 |
| smoser | but... what can you do. | 03:31 |
| smoser | injection is busted in more ways than one :) | 03:31 |
| mikal | Hmmm, I hadn't thought of permissions | 03:31 |
| mikal | I think you have to guess that, right? | 03:31 |
| smoser | the /root/.ssh/authorized_keys | 03:31 |
| smoser | you just have to know that | 03:31 |
| mikal | How do you know the userid to inject the ssh key for example? | 03:31 |
| smoser | well, ubuntu images (right now) just "know" that the user is ubuntu | 03:32 |
| smoser | but that is configurable via user-data | 03:32 |
| mikal | Ahhh, ok | 03:32 |
| smoser | by quantal i'll have it able to create the user | 03:32 |
| mikal | The user data could be yet more json to tell you all that | 03:32 |
| smoser | right now it just expects that is done. | 03:32 |
| smoser | right. | 03:32 |
| smoser | but then if the user-data is json... it should just be cloud-init data, and you can solve your "injection" that way. | 03:32 |
| mikal | True | 03:33 |
| smoser | config drive injection is really not what people want when they think of injection | 03:33 |
| mikal | The key is known to the instance IIRC | 03:33 |
| smoser | its just more user-data | 03:33 |
| mikal | So there's no reason for it not to be in the json metadata | 03:33 |
| smoser | right. the public key the instance is laucned with will be in the metadata.js files | 03:33 |
| smoser | and can also be in cloud-config (for cloud-init) | 03:33 |
| smoser | all sorts of ways | 03:33 |
| smoser | you can make a part-handler for cloud-init that takes a mimetype=tar and tar -C / xvpf - | 03:34 |
| smoser | so, yeah, lets break injection | 03:34 |
| smoser | in config-drive-v1 terms | 03:34 |
| mikal | Let's try and murder it in grizzly | 03:34 |
| smoser | and see if anyone complains. | 03:34 |
| mikal | How does ec2 deal with these problems? | 03:34 |
| mikal | Do they do injection? | 03:34 |
| smoser | they have user-data | 03:34 |
| smoser | thats all you need. | 03:34 |
| smoser | they did things right | 03:34 |
| smoser | not this garbage. | 03:35 |
| smoser | you get 64k of stuff to pass to an instance. | 03:35 |
| smoser | the instance has to then pay attention to that and do what you told it | 03:35 |
| *** ncode has quit IRC | 03:35 | |
| mikal | Ahhh, interesting | 03:35 |
| mikal | So removing this doesn't make us ec2 incomptable | 03:35 |
| mikal | It kinda makes us more compatable | 03:35 |
| smoser | with config-drive-v2 and "injection" we're essentially making that number 64M | 03:35 |
| mikal | Only 64mb with vfat | 03:35 |
| smoser | but we're not putting /files in the metadata service | 03:35 |
| mikal | With iso the skys the limit... | 03:35 |
| smoser | true. | 03:35 |
| smoser | you should limit that. | 03:36 |
| mikal | We don't enforce any limit at the disk side | 03:36 |
| mikal | Perhaps the DB does | 03:36 |
| smoser | the api should imit it | 03:36 |
| smoser | otherwise i can dos you that way | 03:36 |
| * mikal goes to check the db model | 03:36 | |
| mikal | user-data is a text column | 03:36 |
| mikal | So you can probably DoS the DB as well... | 03:37 |
| mikal | Oh, I lie | 03:37 |
| mikal | Text is 64kb in mysql | 03:37 |
| *** zhuadl has quit IRC | 03:39 | |
| *** Mandell has joined #openstack-dev | 03:53 | |
| *** Mandell has quit IRC | 03:59 | |
| *** Dr_Who has quit IRC | 03:59 | |
| *** thingee is now known as thingee_zz | 04:10 | |
| *** danwent has joined #openstack-dev | 04:13 | |
| *** nati_ueno has quit IRC | 04:19 | |
| *** samkottler has quit IRC | 04:25 | |
| *** littleidea_ has joined #openstack-dev | 04:25 | |
| *** littleidea has quit IRC | 04:28 | |
| *** littleidea_ is now known as littleidea | 04:28 | |
| *** matwood has quit IRC | 04:33 | |
| *** garyk has quit IRC | 04:42 | |
| *** andrewsmedina has quit IRC | 04:46 | |
| *** zhuadl has joined #openstack-dev | 04:49 | |
| *** troytoman is now known as troytoman-away | 04:57 | |
| *** matwood has joined #openstack-dev | 05:02 | |
| *** Mandell has joined #openstack-dev | 05:05 | |
| *** matwood has quit IRC | 05:17 | |
| *** matwood has joined #openstack-dev | 05:17 | |
| *** samkottler has joined #openstack-dev | 05:19 | |
| *** samkottler is now known as samkottler|nap | 05:19 | |
| *** matiu has quit IRC | 05:23 | |
| *** rohitk has joined #openstack-dev | 05:25 | |
| *** mrunge has joined #openstack-dev | 05:33 | |
| *** amotoki has joined #openstack-dev | 05:34 | |
| *** garyk has joined #openstack-dev | 05:35 | |
| *** erikzaadi has joined #openstack-dev | 05:41 | |
| *** zaitcev has quit IRC | 05:42 | |
| *** adjohn has joined #openstack-dev | 05:43 | |
| *** Hitesh has joined #openstack-dev | 05:53 | |
| Hitesh | Hello | 05:54 |
| Hitesh | I got stucked while installing devstack script | 05:54 |
| Hitesh | any pointer on this -> lockfile.LockFailed: failed to create /opt/stack/horizon/openstack_dashboard/local/hitesh-openstack.MainThread-3623 | 05:54 |
| *** s34n has quit IRC | 05:55 | |
| *** s34n_ has joined #openstack-dev | 05:56 | |
| *** hattwick has quit IRC | 06:00 | |
| Hitesh | danwent: yOU THERE? | 06:04 |
| *** steveb_ has quit IRC | 06:05 | |
| *** Hitesh has quit IRC | 06:06 | |
| *** gargya has joined #openstack-dev | 06:09 | |
| *** colinmcnamara has quit IRC | 06:13 | |
| *** adjohn has quit IRC | 06:19 | |
| *** markmc has joined #openstack-dev | 06:21 | |
| *** littleidea has quit IRC | 06:22 | |
| *** danwent has quit IRC | 06:25 | |
| *** colinmcnamara has joined #openstack-dev | 06:26 | |
| *** vuntz has quit IRC | 06:41 | |
| *** rafaduran has joined #openstack-dev | 06:46 | |
| *** danwent has joined #openstack-dev | 06:47 | |
| *** colinmcnamara has quit IRC | 06:47 | |
| *** alex88 has joined #openstack-dev | 07:03 | |
| *** vanchester has joined #openstack-dev | 07:04 | |
| *** markmcclain has quit IRC | 07:06 | |
| *** gargya has quit IRC | 07:07 | |
| *** reidrac has joined #openstack-dev | 07:07 | |
| *** EmilienM has joined #openstack-dev | 07:09 | |
| *** matwood has quit IRC | 07:14 | |
| *** avishay has joined #openstack-dev | 07:35 | |
| *** sunxin has quit IRC | 07:37 | |
| *** sulochan has joined #openstack-dev | 07:59 | |
| *** darraghb has joined #openstack-dev | 08:02 | |
| *** derekh has joined #openstack-dev | 08:03 | |
| *** pixelbeat has joined #openstack-dev | 08:12 | |
| *** Ryan_Lane has quit IRC | 08:18 | |
| *** steveb_ has joined #openstack-dev | 08:19 | |
| *** avishay has quit IRC | 08:41 | |
| *** blamar has quit IRC | 08:42 | |
| *** hattwick has joined #openstack-dev | 08:43 | |
| *** eglynn has joined #openstack-dev | 08:51 | |
| *** avishay has joined #openstack-dev | 08:54 | |
| *** winston-d has quit IRC | 08:55 | |
| *** danpb has joined #openstack-dev | 08:56 | |
| *** danpb has quit IRC | 08:57 | |
| *** danpb has joined #openstack-dev | 08:57 | |
| *** steveb_ has quit IRC | 09:02 | |
| *** s34n_ has quit IRC | 09:04 | |
| *** s34n has joined #openstack-dev | 09:05 | |
| *** Slower has quit IRC | 09:09 | |
| *** amotoki has left #openstack-dev | 09:09 | |
| *** Slower has joined #openstack-dev | 09:09 | |
| *** amotoki has joined #openstack-dev | 09:09 | |
| *** alexpilotti has quit IRC | 09:13 | |
| *** danwent has quit IRC | 09:17 | |
| *** Slower has quit IRC | 09:32 | |
| *** Slower has joined #openstack-dev | 09:32 | |
| *** alexpilotti has joined #openstack-dev | 09:46 | |
| *** zhuadl has quit IRC | 09:55 | |
| *** asalkeld has quit IRC | 09:58 | |
| *** kyriakos has joined #openstack-dev | 10:17 | |
| *** vuntz has joined #openstack-dev | 10:29 | |
| *** asalkeld has joined #openstack-dev | 10:53 | |
| *** sandywalsh has quit IRC | 11:30 | |
| *** linmin_ has quit IRC | 11:39 | |
| *** adalbas has joined #openstack-dev | 11:47 | |
| *** rkukura has quit IRC | 11:52 | |
| *** Dr_Who has joined #openstack-dev | 11:57 | |
| *** Dr_Who has joined #openstack-dev | 11:57 | |
| *** maurosr has joined #openstack-dev | 12:02 | |
| avishay | mikal: ping | 12:05 |
| *** markvoelker has joined #openstack-dev | 12:06 | |
| *** matwood has joined #openstack-dev | 12:07 | |
| *** tserong has quit IRC | 12:13 | |
| *** dhellmann_ has joined #openstack-dev | 12:16 | |
| *** tserong has joined #openstack-dev | 12:18 | |
| *** mnewby has joined #openstack-dev | 12:20 | |
| *** dhellmann has quit IRC | 12:20 | |
| *** dhellmann_ is now known as dhellmann | 12:20 | |
| *** dhellmann has quit IRC | 12:21 | |
| *** samkottler|nap has quit IRC | 12:22 | |
| *** fsouza has joined #openstack-dev | 12:22 | |
| *** samkottler|nap has joined #openstack-dev | 12:25 | |
| *** samkottler|nap is now known as samkottler | 12:25 | |
| *** zaneb has joined #openstack-dev | 12:27 | |
