| *** yb_ has quit IRC | 00:00 | |
| *** patelna has quit IRC | 00:01 | |
| *** nati_ueno has quit IRC | 00:02 | |
| *** itooon has quit IRC | 00:02 | |
| *** ywu_ has quit IRC | 00:03 | |
| *** littleidea has joined #openstack-dev | 00:03 | |
| *** ywu_ has joined #openstack-dev | 00:03 | |
| rmk | danwent: would the use-count be tracked in a new column? | 00:03 |
|---|---|---|
| *** littleidea has quit IRC | 00:04 | |
| danwent | rmk: wrong nick? | 00:04 |
| rmk | err dansmith | 00:04 |
| rmk | yes sorry | 00:04 |
| danwent | np | 00:04 |
| *** jog0 has quit IRC | 00:04 | |
| dansmith | rmk: yeah, I'd expect that'd be best | 00:06 |
| *** littleidea has joined #openstack-dev | 00:06 | |
| *** littleidea has left #openstack-dev | 00:06 | |
| dansmith | rmk: I'm sure there's a good reason why such a scheme wouldn't work or be ideal, but I'm too dumb to know :) | 00:06 |
| rmk | dansmith: I like the idea -- my only concern is that we're trying to get something suitable for backporting, because this patch is likely going into Folsom (https://review.openstack.org/#/c/15153/) | 00:07 |
| *** nunosantos has joined #openstack-dev | 00:08 | |
| rmk | dansmith: and I think changing the schema isn't something we can consider as a backport | 00:08 |
| dansmith | rmk: sure, the refcount approach isn't something you'd want to backport | 00:08 |
| rmk | nods | 00:08 |
| *** slackguru has quit IRC | 00:08 | |
| dansmith | rmk: right, I'm just commenting how it might be done going forward | 00:08 |
| rmk | yeah I think for grizzly, your way is better/cleaner | 00:08 |
| *** devcamca1 has joined #openstack-dev | 00:08 | |
| rmk | so there isn't a bunch of logic all over the place about when you want/dont want deleted flavors | 00:09 |
| *** ywu_ has quit IRC | 00:09 | |
| *** nunosantos has quit IRC | 00:09 | |
| *** ywu_ has joined #openstack-dev | 00:09 | |
| *** nunosantos has joined #openstack-dev | 00:10 | |
| *** JStoker has quit IRC | 00:12 | |
| *** jcmartin has quit IRC | 00:13 | |
| *** JStoker has joined #openstack-dev | 00:14 | |
| openstackgerrit | A change was merged to openstack/openstack-common: Removes redundant string cast https://review.openstack.org/15086 | 00:17 |
| *** ywu_ has quit IRC | 00:18 | |
| *** devcamca1 has quit IRC | 00:18 | |
| *** ywu_ has joined #openstack-dev | 00:18 | |
| *** gongysh has joined #openstack-dev | 00:18 | |
| *** johnpostlethwait has quit IRC | 00:19 | |
| *** littleidea has joined #openstack-dev | 00:20 | |
| *** JStoker has quit IRC | 00:21 | |
| *** littleidea has left #openstack-dev | 00:21 | |
| *** d3llf has quit IRC | 00:27 | |
| rmk | vishy: Thanks for that suggestion -- I was going to lose my mind updating tests | 00:31 |
| rmk | This is way easier | 00:31 |
| *** stevebake has quit IRC | 00:33 | |
| *** JStoker_ has joined #openstack-dev | 00:34 | |
| *** jrodom has joined #openstack-dev | 00:35 | |
| *** arata has quit IRC | 00:38 | |
| *** arata has joined #openstack-dev | 00:39 | |
| *** danwent has quit IRC | 00:40 | |
| *** arata has quit IRC | 00:40 | |
| *** jrodom has quit IRC | 00:42 | |
| *** jtran has quit IRC | 00:43 | |
| *** maurosr has quit IRC | 00:44 | |
| sandywalsh_ | Kiall: hey! | 00:45 |
| *** Ryan_Lane has quit IRC | 00:47 | |
| *** anniec has quit IRC | 00:47 | |
| *** Ryan_Lane has joined #openstack-dev | 00:48 | |
| *** samkottler|out has joined #openstack-dev | 00:49 | |
| *** devcamca1 has joined #openstack-dev | 00:50 | |