| *** mrunge has quit IRC | 12:30 | |
| *** samkottl1r has joined #openstack-dev | 12:33 | |
| *** samkottler has quit IRC | 12:35 | |
| *** samkottl1r is now known as samkottler | 12:35 | |
| *** dolphm_ has joined #openstack-dev | 12:36 | |
| *** wiliam has joined #openstack-dev | 12:39 | |
| *** dprince has joined #openstack-dev | 12:39 | |
| *** fsouza has quit IRC | 12:40 | |
| *** rods has joined #openstack-dev | 12:47 | |
| *** roge has joined #openstack-dev | 12:55 | |
| *** dprince has quit IRC | 13:01 | |
| *** markmcclain has joined #openstack-dev | 13:07 | |
| *** markmcclain has quit IRC | 13:08 | |
| *** dprince has joined #openstack-dev | 13:08 | |
| *** avishay has quit IRC | 13:09 | |
| *** mcolombo has quit IRC | 13:10 | |
| *** dolphm_ has quit IRC | 13:12 | |
| *** maurosr has quit IRC | 13:14 | |
| *** maurosr has joined #openstack-dev | 13:14 | |
| *** dolphm has joined #openstack-dev | 13:15 | |
| *** dolphm has quit IRC | 13:15 | |
| *** oneiroi|gone is now known as oneiroi | 13:16 | |
| *** andrewsmedina has joined #openstack-dev | 13:18 | |
| *** fsouza has joined #openstack-dev | 13:20 | |
| *** rohitk_1 has joined #openstack-dev | 13:25 | |
| *** rpedde_away is now known as rpedde | 13:25 | |
| *** rohitk has quit IRC | 13:26 | |
| *** cloudvirt has joined #openstack-dev | 13:30 | |
| *** ncode has joined #openstack-dev | 13:34 | |
| *** andrewsmedina has left #openstack-dev | 13:34 | |
| *** lts has joined #openstack-dev | 13:36 | |
| *** ayoung has joined #openstack-dev | 13:36 | |
| *** fsouza has quit IRC | 13:37 | |
| *** fsouza has joined #openstack-dev | 13:38 | |
| *** markmcclain has joined #openstack-dev | 13:43 | |
| *** flaviamissi has joined #openstack-dev | 13:47 | |
| *** halfss has joined #openstack-dev | 13:49 | |
| *** samkottl1r has joined #openstack-dev | 13:50 | |
| *** samkottler is now known as samkottler_ | 13:50 | |
| *** samkottl1r is now known as samkottler | 13:50 | |
| ameade | vishy: Hey so i'm a little unclear on something, should making an unversioned request to a bookmark link work? | 13:50 |
| *** samkottler has quit IRC | 13:51 | |
| *** samkottler has joined #openstack-dev | 13:51 | |
| *** samkottler_ has quit IRC | 13:51 | |
| *** dolphm_ has joined #openstack-dev | 13:51 | |
| *** samkottler has quit IRC | 13:51 | |
| *** dubsquared has joined #openstack-dev | 13:51 | |
| *** samkottler has joined #openstack-dev | 13:52 | |
| *** milner_ has quit IRC | 13:53 | |
| *** edygarcia has joined #openstack-dev | 13:53 | |
| *** markmcclain has quit IRC | 13:54 | |
| *** mtreinish has joined #openstack-dev | 13:57 | |
| sdague | vishy: so, can you re rebase your patches to get them back under your name? | 13:59 |
| *** tongli has joined #openstack-dev | 14:03 | |
| *** markmcclain has joined #openstack-dev | 14:04 | |
| *** milner_ has joined #openstack-dev | 14:07 | |
| *** rkukura has joined #openstack-dev | 14:07 | |
| *** sandywalsh has joined #openstack-dev | 14:08 | |
| *** openstackgerrit has quit IRC | 14:08 | |
| *** openstackgerrit has joined #openstack-dev | 14:08 | |
| *** halfss has quit IRC | 14:12 | |
| *** dhellmann has joined #openstack-dev | 14:12 | |
| *** EmilienM has left #openstack-dev | 14:12 | |
| markmc | sdague, do you see your patch set 5 in the UI? | 14:14 |
| markmc | sdague, in this one https://review.openstack.org/10814 ? | 14:14 |
| markmc | sdague, ah, no - you don't, since you commented on patch set 4 | 14:15 |
| markmc | sdague, btw, you wanted 'git review -R' | 14:15 |
| sdague | I see a patch 4 draft | 14:15 |
| * markmc wished that was the default :) | 14:15 | |
| sdague | markmc: yep, got that now :) | 14:15 |
| markmc | sdague, patch 4 draft? wow | 14:15 |
| markmc | sdague, could you screenshot? | 14:15 |
| sdague | sure | 14:15 |
| sdague | markmc: http://img6.imagebanana.com/img/7nblrznu/screenshot_236.png | 14:17 |
| markmc | sdague, oh, I see that too :) | 14:17 |
| markmc | sdague, funky | 14:18 |
| *** johnpur has joined #openstack-dev | 14:18 | |
| *** ChanServ sets mode: +v johnpur | 14:18 | |
| sdague | so, here is the sequence that got us here, maybe we can figure out a way to back it out before vishy gets in. | 14:18 |
| markmc | jeblair, seen https://bugs.launchpad.net/openstack-ci/+bug/1034761 ? | 14:18 |
| uvirtbot | Launchpad bug 1034761 in openstack-ci "gerrit: missing patchset bizarroness" [Undecided,New] | 14:18 |
| sdague | I originally did a -D upload because I assumed that it would be kind of safe | 14:19 |
| *** samkottl1r has joined #openstack-dev | 14:19 | |
| sdague | then did a normal git review, I didn't get the -R piece of advice until this was part way through | 14:19 |
| *** samkottl1r has quit IRC | 14:19 | |
| sdague | so is there a way that you think I could get rid of my rebases? or will vish need to just push new rebases? | 14:20 |
| dansmith | sdague: did your reviews actually affect vish's? | 14:20 |
| dansmith | or just make your review look funky? | 14:20 |
| sdague | dansmith: yes, it has seemed to | 14:20 |
| dansmith | hmm, that seems majorly problematic, if one user can hose another like that.. | 14:20 |
| *** samkottl1r has joined #openstack-dev | 14:20 | |
| *** dubsquared has quit IRC | 14:21 | |
| markmc | sdague, I think the issue here is that only nova-core can upload revisions to reviews started by others | 14:22 |
| markmc | sdague, my guess is that it allowed you do it, but then got confused | 14:22 |
| *** dubsquared has joined #openstack-dev | 14:22 | |
| markmc | sdague, I guess I could push a new set of commits and that might work, but may as well just let vish do that | 14:22 |
| sdague | markmc: ok, for future, is there something I can through in .gitconfig to make -R be the default git review behavior | 14:22 |
| *** blamar has joined #openstack-dev | 14:23 | |
| *** samkottler has quit IRC | 14:23 | |
| *** blamar has quit IRC | 14:23 | |
| markmc | sdague, hmm, look at that - I think there is | 14:23 |
| * markmc tries to figure it out | 14:23 | |
| sdague | yeh, sorry for screwing up the system here, wasn't trying to make everyone's life difficult | 14:23 |
| *** samkottl1r is now known as samkottler | 14:24 | |
| *** samkottler has joined #openstack-dev | 14:24 | |
| markmc | sdague, think we can add defaultrebase=false to .gitreview | 14:24 |
| *** ewindisch has joined #openstack-dev | 14:24 | |
| sdague | ok, so that would go in the repos, that would be great | 14:24 |
| markmc | sdague, yes, that would be superb | 14:25 |
| markmc | sdague, the autorebasing thing really messes up a nice gerrit feature too | 14:25 |
| sdague | markmc: I'm happy to run around and do that | 14:25 |
| markmc | sdague, you can compare different patch sets within a review | 14:25 |
| markmc | sdague, but if there's a rebase, you get all those changes mixed into the diff | 14:25 |
| markmc | sdague, now, what would be nice if git-review could check for merge conflicts before uploading | 14:26 |
| markmc | sdague, i.e. "you change conflicts with master, please rebase before uploading" | 14:26 |
| sdague | I thought I've hit a currency issue before | 14:26 |
| sdague | where it did complain about that | 14:27 |
| sdague | though, in reality, given how long lived a lot of the reviews are, it seems like it wouldn't cover a lot of the conflicts anyway | 14:27 |
| markmc | sdague, yeah, I guess "check for merge conflicts" is the only argument I can see for auto-rebasing | 14:28 |
| vishy | I can pull my patches back sure | 14:29 |
| sdague | markmc: ok, so if you like, I can register the issue, and get those .gitreview patches up for review on all the projects, to hopefully avoid this in the future | 14:29 |
| vishy | but i'm still listed as the author so i don't know if it matters | 14:29 |
| sdague | vishy: it looks like it caused a consistency issue | 14:29 |
| *** zaneb has quit IRC | 14:29 | |
| markmc | vishy, we can't vote on those patches | 14:29 |
| sdague | per comments from markmc further up | 14:29 |
| markmc | vishy, i.e. in https://review.openstack.org/10814 there is only patch set 4 and we can't vote on it | 14:30 |
| markmc | vishy, there's a hidden patch set 5 | 14:30 |
| vishy | markmc, sdague: weird | 14:30 |
| sdague | agreed :) | 14:30 |
| vishy | ok I will redo mine | 14:30 |
| markmc | vishy, right, see https://bugs.launchpad.net/openstack-ci/+bug/1034761 | 14:30 |
| uvirtbot | Launchpad bug 1034761 in openstack-ci "gerrit: missing patchset bizarroness" [Undecided,New] | 14:30 |
| vishy | sdague: you will have to cherry pick your patch over mine again | 14:30 |
| sdague | yep, that's fine | 14:31 |
| *** Ruetobas has quit IRC | 14:31 | |
| dansmith | hmm, I don't even see patch set 4 on that review, only 1-3 | 14:31 |
| *** zaneb has joined #openstack-dev | 14:31 | |
| vishy | sdague, markmc: done | 14:32 |
| vanchester | hello. In keystone is there any tests for contrib modules? | 14:32 |
| *** littleidea has joined #openstack-dev | 14:33 | |
| markmc | vishy, cool | 14:33 |
| markmc | vishy, but, gah - my inline comments got lost somehow | 14:33 |
| vishy | markmc: fyi no I don't think that works | 14:33 |
| markmc | vishy, the controller sharing thing? | 14:34 |
| ttx | dprince: saw my message yesterday about the cinder deprecation support draft review ? | 14:34 |
| vishy | markmc: the problem is that the extensions are stored in the resource | 14:34 |
| *** e1mer has quit IRC | 14:34 | |
| vishy | markmc: oh, you're saying that you we keep the other parts of the inherits | 14:34 |
| vishy | markmc: I suppose that is fine, but it doesn't allow for actually extending the controller if that is the desire | 14:35 |
| dprince | ttx: I did. Its in the queue. :) | 14:35 |
| ttx | dprince: ok :) | 14:35 |
| markmc | vishy, yes, still inherit the resource but share the controller | 14:35 |
| markmc | vishy, annoyingly, the resource is a controller from routers.Mapper POV :) | 14:35 |
| *** matiu has joined #openstack-dev | 14:35 | |
| * ttx is like TCP, gets worried when he doesn't receive ACKs | 14:36 | |
| vishy | markmc: as in if you wanted to add an extra method or some such to the servers controller. I removed the places where we do that right now, but I expect someone will want to do it again. | 14:36 |
| *** Ruetobas has joined #openstack-dev | 14:36 | |
| markmc | vishy, ok, we'd need a "if resource.controller is None: resource.controller = inherits.controller" | 14:36 |
| markmc | ttx, did you get those UDP jokes, btw? | 14:37 |
| *** Mandell has quit IRC | 14:37 | |
| vishy | markmc: makes sense. I will change it and pull in the other change as well | 14:37 |
| markmc | vishy, cool | 14:37 |
| vishy | markmc: I was saving the resources because i needed them in an old version, but i worked around it with inherits | 14:38 |
| markmc | vishy, yeah, I figured | 14:38 |
| ttx | markmc: heh | 14:38 |
| * dprince apparently is more like UDP (sometimes just gets lost) | 14:38 | |
| markmc | vishy, my inline comment that got lost was a "it'd be nice if we could clean this up" thing | 14:39 |
| markmc | vishy, the get_methods() method returning a method and extensions | 14:39 |
| markmc | vishy, and you ignore the method from inherits | 14:39 |
| markmc | vishy, would be nice to have separate get_methods() and get_extensions() | 14:39 |
| vishy | markmc: yes, although i don't know what the hell to do if the method is different | 14:40 |
| vishy | markmc: as in, which one do you call? both of them? | 14:40 |
| markmc | vishy, you'd only call get_extension() on inherits | 14:40 |
| markmc | extensions | 14:40 |
| *** littleidea has quit IRC | 14:41 | |
| markmc | vishy, would the logic be simply look in self.wsgi_extensions first and then fall back to self.wsgi_action_extensions ? | 14:42 |
| markmc | vishy, it's not clear from the current code | 14:42 |
| markmc | vishy, anyway, no big deal if you want to ignore that | 14:42 |
| *** littleidea has joined #openstack-dev | 14:43 | |
| *** markmcclain1 has joined #openstack-dev | 14:44 | |
| *** markmcclain has quit IRC | 14:46 | |
| jeblair | 14:22 < markmc> sdague, I think the issue here is that only nova-core can upload revisions to reviews started by others | 14:47 |
| markmc | that's what I said! | 14:48 |
| markmc | oh wait, it is what I said | 14:48 |
| jeblair | i know, i'm quoting you and writing a response | 14:48 |
| markmc | :) | 14:48 |
| jeblair | i just got up, i'm trying to establish context. :) | 14:48 |
| jeblair | markmc, sdague: ^ actually, anyone can upload a new patchset to an existing change (or if not, it's a bug) | 14:49 |
| markmc | I can keep asking dumb questions and interrupting you writing your response, if that helps | 14:49 |
| markmc | bah, too late | 14:49 |
| jeblair | whew | 14:49 |
| * jeblair warms up fingers | 14:49 | |
| markmc | jeblair, ah, ok | 14:49 |
| markmc | it used to be nova-core only, right? | 14:49 |
| * markmc shrugs | 14:49 | |
| *** dave_mcn has joined #openstack-dev | 14:50 | |
| jeblair | markmc: ah, it may have been the case that only -core could "forge author identity" which might have had the same effect i think. but that should no longer be the case. | 14:51 |
| *** mjfork has joined #openstack-dev | 14:51 | |
| markmc | jeblair, that's the thing, yeah | 14:51 |
| dave_mcn | Hey all, wondering if anyone is available to review a change (https://review.openstack.org/#/c/11029/)? | 14:51 |
| * jeblair goes back to trying to understand the rest of that conversation. | 14:52 | |
| markmc | jeblair, the bug in launchpad should cover it | 14:52 |
| *** datsun180b has joined #openstack-dev | 14:54 | |
| *** erikzaadi has quit IRC | 14:56 | |
| jeblair | markmc: my understanding is 10815,7 was based on 10814,4 -- and both of those were drafts | 14:58 |
| jeblair | markmc: is that incorrect? | 14:58 |
| markmc | jeblair, hmm, no | 15:01 |
| markmc | jeblair, sdague's 10814,4 wasn't visible to me earlier | 15:01 |
| markmc | jeblair, could only see it with 'gerrit query' | 15:01 |
| jeblair | markmc: it's still not visible to me, i made some assumptions there. but the bug didn't mention 10815,7 which is also invisible. | 15:02 |
| vishy | markmc: ok made the changes | 15:02 |
| vishy | markmc: might want to wait for jenkins though. I didn't run all of the tests :) | 15:02 |
| *** jaypipes has joined #openstack-dev | 15:02 | |
| markmc | jeblair, yes, that was part of the series of rebased commits sdague uploaded | 15:03 |
| markmc | jeblair, I do see them in UI now | 15:03 |
| jeblair | markmc: really, i don't see a patchs set 7 for 10815 | 15:03 |
| markmc | jeblair, it shows up like in sdague's screenshot with DRAFT | 15:03 |
| jeblair | oh, you're on the review. | 15:04 |
| vishy | jeblair: do you have any idea why the first patch in a series doesn't get the helpful "Dependencies/ Needed By" ? | 15:04 |
| markmc | jeblair, that could be, I wasn't originally | 15:04 |
| markmc | jeblair, right, and it's a draft | 15:04 |
| markmc | jeblair, ok, all makes sense now | 15:04 |
| markmc | jeblair, sdague just did 'git review -D' | 15:04 |
| *** derekh has quit IRC | 15:04 | |
| *** reidrac has quit IRC | 15:04 | |
| jeblair | markmc: heh, i added myself to the reviews so now i can see the draft. :) | 15:04 |
| markmc | vishy, it only shows non-merged patches | 15:05 |
| markmc | vishy, it should get the Needed By thing | 15:05 |
| vishy | markmc: but it doesn't | 15:05 |
| markmc | vishy, load https://review.openstack.org/#/c/10814/ and hit 'd' | 15:05 |
| markmc | vishy, that expands the section | 15:05 |
| vishy | markmc: the first patch never does, it only seems to show Needed By if it allso has a depends on | 15:05 |
| vishy | markmc: ah, just not expanded by default | 15:06 |
| vishy | markmc: odd that it is when there is a depends on | 15:06 |
| markmc | vishy, I guess it's only expanded by default if there's something in Depends On | 15:06 |
| jeblair | vishy, markmc: oddly enough -- gerrit does actually show the "depends on" for every patch, even when it's merged, in the ssh query output | 15:07 |
| *** matiu has quit IRC | 15:07 | |
| *** ecarlin has joined #openstack-dev | 15:07 | |
| *** matiu has joined #openstack-dev | 15:07 | |
| *** matiu has quit IRC | 15:07 | |
| *** matiu has joined #openstack-dev | 15:07 | |
| jeblair | markmc: so now that i can see all the patchsets, it does look like 10815,7 was based on 10814,4 and both were drafts | 15:08 |
| vishy | markmc, jeblair: are the 'os- and inherits misaligned for you on the right side on line 592 here? https://review.openstack.org/#/c/10816/9/nova/api/openstack/compute/contrib/volumes.py | 15:08 |
| markmc | jeblair, I've closed the bug as Invalid already btw :) | 15:08 |
| markmc | vishy, looks fine here | 15:09 |
| jeblair | markmc: ok. i think the underlying bug here is that you really shouldn't be able to add a draft patchset to a non-draft review. that _always_ makes things weird. | 15:09 |
| vishy | weird, must be a weird rendering issue on my machine | 15:09 |
| markmc | jeblair, autorebase die in a fire! | 15:09 |
| markmc | jeblair, :) | 15:09 |
| sdague | jeblair: right, so the problem was I did a draft review, assuming that would cause less damage if something went wrong. And the opposite was true :) | 15:11 |
| jeblair | markmc: so what did autorebase screw up here? | 15:11 |