| *** mnewby has quit IRC | 00:50 | |
| *** EmilienM has left #openstack-dev | 00:51 | |
| *** roge has joined #openstack-dev | 00:53 | |
| *** devcamca1 has quit IRC | 00:57 | |
| *** jrodom has joined #openstack-dev | 00:58 | |
| *** maurosr has joined #openstack-dev | 00:58 | |
| *** dims has joined #openstack-dev | 00:58 | |
| gongysh | sdague: ping | 01:00 |
| *** nati_ueno has joined #openstack-dev | 01:01 | |
| gongysh | nati_neno: ping | 01:02 |
| *** jrodom has quit IRC | 01:03 | |
| nati_ueno | gongysh: pong | 01:03 |
| gongysh | https://review.openstack.org/#/c/15100/ | 01:03 |
| gongysh | do u known how to allow jenkins to reverify the patch | 01:04 |
| nati_ueno | just comment "reverify" :) | 01:05 |
| nati_ueno | let me try | 01:05 |
| *** roadTripper has quit IRC | 01:06 | |
| *** vipuls has quit IRC | 01:07 | |
| *** Ryan_Lane has quit IRC | 01:08 | |
| *** SlickNik has quit IRC | 01:09 | |
| *** winston-d has joined #openstack-dev | 01:10 | |
| *** danwent has joined #openstack-dev | 01:14 | |
| *** edygarcia has quit IRC | 01:15 | |
| *** Ryan_Lane has joined #openstack-dev | 01:16 | |
| *** xchu has joined #openstack-dev | 01:16 | |
| *** darjeeling has quit IRC | 01:18 | |
| *** JStoker_ is now known as JStoker | 01:19 | |
| *** Ryan_Lane has quit IRC | 01:20 | |
| *** Tross has quit IRC | 01:20 | |
| *** devcamca1 has joined #openstack-dev | 01:24 | |
| *** gyee has quit IRC | 01:29 | |
| *** mlavalle has left #openstack-dev | 01:29 | |
| *** wenjianhn has joined #openstack-dev | 01:34 | |
| openstackgerrit | A change was merged to openstack/keystone: Imported Translations from Transifex https://review.openstack.org/15161 | 01:35 |
| *** devcamca1 has quit IRC | 01:35 | |
| *** dachary has quit IRC | 01:38 | |
| *** dachary1 has joined #openstack-dev | 01:38 | |
| *** Mandell has quit IRC | 01:38 | |
| *** redbeard2 has quit IRC | 01:39 | |
| *** edygarcia has joined #openstack-dev | 01:41 | |
| *** adalbas has quit IRC | 01:43 | |
| *** yamahata has joined #openstack-dev | 01:43 | |
| *** timello has joined #openstack-dev | 01:45 | |
| *** roge has quit IRC | 01:46 | |
| *** adjohn has quit IRC | 01:46 | |
| *** maurosr has quit IRC | 01:53 | |
| *** Tross has joined #openstack-dev | 01:53 | |
| *** Tross has quit IRC | 01:53 | |
| *** maxim has joined #openstack-dev | 02:01 | |
| *** maxim is now known as Guest5053 | 02:01 | |
| *** devcamca1 has joined #openstack-dev | 02:02 | |
| *** benner_ has joined #openstack-dev | 02:03 | |
| *** benner has quit IRC | 02:05 | |
| *** openstackgerrit has quit IRC | 02:05 | |
| *** larsks has joined #openstack-dev | 02:05 | |
| *** openstackgerrit has joined #openstack-dev | 02:05 | |
| larsks | Anybody around? I'm trying to locate the commit containing the version of nova/virt/disk/api.py that's installed on my system, using the python-nova-2012.1.3-1 package from EPEL. | 02:06 |
| larsks | I can't find the corresponding version of this file in the repository. | 02:06 |
| *** bing_bu has joined #openstack-dev | 02:07 | |
| openstackgerrit | A change was merged to openstack/keystone: move hashing user password functions to common/utils https://review.openstack.org/15040 | 02:07 |
| *** entropyworks is now known as Aaton | 02:12 | |
| *** devcamca1 has quit IRC | 02:13 | |
| clarkb | larsks: should be under https://github.com/openstack/nova/tree/2012.1.3 I would think | 02:16 |
| larsks | clarkb: Thanks, I'll take a look. I was poking through stable/essex. | 02:23 |
| *** nati_uen_ has joined #openstack-dev | 02:23 | |
| clarkb | larsks: the tag would be against a commit in stable/essex, but looking at that version number it is probably the tagged commit | 02:23 |
| *** nati_ueno has quit IRC | 02:24 | |
| larsks | clarkb: Still no dice. I checked out 2012.1.3, but the installed file contains a "_setup_selinux_for_keys" method not defined in the 2012.1.3 tag in the repository. | 02:25 |
| larsks | I iterated across all the commits in stable/essex touching that file, and none of them matched :(. Trying to pick up the source rpm right now to see if there are any patches getting applied as part of the packaging process. | 02:25 |
| *** ayoung has quit IRC | 02:28 | |
| *** ywu_ has quit IRC | 02:28 | |
| larsks | clarkb: Oh god, it's 2012.1.3 plus a random series of patches, resulting in a file that doesn't actually exist in the repository. Shoot me now. | 02:30 |
| *** timello has quit IRC | 02:32 | |
| *** dachary1 has quit IRC | 02:33 | |
| *** dachary has joined #openstack-dev | 02:33 | |
| *** xchu has quit IRC | 02:36 | |
| *** alexpilotti has quit IRC | 02:38 | |
| *** Tross has joined #openstack-dev | 02:41 | |
| *** mjfork has quit IRC | 02:45 | |
| *** devcamca1 has joined #openstack-dev | 02:45 | |
| *** ewindisch has quit IRC | 02:47 | |
| *** edygarcia has quit IRC | 02:48 | |
| *** xchu has joined #openstack-dev | 02:49 | |
| *** devcamca1 has quit IRC | 02:53 | |
| *** sandywalsh_ has quit IRC | 02:59 | |
| *** e1mer has joined #openstack-dev | 03:04 | |
| *** maurosr has joined #openstack-dev | 03:11 | |
| *** e1mer has quit IRC | 03:12 | |
| *** maurosr has quit IRC | 03:15 | |
| *** pixelbeat has quit IRC | 03:16 | |
| *** melwitt has quit IRC | 03:16 | |
| *** maurosr has joined #openstack-dev | 03:18 | |
| *** devcamca1 has joined #openstack-dev | 03:20 | |
| *** novas0x2a|laptop has quit IRC | 03:20 | |
| *** tgall_foo has quit IRC | 03:20 | |
| *** novas0x2a|laptop has joined #openstack-dev | 03:20 | |
| *** tgall_foo has joined #openstack-dev | 03:21 | |
| *** maurosr has quit IRC | 03:22 | |
| *** edygarcia has joined #openstack-dev | 03:24 | |
| *** maurosr has joined #openstack-dev | 03:25 | |
| *** Mandell has joined #openstack-dev | 03:27 | |
| *** edygarcia has quit IRC | 03:27 | |
| *** maurosr has quit IRC | 03:29 | |
| *** maurosr has joined #openstack-dev | 03:30 | |
| winston-d | russellb, ping | 03:30 |
| *** tgall_foo has quit IRC | 03:30 | |
| *** samkottler|out is now known as samkottler | 03:31 | |
| *** devcamca1 has quit IRC | 03:32 | |
| *** xchu has quit IRC | 03:34 | |
| *** maurosr has quit IRC | 03:34 | |
| *** maurosr has joined #openstack-dev | 03:35 | |
| *** ewindisch has joined #openstack-dev | 03:36 | |
| openstackgerrit | A change was merged to openstack/cinder: Log the body of an /action https://review.openstack.org/14566 | 03:44 |
| *** tmishler has joined #openstack-dev | 03:45 | |
| *** xchu has joined #openstack-dev | 03:47 | |
| openstackgerrit | A change was merged to openstack/cinder: Allow the user to update a volume's metadata https://review.openstack.org/14772 | 03:48 |
| *** tmishler has quit IRC | 03:56 | |
| *** devcamca1 has joined #openstack-dev | 04:02 | |
| *** mnewby has joined #openstack-dev | 04:02 | |
| *** maurosr has quit IRC | 04:04 | |
| *** devcamca1 has quit IRC | 04:11 | |
| jimfehlig | k | 04:13 |