| vishy | weird, long runs of whitespace look totally borked on my machine | 15:12 |
| markmc | jeblair, it meant sdague unintentionally rebased vishy's patches | 15:12 |
| jeblair | vishy: i see that they are not horizontally aligned in both the old/new columns. there is an empty line in the new column (that isn't really there) due to the removal of a line of code | 15:12 |
| vishy | stupid chrome canary :) | 15:12 |
| sdague | heh, well it's called canary for a reason | 15:13 |
| vishy | looks fine in regular chrome | 15:13 |
| vishy | regular chrome doesn't support the retina display yet :( | 15:13 |
| markmc | mac weanies | 15:13 |
| * markmc sighs | 15:13 | |
| vishy | oh sweet, it has been updated | 15:14 |
| * vishy switches back | 15:14 | |
| jeblair | markmc: 10815 was still based on 10814 -- was it a problem that 10814's parent changed? | 15:14 |
| jeblair | markmc: (I wouldn't mind if auto-rebase were turned off, just making sure i understand what happened here) | 15:15 |
| sdague | jeblair: so what I was trying to do was add another review dependent on 10816, because I needed infrastructure that vishy wrote to do additional work | 15:16 |
| markmc | jeblair, sdague just wanted to upload 11051 | 15:16 |
| jeblair | oh! | 15:16 |
| sdague | so I pulled his work into a seperate branch, and was trying to tag just a review on top of it | 15:16 |
| jeblair | so the 814 and 815 rebased draft uploads were accidental, caused by auto-rebase! | 15:17 |
| markmc | jeblair, and didn't intend on rebasing 10814,10815 and 10816 | 15:17 |
| sdague | yes, exactly | 15:17 |
| markmc | jeblair, right :) | 15:17 |
| jeblair | markmc: evil indeed. :( | 15:17 |
| markmc | jeblair, die in a fire! | 15:17 |
| markmc | not you, autorebase | 15:17 |
| markmc | you understand | 15:17 |
| vishy | time to change git review to only rebase if it has to | 15:17 |
| jeblair | mtaylor: ^ | 15:18 |
| sdague | right, so markmc found that we could default it off in the configs | 15:18 |
| * jeblair throws mtaylor into the fire | 15:18 | |
| *** cloudvirt has quit IRC | 15:18 | |
| markmc | vishy, yeah, we can add defaultrebase=false in .gitreview - sdague is on it | 15:18 |
| sdague | though, the issue that seems to happen there, is there doesn't seem to be a flag on git review to fork it to rebase | 15:18 |
| sdague | there is only -R to force it to not | 15:18 |
| *** hemna has joined #openstack-dev | 15:19 | |
| markmc | sdague, 'git rebase' works just fine without git-review's help :) | 15:19 |
| sdague | is that going to get us wedged in a different way? | 15:19 |
| sdague | yep, true | 15:19 |
| markmc | you always want to test after rebase | 15:19 |
| sdague | agreed | 15:19 |
| markmc | never makes sense to rebase and upload as one atomic thing | 15:19 |
| sdague | agreed. | 15:20 |
| dansmith | nah, my code never needs to be retested after a rebase | 15:20 |
| sdague | dansmith: is that part of your chuck norris placard? | 15:21 |
| markmc | obvious troll is obvious | 15:21 |
| * markmc throws dansmith in after mtaylor | 15:21 | |
| sdague | "They say his code always rebases cleaning, and never needs retesting" | 15:21 |
| sdague | "They say his code always rebases cleanly, and never needs retesting" | 15:21 |
| dansmith | heh | 15:22 |
| mtaylor | vishy, jeblair: I agree | 15:23 |
| *** sacharya has joined #openstack-dev | 15:23 | |
| mtaylor | markmc, vishy: I believe we have hit the point where the default rebase is causing more pain than it is alleviating | 15:24 |
| markmc | mtaylor, a "your commit conflicts with master, you should rebase" warning check thing would be nice though | 15:24 |
| vishy | mtaylor: i think a huge help would be if rebase succeeds, git reset --hard ORIG_HEAD (or equivalent) | 15:24 |
| markmc | mtaylor, i.e. do a throwaway test merge before uploading | 15:24 |
| markmc | vishy, or that :) | 15:25 |
| vishy | mtaylor: or what markmc said | 15:25 |
| *** dprince_ has joined #openstack-dev | 15:25 | |
| mtaylor | hrm. now I have to choose between vishy and markmc | 15:25 |
| vishy | but not rebasing if the rebase is clean seems ++ | 15:25 |
| markmc | well, gerrit does a merge in the end | 15:25 |
| *** dprince has quit IRC | 15:26 | |
| markmc | so it's merge conflicts you want to avoid, not rebase conflicts :) | 15:26 |
| mtaylor | markmc: we've got some projects testing the cherry-pick state | 15:26 |
| * markmc wins! | 15:26 | |
| markmc | mtaylor, actually, a throwaway rebase tests all patches in the series for conflicts | 15:26 |
| mtaylor | markmc: openstack-ci-puppet has been cherry-picking instead of merging for about a week now | 15:26 |
| markmc | mtaylor, it's conceivable that the patch series doesn't conflict, but one of the patches does | 15:27 |
| mtaylor | markmc: and yes, a throwaway rebase might honestly get the point that we want to help with | 15:27 |
| markmc | mtaylor, and gerrit is merging them individually, so ... | 15:27 |
| mtaylor | ++ | 15:27 |
| mtaylor | markmc: btw - the cherry pick strategy is working well, but it does mean that _every_ patch gets cherry-pick applied, so if you're working on dependent patch sets, do you kind of need to understand rebasing better | 15:28 |
| *** dave_mcn has quit IRC | 15:28 | |
| mtaylor | markmc: but, check our recent commits: https://github.com/openstack/openstack-ci-puppet/commits/master | 15:29 |
| *** cp16net|away is now known as cp16net | 15:29 | |
| markmc | mtaylor, jeblair switched cinder over to cherry-picking mode briefly when I had a big queue of patches | 15:29 |
| markmc | mtaylor, something weird happened where patches go applied out of order | 15:30 |
| markmc | jeblair, it was cherry-picking mode you switched to, right ? | 15:30 |
| mtaylor | markmc: yeah, we've solved that part now | 15:30 |
| markmc | mtaylor, ok | 15:30 |
| mtaylor | markmc: zuul understands patch dependency order now | 15:30 |
| markmc | mtaylor, cherry-picking mode mimics the old "apply patches from the mailing list" behaviour | 15:30 |
| mtaylor | markmc: yeah | 15:30 |
| markmc | mtaylor, which I'm fine with, but some folks would think it's evil incarnated | 15:31 |
| mtaylor | markmc: yes. it is a different feel | 15:31 |
| danpb | nonsense, rebasing is evil incarnated :-) | 15:31 |
| markmc | mtaylor, you lose the history of exactly what the submitter tested with | 15:31 |
| markmc | danpb, that's what cherry-pick mode is, basically | 15:31 |
| danpb | markmc s/rebasing/merging/ | 15:31 |
| markmc | mtaylor, i.e. a rebase/merge can conceivably break your patch | 15:31 |
| *** Gordonz has joined #openstack-dev | 15:31 | |
| mtaylor | markmc: that's why we do that work and then test it | 15:32 |
| markmc | mtaylor, but if your patch was merged, not rebased you can point to the history and say "look, my patch *did* work!" | 15:32 |
| mtaylor | rather than testing the patch in the form thatit's sumbmitted | 15:32 |
| markmc | mtaylor, but danpb's response to that one is "I don't care" :) | 15:32 |
| jeblair | yeah, the gate tests always merge the change as gerrit is going to merge it before testing | 15:32 |
| mtaylor | markmc: we actually don't lose that history inside of gerrit | 15:32 |
| danpb | markmc i just don't think that's a particularl useful ability | 15:32 |
| mtaylor | markmc: you can still see what the parent of the patchset that you submitted was | 15:33 |
| markmc | mtaylor, yeah, I'm talking about e.g. doing git-bisect on some regression you find later | 15:33 |
| danpb | as compared to the horrible thinkgs merging does to the history | 15:33 |
| markmc | mtaylor, you land on my patch, blame me ... but it was the rebase/merge's fault, not mine | 15:33 |
| markmc | mtaylor, but TBH, I'd prefer the cleaner history :) | 15:33 |
| jeblair | markmc: so there is at least one test of the code _excatly_ as its committed, before its committed. | 15:33 |
| mtaylor | markmc: indeed. although honestly I don't care about blame | 15:33 |
| mtaylor | markmc: ++ | 15:33 |
| markmc | mtaylor, you only see it in gerrit, not git | 15:33 |
| markmc | mtaylor, blame is why we have Author: tags :) | 15:34 |
| mtaylor | markmc: the bisect will still find the code change where the failure is, so you can more easily find what broke | 15:34 |
| mtaylor | markmc: :) | 15:34 |
| danpb | markmc whether its the merges fault, or the original patch authors fault is not important IMHO | 15:34 |
| mtaylor | danpb: ++ | 15:34 |
| danpb | markmc merely identifying the point at which the fault occurred is the important thing | 15:34 |
| mtaylor | the important thing is finding the error and being able to fix it | 15:34 |