| *** maurosr has joined #openstack-dev | 04:17 | |
| *** ewindisch has quit IRC | 04:18 | |
| *** maurosr has quit IRC | 04:22 | |
| *** maurosr has joined #openstack-dev | 04:23 | |
| *** lloydde has joined #openstack-dev | 04:23 | |
| *** nunosantos has quit IRC | 04:25 | |
| *** maurosr has quit IRC | 04:27 | |
| *** lloydde has quit IRC | 04:27 | |
| *** vipuls has joined #openstack-dev | 04:27 | |
| *** aeperezt has quit IRC | 04:37 | |
| *** devcamca1 has joined #openstack-dev | 04:38 | |
| *** metral has quit IRC | 04:39 | |
| *** maurosr has joined #openstack-dev | 04:39 | |
| *** metral has joined #openstack-dev | 04:40 | |
| *** maurosr has quit IRC | 04:44 | |
| *** Tross has quit IRC | 04:45 | |
| *** Mandell has quit IRC | 04:46 | |
| *** maurosr has joined #openstack-dev | 04:47 | |
| *** dani4571 has joined #openstack-dev | 04:47 | |
| *** dani4571 has quit IRC | 04:47 | |
| *** adjohn has joined #openstack-dev | 04:48 | |
| *** itooon has joined #openstack-dev | 04:49 | |
| *** devcamca1 has quit IRC | 04:51 | |
| *** maurosr has quit IRC | 04:51 | |
| *** armaan has joined #openstack-dev | 04:54 | |
| *** armaan has left #openstack-dev | 04:54 | |
| *** maurosr has joined #openstack-dev | 04:54 | |
| *** adjohn has quit IRC | 04:55 | |
| *** hattwick has quit IRC | 04:59 | |
| *** Tross has joined #openstack-dev | 04:59 | |
| *** johnpostlethwait has joined #openstack-dev | 05:01 | |
| *** jimfehlig has quit IRC | 05:07 | |
| *** thingee has quit IRC | 05:10 | |
| *** davidha has quit IRC | 05:19 | |
| *** devcamca1 has joined #openstack-dev | 05:19 | |
| *** davidha has joined #openstack-dev | 05:20 | |
| *** adjohn has joined #openstack-dev | 05:29 | |
| *** devcamca1 has quit IRC | 05:29 | |
| *** utlemming has quit IRC | 05:30 | |
| *** yjiang5 has joined #openstack-dev | 05:30 | |
| *** adjohn has quit IRC | 05:31 | |
| *** darjeeling has joined #openstack-dev | 05:38 | |
| *** hattwick has joined #openstack-dev | 05:42 | |
| *** jeffrey4l has joined #openstack-dev | 05:52 | |
| *** devcamca1 has joined #openstack-dev | 05:58 | |
| *** vipuls has quit IRC | 06:01 | |
| *** danwent has quit IRC | 06:02 | |
| *** adjohn has joined #openstack-dev | 06:06 | |
| *** devcamca1 has quit IRC | 06:10 | |
| *** thovden has joined #openstack-dev | 06:12 | |
| *** johnpostlethwait has quit IRC | 06:17 | |
| *** thovden has quit IRC | 06:22 | |
| *** darjeeling has quit IRC | 06:23 | |
| *** samkottler is now known as samkottler|afk | 06:30 | |
| *** darjeeling has joined #openstack-dev | 06:31 | |
| *** adjohn has quit IRC | 06:34 | |
| *** maurosr has quit IRC | 06:35 | |
| *** devcamca1 has joined #openstack-dev | 06:36 | |
| *** devcamca1 has quit IRC | 06:49 | |
| *** jakedahn_zz is now known as jakedahn | 06:52 | |
| *** garyk has joined #openstack-dev | 07:04 | |
| *** utlemming has joined #openstack-dev | 07:04 | |
| *** utlemming has quit IRC | 07:09 | |
| *** devcamca1 has joined #openstack-dev | 07:15 | |
| *** Ryan_Lane has joined #openstack-dev | 07:20 | |
| *** mrunge has joined #openstack-dev | 07:23 | |
| *** devcamca1 has quit IRC | 07:28 | |
| *** ondergetekende has joined #openstack-dev | 07:34 | |
| *** bing_bu has quit IRC | 07:36 | |
| *** garyk has quit IRC | 07:38 | |
| *** dachary has quit IRC | 07:39 | |
| *** dachary has joined #openstack-dev | 07:41 | |
| *** reidrac has joined #openstack-dev | 07:49 | |
| *** Ryan_Lane has quit IRC | 07:49 | |