| markmc | danpb, mtaylor, it actually is useful information | 15:35 |
| danpb | whether your patch originally worked, prior to it being applied is not important IMHO, only that once applied it is broken | 15:35 |
| markmc | danpb, mtaylor, you know that it's a conflict between two independent changes | 15:35 |
| danpb | i don't think that tells you anything useful | 15:35 |
| markmc | danpb, mtaylor, think about looking at the result of git-bisect | 15:36 |
| danpb | that you don't already find out otherwise | 15:36 |
| markmc | danpb, mtaylor, if it's a cherry-picked patch, you try and figure out why the patch was broken | 15:36 |
| markmc | danpb, mtaylor, if it's a merge, you know it's a conflict between two different patches | 15:36 |
| markmc | otherwise the bisect would have landed on the patch itself, not the merge | 15:37 |
| danpb | i've never really found a case where that distinct is relevant or useful | 15:37 |
| markmc | but again, I'd say "a cleaner history is more useful" | 15:37 |
| danpb | maybe it could be, but I've never seen it | 15:37 |
| markmc | that's because you're not a kernel developer | 15:37 |
| markmc | and they're special | 15:37 |
| mtaylor | yeah - I can see what markmc is talking about in theory, but also agree, the clean history is more useful | 15:37 |
| jeblair | markmc: when you brought it up on the mailing list, there seems to have been some consensus that people liked the clean history of cherry-pick, and i don't recall any objections. | 15:37 |
| markmc | jeblair, yeah, don't mind me | 15:38 |
| * jeblair considers markmc the leading champion of cherry-pick | 15:38 | |
| markmc | I'm mostly just warning you folks about how kernel developers are going to tell us we suck for doing this | 15:38 |
| * danpb doesn't much care for the kernel developers opinions on non-kernel projects | 15:39 | |
| markmc | danpb, you know kernel developers are infallible, right? | 15:40 |
| jeblair | turns out there's more than one way to run a project. and git even supports being used in more than one way. :) | 15:40 |
| markmc | (that pause in conversation was brought to you by my laptop swapping like mad while I copy a disk image) | 15:40 |
| danpb | markmc if they're infallible maybe one of them can be Pope ;-P | 15:40 |
| vishy | mtaylor, jeblair, markmc: https://review.openstack.org/11081 | 15:41 |
| vishy | straw man :) | 15:42 |
| * markmc imagines us all lining up to kiss linus's hand | 15:42 | |
| *** maploin has joined #openstack-dev | 15:42 | |
| *** maploin has quit IRC | 15:42 | |
| *** maploin has joined #openstack-dev | 15:42 | |
| dansmith | well, probably better than RMS'.. I don't think he showers much.. | 15:42 |
| mtaylor | vishy: I likey | 15:46 |
| mtaylor | dansmith: I've touched RMS' hand before. I do not want to kiss it | 15:46 |
| dansmith | heh | 15:47 |
| mtaylor | vishy: if the rebase changes fails, should we rebase --abort for people? or leave it in rebase conflict state? | 15:48 |
| mtaylor | (in this new model) | 15:48 |
| *** spiffxp has joined #openstack-dev | 15:48 | |
| vishy | mtaylor: hmm, maybe abort with a message to rebase manually might be nicer, but it kind of makes --no-rebase useless | 15:51 |
| *** primeministerp has quit IRC | 15:54 | |
| *** primeministerp has joined #openstack-dev | 15:54 | |
| *** alex88 has quit IRC | 15:55 | |
| *** sdake has quit IRC | 15:57 | |
| *** maploin has quit IRC | 15:58 | |
| *** rafaduran has quit IRC | 15:59 | |
| *** sdake has joined #openstack-dev | 15:59 | |
| *** Rajesh has joined #openstack-dev | 16:02 | |
| *** Rajesh is now known as Guest38070 | 16:02 | |
| *** ayoung is now known as ayoung-afk | 16:02 | |
| *** cloudvirt has joined #openstack-dev | 16:08 | |
| *** ewindisch has quit IRC | 16:09 | |
| *** ecarlin has quit IRC | 16:09 | |
| *** lloydde has joined #openstack-dev | 16:10 | |
| *** Mandell has joined #openstack-dev | 16:14 | |
| *** rnirmal has joined #openstack-dev | 16:16 | |
| *** Gordonz has quit IRC | 16:17 | |
| *** ewindisch has joined #openstack-dev | 16:20 | |
| *** primeministerp has joined #openstack-dev | 16:23 | |
| *** danwent has joined #openstack-dev | 16:24 | |
| *** garyk has quit IRC | 16:24 | |
| *** danwent has quit IRC | 16:24 | |
| *** primeministerp has quit IRC | 16:27 | |
| *** primeministerp has joined #openstack-dev | 16:28 | |
| *** dubsquared1 has joined #openstack-dev | 16:30 | |
| *** dubsquared has quit IRC | 16:30 | |
| *** EmilienM has joined #openstack-dev | 16:32 | |
| *** Shrews has quit IRC | 16:35 | |
| *** EmilienM has quit IRC | 16:36 | |
| *** utlemming has quit IRC | 16:38 | |
| *** colinmcnamara has joined #openstack-dev | 16:39 | |
| *** ecarlin has joined #openstack-dev | 16:42 | |
| *** Ryan_Lane has joined #openstack-dev | 16:42 | |
| *** danwent has joined #openstack-dev | 16:43 | |
| *** cloudvirt has quit IRC | 16:45 | |
| *** rnirmal has quit IRC | 16:46 | |
| *** anniec has joined #openstack-dev | 16:46 | |
| *** EmilienM has joined #openstack-dev | 16:47 | |
| *** Shrews has joined #openstack-dev | 16:48 | |
| *** utlemming has joined #openstack-dev | 16:49 | |
| *** adjohn has joined #openstack-dev | 16:50 | |
| *** cloudvirt has joined #openstack-dev | 16:53 | |
| *** rohitk_1 has quit IRC | 16:55 | |
| *** sulochan has quit IRC | 16:56 | |
| *** inteq has joined #openstack-dev | 16:57 | |
| *** samkottler is now known as samkottler|lunch | 16:58 | |
| *** nati_ueno has joined #openstack-dev | 16:59 | |
| *** rohitk has joined #openstack-dev | 17:00 | |
| *** thingee_zz is now known as thingee | 17:03 | |
| *** dolphm_ has quit IRC | 17:03 | |
| *** markmcclain1 has quit IRC | 17:03 | |
| *** anniec_ has joined #openstack-dev | 17:06 | |
| *** ecarlin_ has joined #openstack-dev | 17:06 | |
| *** ecarlin has quit IRC | 17:06 | |
| *** ecarlin_ is now known as ecarlin | 17:06 | |
| *** issackel_ has joined #openstack-dev | 17:07 | |
| *** ecarlin has quit IRC | 17:07 | |
| *** ecarlin has joined #openstack-dev | 17:07 | |
| *** anniec has quit IRC | 17:08 | |
| *** anniec_ is now known as anniec | 17:08 | |
| *** nunosantos has joined #openstack-dev | 17:09 | |
| *** PotHix has joined #openstack-dev | 17:12 | |
| *** hub_cap has joined #openstack-dev | 17:13 | |
| *** colinmcnamara has quit IRC | 17:14 | |
| *** garyk has joined #openstack-dev | 17:15 | |
| vishy | jgriffith: ping | 17:18 |
| jgriffith | vishy: hey | 17:19 |
| *** dprince_ has quit IRC | 17:19 | |
| *** dprince has joined #openstack-dev | 17:19 | |
| vishy | jgriffith: have we cleaned out the cinder db and migrations yet? | 17:19 |
| *** dubsquared1 has quit IRC | 17:19 | |
| jgriffith | vishy: Yep, unless there's stuff I missed | 17:20 |
| jgriffith | vishy: I went threw those a couple months ago | 17:20 |
| jgriffith | vishy: We have a single db migration file right now | 17:21 |
| vishy | jgriffith: cool, couldn't remember | 17:21 |
| vishy | jgriffith: how about keeping pace with common? | 17:21 |
| jgriffith | vishy: Pretty good there thanks to markmc and vincent | 17:21 |
| jgriffith | vishy: I'll be due to take another pass through those next week I think | 17:22 |
| *** jdurgin has joined #openstack-dev | 17:23 | |
| *** Guest38070 has quit IRC | 17:24 | |
| *** rohitk has quit IRC | 17:24 | |
| *** johnpur has left #openstack-dev | 17:24 | |
| jgriffith | vishy: If you're noticing something I'm missing let me know | 17:25 |
| *** jaypipes has quit IRC | 17:25 | |
| *** colinmcnamara has joined #openstack-dev | 17:30 | |
| *** harlowja has joined #openstack-dev | 17:30 | |
| *** ayoung-afk is now known as ayoung | 17:31 | |
| *** darraghb has quit IRC | 17:31 | |
| hub_cap | im interested in adding a library to a project, whats the process for petitioning to add a library? | 17:33 |
| *** joesavak has joined #openstack-dev | 17:34 | |
| *** markmcclain has joined #openstack-dev | 17:36 | |
| vishy | hub_cap: it is a prop into the common library project now http://lists.openstack.org/pipermail/openstack-dev/2012-August/000424.html | 17:36 |
| *** negronjl has quit IRC | 17:36 | |
| hub_cap | <3 vishy | 17:36 |
| *** negronjl has joined #openstack-dev | 17:36 | |
| hub_cap | thats good actually cuz i want to add a library to the keystone middleware (hopefully i wont get ran out of town with torches and pitchforks) | 17:37 |
| *** danpb has quit IRC | 17:38 | |
| *** hub_cap has quit IRC | 17:39 | |
| *** ecarlin has quit IRC | 17:39 | |
| *** hub_cap has joined #openstack-dev | 17:39 | |
| hub_cap | so vishy mtaylor if i want to add a library to openstack/requirements, should i have a BP detailing why, i assume? Whats the 'process' persay | 17:42 |
| hub_cap | just prop it? BP it? etc... | 17:42 |
| *** jaypipes has joined #openstack-dev | 17:45 | |
| *** ayoung is now known as ayoung-afk | 17:50 | |
| *** joesavak has quit IRC | 17:58 | |
| *** kyriakos has quit IRC | 17:59 | |
| *** zaneb has quit IRC | 18:00 | |
| *** joesavak has joined #openstack-dev | 18:02 | |
| *** colinmcnamara has quit IRC | 18:06 | |
| *** epim has joined #openstack-dev | 18:08 | |
| *** dubsquared has joined #openstack-dev | 18:18 | |
| *** samkottler|lunch is now known as samkottler | 18:20 | |
| *** harlowja has quit IRC | 18:21 | |
| *** joesavak has quit IRC | 18:24 | |
| adam_g | hub_cap: just curious, which library? | 18:25 |
| *** harlowja has joined #openstack-dev | 18:26 | |
| hub_cap | python-beaker adam_g | 18:26 |
| hub_cap | a generic cache library so we can specify cache types | 18:26 |
| hub_cap | ie memory, db, memcache | 18:26 |
| adam_g | cool, thanks | 18:27 |
| hub_cap | np, seem sane adam_g? | 18:27 |
| *** colinmcnamara has joined #openstack-dev | 18:27 | |
| *** sdake has quit IRC | 18:28 | |
| *** sdake has joined #openstack-dev | 18:28 | |
| *** dubsquared has quit IRC | 18:31 | |
| *** zaitcev has joined #openstack-dev | 18:32 | |
| *** joesavak has joined #openstack-dev | 18:32 | |
| *** jsavak has joined #openstack-dev | 18:35 | |
| *** joesavak has quit IRC | 18:37 | |
| *** dubsquared has joined #openstack-dev | 18:42 | |
| *** wiliam has quit IRC | 18:43 | |
| *** hub_cap is now known as hub_cap_onoes_go | 18:44 | |
| mtaylor | hub_cap_onoes_go: I'd just prop it | 18:44 |
| mtaylor | hub_cap_onoes_go: we just need to ensure that the right folks are in the review team for the project :) | 18:45 |
| *** dolphm has joined #openstack-dev | 18:51 | |
| *** erikzaadi has joined #openstack-dev | 18:51 | |
| *** hub_cap_onoes_go is now known as hub_cap | 18:53 | |
| hub_cap | mtaylor: <3 | 18:53 |
| *** adiantum has joined #openstack-dev | 18:53 | |
| *** dubsquared has quit IRC | 18:58 | |
| *** dhellmann_ has joined #openstack-dev | 18:58 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:00 | |
| *** dhellmann has quit IRC | 19:01 | |
| *** dhellmann_ is now known as dhellmann | 19:01 | |
| *** dubsquared has joined #openstack-dev | 19:04 | |
| *** jsavak has quit IRC | 19:04 | |
| *** colinmcnamara has quit IRC | 19:05 | |
| *** matiu has quit IRC | 19:12 | |
| *** johnpostlethwait has joined #openstack-dev | 19:15 | |
| *** ExxonValdeez has left #openstack-dev | 19:16 | |
| *** steveb_ has joined #openstack-dev | 19:18 | |
| *** erikzaadi has quit IRC | 19:21 | |
| *** ayoung-afk is now known as ayoung | 19:21 | |
| *** colinmcnamara has joined #openstack-dev | 19:21 | |
| *** joesavak has joined #openstack-dev | 19:27 | |
| *** matwood has quit IRC | 19:27 | |
| *** adiantum_ has joined #openstack-dev | 19:27 | |
| *** ecarlin has joined #openstack-dev | 19:29 | |
| *** adiantum has quit IRC | 19:30 | |
| *** joesavak has quit IRC | 19:31 | |
| *** pixelbeat has quit IRC | 19:32 | |
| *** joesavak has joined #openstack-dev | 19:33 | |
| sandywalsh | so, if I want to add something to nova.openstack.common ... do I add it in nova or openstack-common? (is the nova.openstack.common in nova some sort of symlink?) | 19:34 |
| sandywalsh | (like a git submodule) | 19:35 |
| jgriffith | sdague: Oops, looks like you were doing all of them | 19:36 |
| jgriffith | sdague: Well, Cinder's done anyway. | 19:36 |
| sdague | jgriffith: yep, though for some reason cinder didn't auto update | 19:37 |
| sdague | jgriffith: is cinder's linkage from commit messages working? | 19:37 |
| jgriffith | sdague: should be, unless something broke recently | 19:37 |
| sdague | https://bugs.launchpad.net/quantum/+bug/1035035 - all the projects except cinder updated when I pushed patches | 19:37 |
| uvirtbot | Launchpad bug 1035035 in openstack-common "git review autorebase should "die in a fire"" [Undecided,In progress] | 19:37 |
| sdague | sandywalsh: you add it to openstack-common, then sync it to nova after it lands in common | 19:37 |
| *** johnpostlethwait has quit IRC | 19:38 | |
| jgriffith | sdague: yeah, something's broke | 19:38 |
| sdague | sandywalsh: http://wiki.openstack.org/CommonLibrary | 19:38 |
| sandywalsh | sdague: thanks, I'll have a look | 19:38 |
| jgriffith | sdague: But I don't see your patch in the dashboard eitehr?? | 19:38 |
| jgriffith | sdague: NM got it | 19:39 |
| sdague | jgriffith: https://review.openstack.org/#/c/11097/ | 19:39 |
| jgriffith | sdague: yup, notifications don't seem to be working on Cinder :( | 19:39 |
| sdague | jgriffith: well that's one way to find project inconsitencies :) | 19:40 |
| jgriffith | haha | 19:40 |
| garyk | nati_ueno: ping | 19:43 |
| annegentle | does any one know if the os-volumes extension creates volumes using the Compute API? | 19:44 |
| annegentle | and if so, is it the "same thing" as using the Volume API's call to make a volume? | 19:45 |
| *** dprince has quit IRC | 19:45 | |
| mtaylor | jgriffith: vish has already submitted a patch for autorebase die in a fire... | 19:46 |
| mtaylor | jgriffith: https://review.openstack.org/11081 | 19:46 |
| jgriffith | mtaylor: bahh... vishy, sdague and me now | 19:46 |
| jgriffith | mtaylor: Any idea why notifications have stopped on Cinder? | 19:47 |
| jeblair | jgriffith: you mean bug links? | 19:47 |
| jgriffith | jeblair: bug links and email notifications of gerrit submissions | 19:47 |
| jeblair | jgriffith: hudson-openstack was missing from cinder-bugs group; i just added it; should be fixed for future events now | 19:47 |
| jgriffith | jeblair: yeah, figured it out | 19:48 |
| jgriffith | jeblair: thanks!!! | 19:48 |
| jeblair | np, i'm sure it's something i forgot to do a while back. | 19:48 |
| jeblair | jgriffith: email from gerrit, otoh, is different | 19:48 |
| jgriffith | jeblair: Actually I think we had it initially then turned it off in one of our squashes | 19:48 |
| jeblair | jgriffith: don't know why that wouldn't work -- you've checked that you have the project watches set up the way you want? | 19:48 |
| jeblair | jgriffith: it's a per-user/per-repo option | 19:49 |
| jeblair | https://review.openstack.org/#/settings/projects | 19:49 |
| jgriffith | jeblair: Yeah, I'm good there. Was fine until this afternoon | 19:49 |
| *** colinmcnamara has quit IRC | 19:49 | |
| *** thingee is now known as thingee_zz | 19:49 | |
| jgriffith | jeblair: Lemme give it a while and see if it's just gmail/chrome crap | 19:49 |
| *** mnewby has quit IRC | 19:49 | |
| jgriffith | jeblair: Well crap... I've missed all sorts of email this afternoon | 19:50 |
| nati_ueno | garyk: pong | 19:51 |
| jeblair | jgriffith: looks like gmail is at least accepting mail for you from the gerrit server in general | 19:51 |
| jgriffith | jeblair: yeah, I logged out back in and got a ton'o emails | 19:52 |
| jgriffith | jeblair: Problem was on my end, sorry | 19:52 |
| jeblair | jgriffith: go ajax! :) | 19:52 |
| *** matwood has joined #openstack-dev | 19:52 | |
| sdague | mtaylor: oh, I thought we were going to fix it client side | 19:55 |
| mtaylor | sdague: we are | 19:56 |
| mtaylor | sdague: patch has been submitted, I like it | 19:56 |
| mtaylor | sdague: hopefully we'll land that and one other today and then cut a new git review release | 19:57 |
| *** cdub has quit IRC | 19:57 | |
| sdague | mtaylor: ok, so the .gitreview patches that I just sent in aren't needed? | 19:57 |
| sdague | markmc realized we could just turn it off there as well | 19:57 |
| mtaylor | sdague: nope we should have something out by today | 19:57 |
| *** ayoung_ has joined #openstack-dev | 19:58 | |
| *** dolphm has quit IRC | 19:58 | |
| *** cdub has joined #openstack-dev | 19:58 | |
| *** ayoung has quit IRC | 19:58 | |
| sdague | mtaylor: that review from vishy just adds force rebase, right, or am I missing something? | 19:59 |
| mtaylor | sdague: you are. what it actually does is just does a test rebase | 19:59 |
| mtaylor | sdague: if it works, it resets you back to pre-rebase state before submitting | 19:59 |
| mtaylor | sdague: if it doesn't, it'll bomb out and tell you that you are unmergable with upstream | 20:00 |
| *** adiantum_ has quit IRC | 20:00 | |
| mtaylor | so the rebase will only get triggered when you actually _need_ to rebase because gerrit won't be able to merge your patch | 20:00 |
| clarkb | and you would need to do it manually or use the -F option | 20:01 |
| *** jsavak has joined #openstack-dev | 20:02 | |
| vishy | sdague: it rolls back rebases if they are not necessary | 20:03 |