| *** afazekas has joined #openstack-dev | 07:49 | |
| *** brucer is now known as brucer_afk | 07:54 | |
| *** devcamca1 has joined #openstack-dev | 07:55 | |
| *** brucer_afk has quit IRC | 07:56 | |
| *** ijw11 has joined #openstack-dev | 07:57 | |
| *** garyk has joined #openstack-dev | 07:58 | |
| *** reidrac has quit IRC | 08:03 | |
| *** utlemming has joined #openstack-dev | 08:04 | |
| *** utlemming has quit IRC | 08:05 | |
| *** utlemming has joined #openstack-dev | 08:05 | |
| *** ijw11 is now known as ijw | 08:06 | |
| *** jakedahn is now known as jakedahn_zz | 08:06 | |
| *** mindpixel has joined #openstack-dev | 08:07 | |
| *** devcamca1 has quit IRC | 08:07 | |
| *** utlemming has quit IRC | 08:09 | |
| *** reidrac has joined #openstack-dev | 08:11 | |
| *** metral_ has joined #openstack-dev | 08:11 | |
| *** metral has quit IRC | 08:14 | |
| *** metral_ is now known as metral | 08:14 | |
| *** metral_ has joined #openstack-dev | 08:23 | |
| *** bing_bu has joined #openstack-dev | 08:24 | |
| *** spn has quit IRC | 08:24 | |
| *** spn has joined #openstack-dev | 08:25 | |
| *** TerryH has joined #openstack-dev | 08:26 | |
| *** metral has quit IRC | 08:27 | |
| *** metral_ has quit IRC | 08:28 | |
| *** stevebake has joined #openstack-dev | 08:31 | |
| *** bing_bu has quit IRC | 08:33 | |
| *** devcamca1 has joined #openstack-dev | 08:34 | |
| *** dave_mcn has joined #openstack-dev | 08:37 | |
| *** koolhead17 has quit IRC | 08:43 | |
| *** bing_bu has joined #openstack-dev | 08:44 | |
| *** zing has joined #openstack-dev | 08:46 | |
| *** devcamca1 has quit IRC | 08:46 | |
| *** nati_uen_ has quit IRC | 08:50 | |
| *** davidha has quit IRC | 08:53 | |
| *** metral has joined #openstack-dev | 08:54 | |
| *** davidha has joined #openstack-dev | 08:54 | |
| *** winston-d has quit IRC | 08:56 | |
| *** dachary has quit IRC | 08:57 | |
| *** TerryH has quit IRC | 08:57 | |
| *** mnewby has quit IRC | 08:58 | |
| garyk | markmc: ping | 09:09 |
| *** xchu has quit IRC | 09:12 | |
| *** henrynash has joined #openstack-dev | 09:12 | |
| ijw | garyk: you live! | 09:13 |
| garyk | ijw: true. how are you? | 09:14 |
| *** devcamca1 has joined #openstack-dev | 09:14 | |
| ijw | Not bad. Trying to remember what timezones we have on this side of the Atlantic | 09:14 |
| ijw | Go play with the VIF plugging wiki page | 09:14 |
| garyk | ijw: saw your attachment to the blueprint. it is very similar to what i was thinking. you got there before me :) | 09:14 |
| *** darraghb has joined #openstack-dev | 09:14 | |
| garyk | ijw: soon i'll post some comments from the peanut gallery | 09:15 |
| ijw | garyk: that's cos you're slow. Also cos the Quantum meeting seemed to have widespread agreement. Though it needs running past the nova guys. | 09:15 |
| ijw | Fits in well with getting shite out of the hypervisor drivers though | 09:15 |
| *** danpb has joined #openstack-dev | 09:15 | |
| garyk | ijw: agree 100%. Without nova agreement it is all for nothing | 09:16 |
| *** dachary has joined #openstack-dev | 09:17 | |
| garyk | ijw: in the mean time i have added a fix for the linux bridge (needs a little review love) https://review.openstack.org/#/c/14830/ | 09:17 |
| ijw | I have one for libvirt drivers so that they work with the noop-firewall, too | 09:17 |
| *** davidha has quit IRC | 09:17 | |
| ijw | … which won't see the light of day till late next week as things stand cos my life is complicated by these customer things | 09:18 |
| garyk | ijw: :) | 09:18 |
| garyk | ijw: mine is complicated with quantum hanging on boot.... | 09:18 |
| *** zul has joined #openstack-dev | 09:18 | |
| ijw | Yeah, stop breaking things | 09:18 |
| garyk | ijw: how about we swap, i'll deal with the customers and you can help debug the issue | 09:19 |
| ijw | You have no idea how tempting that is | 09:19 |
| ijw | Hm, so this turns the gateway off in some circumstances? | 09:20 |
| garyk | ijw: a boer maak n plan | 09:20 |
| ijw | Dear god, inscrutable phrases in cute Dutch | 09:21 |
| ijw | ik heb helemaal geen idee wat je zei. | 09:21 |
| garyk | ijw: the problem is when the quantum service starts to run prior to qpid. | 09:21 |
| garyk | ijw: i am quite surprised. i understood what you said. | 09:21 |
| *** davidha has joined #openstack-dev | 09:21 | |
| ijw | I know what it means, I've just no idea what it means | 09:22 |
| garyk | ijw: basically it means that we'll get it sorted by hook or by crook... | 09:23 |
| ijw | Ah, fair enough | 09:23 |
| *** davidha has quit IRC | 09:25 | |
| *** pixelbeat has joined #openstack-dev | 09:25 | |
| *** devcamca1 has quit IRC | 09:26 | |
| *** davidha has joined #openstack-dev | 09:27 | |
| *** xchu has joined #openstack-dev | 09:31 | |
| garyk | gongysh: ping | 09:35 |
| *** mmagr has joined #openstack-dev | 09:37 | |
| *** davidha has quit IRC | 09:38 | |
| *** koolhead17 has joined #openstack-dev | 09:40 | |
| openstackgerrit | A change was merged to openstack/quantum: Put gw_port into router dict result. https://review.openstack.org/15101 | 09:40 |
| *** Guest5053 has quit IRC | 09:43 | |
| *** darjeeling has quit IRC | 09:45 | |
| *** henrynash has quit IRC | 09:47 | |
| *** dave_mcn has quit IRC | 09:50 | |
| ijw | garyk: I'm not being thick here - if you have IptablesFirewall set you basically end up with a gateway on every Quantum network, right? | 09:52 |
| ijw | (This is me solving my own problems, nothing to do with your patch now) | 09:52 |
| garyk | ijw: not sure that i follow. | 09:52 |
| *** devcamca1 has joined #openstack-dev | 09:54 | |
| garyk | ijw: my achilles heal at the moment is the security groups and quantum. i think that arosen is a good person to talk to. | 09:54 |
| *** zul has quit IRC | 09:54 | |
| ijw | garyk: Yeah, this is routing rather than secgroups, though it's all a bit intertwingled | 09:55 |
| garyk | ijw: i hear you. one thing that i have learnt over the last few days is that if namespaces is enable/disbaled then the behaviour may also be a little different | 09:56 |
| ijw | namespaces? | 09:56 |
| garyk | ijw: it is something that we certainly need to sort out | 09:56 |
| garyk | by default the l3 agent and dhcp agent use network namespaces to enable overlapping ip's | 09:57 |
| ijw | Yeah, I see | 09:57 |
| garyk | there is an option to disable this (some linux distributions may not support this by default). there are a few problems when it is disbaled | 09:58 |
| *** xchu has quit IRC | 09:58 | |
| *** eglynn has joined #openstack-dev | 09:58 | |
| ijw | garyk: I think my problem is that I tend to only want bits of the L3 features, and sometimes none at all... | 09:59 |
| garyk | ijw: understood. | 10:00 |
| *** shang has joined #openstack-dev | 10:01 | |
| * ijw thinks. | 10:01 | |
| *** e1mer has joined #openstack-dev | 10:02 | |
| *** anniec has joined #openstack-dev | 10:03 | |
| *** eglynn_ has joined #openstack-dev | 10:04 | |