| hub_cap | so mtaylor who do u think the "right" people would be to get the beaker thing looked @? Its going to be used in the middleware, so it doesnt really affect keystone, but it does the projects that use it. I assume glance and nova ptls, or some representation of those teams? cinder possibly as well? im not fully sure what projects use the middleware besides nova | 20:04 |
| vishy | mtaylor: I just realized that the force rebase option is silly, the only time when it does anything is if the rebase fails | 20:04 |
| *** joesavak has quit IRC | 20:04 | |
| vishy | mtaylor: oh actually cancel that I just confused myself for a second :) | 20:04 |
| *** ecarlin has quit IRC | 20:05 | |
| *** ecarlin has joined #openstack-dev | 20:05 | |
| vishy | I can't come up with a situation where it would actually be useful though. | 20:05 |
| *** joesavak has joined #openstack-dev | 20:05 | |
| sdague | well, it's an option that would be needed to counteract defaultrebase=0 in the configs, if that existed for someone | 20:06 |
| clarkb | doesnt -F start the failed rebase for you? | 20:06 |
| vishy | clarkb: no -F leaves the rebase in | 20:07 |
| vishy | -F gives you the current behavior | 20:07 |
| clarkb | ah | 20:07 |
| vishy | which is to rebase no matter what | 20:07 |
| *** jsavak has quit IRC | 20:07 | |
| *** hub-cap has joined #openstack-dev | 20:08 | |
| vishy | i can't really come up with a situation where you would want to do that though | 20:08 |
| *** jsavak has joined #openstack-dev | 20:08 | |
| *** hub_cap has quit IRC | 20:08 | |
| *** hub-cap is now known as hub_cap | 20:08 | |
| *** dubsquared has quit IRC | 20:08 | |
| sdague | mtaylor: people will have to upgrade their git-review to get the change, correct? so do we want to leave the .gitreview patches there? just confirming before I got abandon them | 20:10 |
| sdague | s/got/go/ ... need more caffine | 20:11 |
| hub_cap | vishy: who should i get to review the requirement add for python-beaker on your team? | 20:11 |
| vishy | hub_cap: define my team? | 20:12 |
| hub_cap | nova-core | 20:12 |
| *** joesavak has quit IRC | 20:12 | |
| hub_cap | :) | 20:12 |
| vishy | send me a link? | 20:13 |
| hub_cap | vishy https://review.openstack.org/#/c/11099/ | 20:13 |
| *** joesavak has joined #openstack-dev | 20:13 | |
| hub_cap | its failed in jenkins but mtaylor is fixing that | 20:13 |
| hub_cap | effectively id like to make a cache that is pluggable in the middleware so i can use whatever the Eff i want (that beaker supports :D ) | 20:13 |
| *** jsavak has quit IRC | 20:14 | |
| vishy | hub_cap: it isn't called beaker? | 20:14 |
| hub_cap | https://launchpad.net/ubuntu/precise/i386/python-beaker/1.5.4-4 | 20:14 |
| *** steveb_ has quit IRC | 20:15 | |
| vishy | that is a pip-requires though... | 20:15 |
| hub_cap | oh duh | 20:15 |
| hub_cap | haha welllllll then | 20:15 |
| vishy | :) | 20:15 |
| * vishy is an ace at reviews | 20:15 | |
| hub_cap | yes u are | 20:15 |
| hub_cap | <3 vishy | 20:16 |
| *** ecarlin has quit IRC | 20:22 | |
| *** vanchester has quit IRC | 20:22 | |
| *** matwood has quit IRC | 20:23 | |
| hub_cap | ok vishy jenkins still hates me, but, now pip doesnt https://review.openstack.org/#/c/11099/ | 20:23 |
| *** thingee_zz is now known as thingee | 20:24 | |
| clarkb | I will fix that | 20:26 |
| vishy | i guess my only complaint now is that you don't know the order of the alphabet... | 20:26 |
| vishy | hub_cap: ^^ | 20:26 |
| vishy | :p | 20:27 |
| sdague | :) | 20:27 |
| clarkb | mtaylor: is puppet not doing the bare git repo initing yet? | 20:27 |
| * clarkb does it by hand | 20:28 | |
| hub_cap | OMFG vishy this is what i get when i friggen do this wile trying to listen in on a conf call | 20:29 |
| hub_cap | geesus im a MORON | 20:29 |
| *** joesavak has quit IRC | 20:29 | |
| hub_cap | obviously u havent heard the rule k before c except after u | 20:29 |
| clarkb | hub_cap: if you push another patchset to that change I think jenkins will be happy now. Or you can wait for somoene that is a gerrit Administrator to force replication to happen (mtaylor can do that) | 20:30 |
| hub_cap | haha i _just_ did too, i can amend the comment since it was complaing about word wrap | 20:31 |
| *** epim has quit IRC | 20:31 | |
| *** JStoker has quit IRC | 20:32 | |
| *** epim has joined #openstack-dev | 20:33 | |
| *** vanchester has joined #openstack-dev | 20:34 | |
| hub_cap | ok clarkb vishy ive gotten some updates from jaypipes on the ML, so ill bring that in to consensus before i attempt to make this land | 20:34 |
| jaypipes | hub_cap: only trying to help :) | 20:35 |
| hub_cap | no dude, it does help | 20:37 |
| *** anniec_ has joined #openstack-dev | 20:37 | |
| hub_cap | im not saying anything otehr than id like to get everything worked out, i dont want something to land to be stripped out or not used | 20:37 |
| jaypipes | hub_cap: totally | 20:37 |
| hub_cap | i think its a fyne idea jaypipes | 20:37 |
| hub_cap | ;) | 20:37 |
| jaypipes | hub_cap: I'm 100% supportive of your idea and work | 20:38 |
| hub_cap | i mean, GOD jaypipes always more work for me | 20:38 |
| hub_cap | :D | 20:38 |
| jaypipes | lol | 20:38 |
| hub_cap | ill reply on ML, thx again jaypipes | 20:38 |
| jaypipes | np! | 20:38 |
| *** rkukura has quit IRC | 20:39 | |
| *** JStoker has joined #openstack-dev | 20:40 | |
| *** anniec has quit IRC | 20:40 | |
| *** anniec_ is now known as anniec | 20:40 | |
| *** matwood has joined #openstack-dev | 20:41 | |
| *** jtran has joined #openstack-dev | 20:43 | |
| *** thingee is now known as thingee_zz | 20:47 | |
| *** ExxonValdeez has joined #openstack-dev | 20:49 | |
| *** nati_uen_ has joined #openstack-dev | 20:49 | |
| *** nati_ueno has quit IRC | 20:49 | |
| *** clarkb has quit IRC | 20:51 | |
| *** clarkb1 has joined #openstack-dev | 20:51 | |
| *** alanmac has joined #openstack-dev | 20:51 | |
| *** clarkb1 is now known as clarkb | 20:52 | |
| *** sacharya has quit IRC | 20:53 | |
| hub_cap | clarkb: jenkins hates me https://jenkins.openstack.org/job/gate-requirements-merge/4/consoleFull | 20:55 |
| hub_cap | lets take it to infra channel | 20:56 |
| clarkb | I think that means I missed something | 20:56 |
| *** dprince has joined #openstack-dev | 20:59 | |
| *** ExxonValdeez has left #openstack-dev | 21:00 | |
| vishy | nova meeting in openstack-meeting | 21:00 |
| *** alexpilotti has quit IRC | 21:01 | |
| *** matwood has quit IRC | 21:03 | |
| *** ecarlin has joined #openstack-dev | 21:03 | |
| *** maurosr has quit IRC | 21:04 | |
| *** cbj has quit IRC | 21:04 | |
| *** bhuvan has joined #openstack-dev | 21:06 | |
| bhuvan | dhellmann: do you think, i should fix env() method to handle boolean values set in any env variables? | 21:08 |
| *** mtreinish has quit IRC | 21:08 | |
| bhuvan | like, if the value is 0 or 'False', set it to False; if the value is 1 or 'True', set it to True. | 21:08 |
| *** dolphm_ has joined #openstack-dev | 21:08 | |
| *** matwood has joined #openstack-dev | 21:09 | |
| *** markmcclain has quit IRC | 21:11 | |
| *** lts has quit IRC | 21:20 | |
| *** markmcclain has joined #openstack-dev | 21:23 | |
| *** hub_cap has quit IRC | 21:23 | |
| *** sacharya has joined #openstack-dev | 21:23 | |
| *** vanchester has quit IRC | 21:24 | |
| *** vanchester has joined #openstack-dev | 21:25 | |
| *** alexpilotti has joined #openstack-dev | 21:26 | |
| *** sacharya has quit IRC | 21:26 | |
| *** vanchester has quit IRC | 21:27 | |
| bhuvan | dhellmann: nevermind. posted code snippet to the review request, 9497. | 21:31 |
| *** cloudvirt has quit IRC | 21:31 | |
| *** johnpostlethwait has joined #openstack-dev | 21:34 | |
| *** jakedahn_zz is now known as jakedahn | 21:36 | |
| *** johnpostlethwait has quit IRC | 21:36 | |
| *** rkukura has joined #openstack-dev | 21:40 | |
| *** steveb_ has joined #openstack-dev | 21:41 | |
| *** ecarlin has quit IRC | 21:44 | |
| *** ayoung_ is now known as ayoung | 21:44 | |
| *** epim has quit IRC | 21:45 | |
| *** ecarlin has joined #openstack-dev | 21:46 | |
| *** dhellmann has quit IRC | 21:50 | |
| *** markmcclain has quit IRC | 21:51 | |
| *** thingee_zz is now known as thingee | 21:51 | |
| *** ayoung has quit IRC | 21:56 | |
| *** flaviamissi has quit IRC | 22:00 | |
| *** avishay has joined #openstack-dev | 22:03 | |
| *** Dr_Who has quit IRC | 22:09 | |
| *** fsouza_ has joined #openstack-dev | 22:09 | |
| avishay | mikal: ping | 22:11 |
| *** jgriffith has quit IRC | 22:11 | |
| *** fsouza has quit IRC | 22:12 | |
| *** jgriffith has joined #openstack-dev | 22:12 | |
| *** markmc is now known as mcaway | 22:12 | |
| *** dprince has quit IRC | 22:12 | |