| *** eglynn has quit IRC | 10:04 | |
| *** devcamca1 has quit IRC | 10:05 | |
| *** anniec has quit IRC | 10:05 | |
| *** anniec has joined #openstack-dev | 10:06 | |
| ijw | So at the moment, at L3, we have the gateway for floating IPs/SNAT. The gateway for 169.254.169.254 access. The DHCP server. Anti-spoofing, which needs the VIF's IP address(es). And security groups, which also depend on IP address. I think that's all the features that are currently going. And some of them can't easily be disabled selectively, or independently of each other - they're either on systemwide or entirely off. | 10:07 |
| *** koolhead17 has quit IRC | 10:11 | |
| *** e1mer has quit IRC | 10:12 | |
| *** TerryH has joined #openstack-dev | 10:14 | |
| openstackgerrit | A change was merged to openstack/nova: Fix copy-paste bug in block_device_info_generation https://review.openstack.org/15152 | 10:17 |
| *** Hitesh has joined #openstack-dev | 10:19 | |
| Hitesh | garyk: Hi Gary | 10:19 |
| garyk | hi | 10:20 |
| Hitesh | garyk: Can you redirect me to quantum.conf configuration for Keystone? | 10:20 |
| Hitesh | garyk: I have this http://docs.openstack.org/trunk/openstack-network/admin/content/demo_installions.html | 10:20 |
| Hitesh | garyk: but I think there is no detail information for it | 10:20 |
| garyk | sorry, but i think that is all we have at the moment. maybe there are other refernces in the admin guide. you'll have to check | 10:21 |
| *** anniec has left #openstack-dev | 10:22 | |
| Hitesh | garyk: Ok gary no issue | 10:23 |
| Hitesh | garyk: I'll check itr | 10:23 |
| Hitesh | garyk: Thanks for reply :) | 10:24 |
| garyk | ok, good luck. if it is missing maybe you can add it to the docs. | 10:24 |
| *** koolhead17 has joined #openstack-dev | 10:25 | |
| Hitesh | garyk: Cool | 10:25 |
| *** devcamca1 has joined #openstack-dev | 10:32 | |
| *** arata has joined #openstack-dev | 10:37 | |
| *** devcamca1 has quit IRC | 10:45 | |
| *** henrynash has joined #openstack-dev | 10:47 | |
| *** koolhead17 has quit IRC | 10:54 | |
| mordred | markmc: ping | 10:56 |
| *** wenjianhn has quit IRC | 10:58 | |
| ijw | garyk: I debate your commenting skills, but I'm not going to argue the toss twice... | 11:00 |
| *** alexpilotti has joined #openstack-dev | 11:01 | |
| garyk | ijw: they are terrible. i'll be happy to update with some better comments | 11:01 |
| ijw | Well, if you would, then I think it would help those that follow. | 11:02 |
| garyk | ijw: in the params? or to add explicit text about the filtering? | 11:03 |
| *** novas0x2a|laptop has quit IRC | 11:04 | |
| ijw | The params, really. I kind of guessed that a param called 'gateway' and set to a boolean value probably turned 'gateway' off and on, ditto filtering, but I thought some useful comment about what 'gateway' and 'filter' were might be helpful. (I'm guessing gateway is whether it's run through the gatewaying rules in the kernel for floating-IP and 169.254 access and filtering is obviously whether iptables rules are applied.) | 11:05 |
| ijw | … or it's left to the caller to sort out. | 11:05 |
| garyk | ijw: i'll update. thanks | 11:07 |
| *** eglynn_ has quit IRC | 11:10 | |
| *** roadTripper has joined #openstack-dev | 11:13 | |
| *** devcamca1 has joined #openstack-dev | 11:14 | |
| markmc | mordred, what's up? | 11:24 |
| *** devcamca1 has quit IRC | 11:24 | |
| *** EmilienM has joined #openstack-dev | 11:25 | |