| avishay | If someone could review our volume driver for Folsom I would really appreciate it. It's already merged into Cinder. Thanks! https://review.openstack.org/#/c/10535/ | 22:19 |
| *** ecarlin has quit IRC | 22:20 | |
| *** edygarcia_ has joined #openstack-dev | 22:22 | |
| *** datsun180b has quit IRC | 22:25 | |
| *** edygarcia has quit IRC | 22:26 | |
| *** edygarcia_ is now known as edygarcia | 22:26 | |
| *** sandywalsh has quit IRC | 22:28 | |
| *** tongli has quit IRC | 22:29 | |
| *** dubsquared has joined #openstack-dev | 22:32 | |
| *** e1mer has joined #openstack-dev | 22:38 | |
| *** jog0 has joined #openstack-dev | 22:47 | |
| *** fsouza_ has quit IRC | 22:47 | |
| *** dubsquared has quit IRC | 22:48 | |
| *** oneiroi is now known as oneiroi|gone | 22:49 | |
| *** spiffxp has quit IRC | 22:50 | |
| adam_g | has anyone out there been usingnova on kernels new than 3.2.x? | 22:53 |
| dansmith | vishy: I just slammed up a new version of the task_state revert thing.. you're on the review list | 22:53 |
| *** maurosr has joined #openstack-dev | 22:53 | |
| dansmith | vishy: the testing of it without the lock throwing an exception (which is how it used to be) is a little funky, so I'm sure there's a -1 in there | 22:53 |
| dansmith | vishy: ...and markmc is conveniently not around :P | 22:54 |
| asalkeld | adam_g, fedora 17 (Linux elf 3.4.6-2.fc17.x86_64) | 22:56 |
| *** ewindisch has quit IRC | 22:56 | |
| vishy | dansmith: lol setting none is funny | 22:58 |
| *** nunosantos has quit IRC | 22:58 | |
| dansmith | vishy: ...well, it works :) | 22:59 |
| vishy | dansmith: it won't work if there is a method that doesn't call a driver though :( | 22:59 |
| *** edygarcia has quit IRC | 22:59 | |
| dansmith | vishy: true, but those all should, I think | 22:59 |
| *** epim has joined #openstack-dev | 22:59 | |
| vishy | dansmith: like this one :) https://review.openstack.org/#/c/10908/ | 22:59 |
| *** edygarcia has joined #openstack-dev | 23:00 | |
| *** epim has quit IRC | 23:00 | |
| dansmith | I guess I should make sure than an exception is triggered to avoid a false positive | 23:00 |
| *** epim has joined #openstack-dev | 23:00 | |
| dansmith | but the ones in the list it's testing do, I think | 23:00 |
| *** jtran has quit IRC | 23:00 | |
| dansmith | any other ideas on how to force an exception without disrupting the decorator stack on those methods? | 23:00 |
| vishy | dansmith: I can't really think of anything better except maybe stubbing out wrap_instance_fault to raise an exception | 23:01 |
| dansmith | well, | 23:01 |
| dansmith | that disrupts the decorator stack and doesn't do what you would think | 23:01 |
| dansmith | that's where I started first | 23:01 |
| vishy | he | 23:01 |
| *** alexpilotti has quit IRC | 23:01 | |
| *** jgriffith has quit IRC | 23:02 | |
| dansmith | the other thing would be to make wrap_instance_fault check a global "are we testing" flag and throw a bogus error, but that's rather ugly :) | 23:02 |
| *** epim_ has joined #openstack-dev | 23:02 | |
| adam_g | asalkeld: with which networking setup? | 23:03 |
| vishy | really something like stubs.Set(manager, 'wrap_instance_fault', new_decorator) doesn't work? | 23:03 |
| vishy | where the new_decorator just returns a method that raises an exception when it is called? | 23:03 |
| vishy | dansmith: ^^ | 23:03 |
| dansmith | vishy: no, because by the time we're doing that stub, | 23:03 |
| asalkeld | adam_g, network_manager=nova.network.manager.FlatDHCPManager | 23:03 |
| dansmith | the class method has already been created and decorated | 23:04 |
| vishy | i mean that is super ugly, oh yes the decorator is at import time | 23:04 |
| dansmith | so you can change the decorator, but it's too late | 23:04 |
| *** edygarcia has quit IRC | 23:04 | |
| vishy | dansmith: gotcha yeah i don't really have a better way i guess | 23:04 |
| *** epim has quit IRC | 23:04 | |
| *** epim_ is now known as epim | 23:04 | |
| *** ewindisch has joined #openstack-dev | 23:05 | |
| dansmith | vishy: okay.. I could put a "# NOTE(danms): cash reward offered for a better approach" in there if you like :) | 23:05 |
| vishy | you could stub out _get_lock()! | 23:05 |
| vishy | to raise an exception :p | 23:05 |
| adam_g | asalkeld: no issues with DHCP, huh? just curious.. someone on the list confirmed an issue im seeing with DHCP that seems to show up in 3.3 vanilla kernels. | 23:05 |
| asalkeld | nothing obvious yet | 23:06 |
| *** hemna has quit IRC | 23:06 | |
| asalkeld | There were some intermediate bad kernels | 23:06 |
| *** jog0_ has joined #openstack-dev | 23:06 | |
| asalkeld | but this seems good | 23:06 |
| dansmith | vishy: yeah, although that will only work for things that also check the lock, | 23:07 |
| dansmith | which is probably the same list right now, but might be a smaller total list | 23:07 |
| adam_g | i should say, the problem seems to affect all vanilla kernels ive tested since 3.3 :) | 23:07 |
| dansmith | vishy: but, I'm happy to make that change and add a false positive check in the test | 23:07 |
| *** jog0 has quit IRC | 23:09 | |
| *** jog0_ is now known as jog0 | 23:09 | |
| *** rpedde is now known as rpedde_away | 23:13 | |
| *** lloydde has quit IRC | 23:14 | |
| *** belliott has quit IRC | 23:15 | |
| *** belliott has joined #openstack-dev | 23:16 | |
| *** jog0 has quit IRC | 23:16 | |
| *** jog0 has joined #openstack-dev | 23:16 | |
| *** dolphm_ has quit IRC | 23:17 | |
| *** fsouza has joined #openstack-dev | 23:17 | |
| *** anniec has quit IRC | 23:20 | |
| *** ayoung has joined #openstack-dev | 23:20 | |
| *** johnpostlethwait has joined #openstack-dev | 23:23 | |
| dansmith | vishy: okay, that worked for all the cases in the test.. lucky #10 is up, and with the pep8 issue fixed | 23:24 |
| *** harlowja_ has joined #openstack-dev | 23:25 | |
| *** harlowja has quit IRC | 23:25 | |
| *** harlowja_ is now known as harlowja | 23:25 | |
| *** maurosr has quit IRC | 23:26 | |
| *** spiffxp has joined #openstack-dev | 23:28 | |
| *** pixelbeat has joined #openstack-dev | 23:29 | |
| *** issackel_ has quit IRC | 23:31 | |
| *** cp16net is now known as cp16net|away | 23:33 | |
| *** cp16net|away is now known as cp16net | 23:33 | |
| *** ohnoimdead has joined #openstack-dev | 23:35 | |
| *** ohnoimdead has quit IRC | 23:35 | |
| *** flaviamissi has joined #openstack-dev | 23:35 | |
| *** littleidea has quit IRC | 23:37 | |
| *** ncode has quit IRC | 23:38 | |
| *** markmcclain has joined #openstack-dev | 23:38 | |
| *** littleidea has joined #openstack-dev | 23:39 | |
| cdub | markmcclain: hey, have a moment to help me understand the netns requirement (or point me to the blueprint, etc)? | 23:39 |
| *** eglynn has quit IRC | 23:42 | |
| *** ecarlin has joined #openstack-dev | 23:42 | |
| markmcclain | yes | 23:45 |
| markmcclain | https://blueprints.launchpad.net/quantum/+spec/dhcp-overlapping-ips | 23:47 |
| *** edygarcia has joined #openstack-dev | 23:47 | |
| cdub | markmcclain: cool, looking at it from the point of view of not having ip netns support in various distros, so wanted to see how to deal w/ that (creat alternative, whatever) | 23:47 |
| cdub | markmcclain: thanks | 23:48 |
| markmcclain | the bp is little light | 23:48 |
| cdub | markmcclain: ok, so it is ;) | 23:48 |
| markmcclain | one of the things we discussed is making this optional for older distros | 23:49 |
| *** joesavak has joined #openstack-dev | 23:49 | |
| cdub | markmcclain: the part i don't get is why the dnsmasq process isn't per network, and therefore only handing out addrs for that network (overlapping or not) | 23:49 |
| markmcclain | the dnsmasq process is per network | 23:50 |
| markmcclain | but you can't have two interfaces on the host that have IP addresses on overlapping IP ranges | 23:51 |
| *** jsavak has joined #openstack-dev | 23:51 | |
| cdub | well, you can, routing just gets interesting | 23:52 |
| *** markmcclain has quit IRC | 23:52 | |
| *** markmcclain has joined #openstack-dev | 23:52 | |
| cdub | markmcclain: which interfaces are w/ overlapping ip ranges? | 23:53 |
| *** joesavak has quit IRC | 23:54 | |
| markmcclain | any of them… since Quantum tenants can create any network ip range so the interface can be spun in the same range as the management network | 23:56 |
| *** Mandell_ has joined #openstack-dev | 23:58 | |
| *** Mandell has quit IRC | 23:58 | |
| cdub | the traffic should be isolated by being w/in a quantum network though (i.e. unique vlan) | 23:59 |
| markmcclain | it is until traffic has to traverse the host | 23:59 |
| markmcclain | if the host knows about 2 192.168.1.3's which one does it talk to? | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