| mordred | markmc: hey! ttx tells me we should chat about oslo | 11:27 |
| *** jeffrey4l has quit IRC | 11:27 | |
| markmc | mordred, yeah, I'm thinking the next step is to rename the repo to oslo-incubator | 11:27 |
| mordred | markmc: ok - so ... renaming repositories will require any dev to re-make their gerrit remotes by hand or to reclone, you ok with that? | 11:28 |
| *** davidha has joined #openstack-dev | 11:28 | |
| markmc | mordred, yep, re-clone should be fine for most people | 11:29 |
| markmc | mordred, relatively small number of developers on openstack-common anyway | 11:29 |
| mordred | markmc: ok. we've got a ceilometer rename coming up soon | 11:29 |
| markmc | mordred, really? to what? | 11:29 |
| mordred | markmc: (rename requires a gerrit restart - don't ask) | 11:29 |
| mordred | markmc: stackforge/ceilometer to openstack/ceilometer | 11:30 |
| *** itooon has quit IRC | 11:30 | |
| markmc | mordred, ah, makes sense | 11:30 |
| mordred | markmc: but if we bundle the renames, it'll be less downtime impact | 11:30 |
| *** itooon has joined #openstack-dev | 11:30 | |
| *** itooon has quit IRC | 11:30 | |
| markmc | mordred, fine by me | 11:30 |
| mordred | markmc: cool | 11:30 |
| mordred | markmc: additionally, I'm guessing we're going to want to start creating oslo lib repos too, yeah? | 11:31 |
| *** maxim has joined #openstack-dev | 11:31 | |
| *** maxim is now known as Guest12051 | 11:32 | |
| markmc | mordred, yep, starting with oslo-config once we get the argparse re-write in | 11:32 |
| markmc | mordred, https://github.com/markmc/oslo-config | 11:32 |
| mordred | markmc: ok - so ... you are now the second person who is going to need creating new projects to be easier | 11:33 |
| mordred | markmc: there are about 5 non-automated steps at the moment, if we can finish automating them, then new project creation could all be a single git commit | 11:33 |
| mordred | markmc: I'm going to try to rope lifeless in on helping with finishing that automation code, but I might want to bug you for a little bit of help on it too | 11:34 |
| mordred | between the three of us, it should be a piece of cake | 11:35 |
| markmc | mordred, ok, cool | 11:36 |
| markmc | mordred, I don't expect a rash of oslo repos tho | 11:36 |
| markmc | mordred, at most one or two more during grizzly | 11:37 |
| mordred | markmc: k. well, then you might have less incentive to help me with that automation :) | 11:37 |
| *** mestery_ has joined #openstack-dev | 11:38 | |
| markmc | mordred, btw, https://review.openstack.org/15191 | 11:38 |
| markmc | mordred, if you +1, I'll just approve it | 11:38 |
| *** sagar_nikam has joined #openstack-dev | 11:38 | |
| *** dims has quit IRC | 11:39 | |
| mordred | markmc: ah - I actually meant to chat with you about that... | 11:39 |
| *** mestery has quit IRC | 11:40 | |
| mordred | markmc: I think it's dangerous to hard-pin with an == when we've been doing a >= up until then | 11:41 |
| markmc | mordred, dangerous in what sense? | 11:42 |
| mordred | markmc: so I think we should _add_ a version ceiling, but there are people (like notmyname) who run things on not-latest distro | 11:42 |
| mordred | (swift still runs on lucid in production at rackspace, for instance) | 11:42 |
| mordred | so hard asserting lower bounds arbitrarily is tricky | 11:43 |
| markmc | mordred, what else uses pip-requires other than virtualenv builders like tox? | 11:43 |
| markmc | mordred, AFAICT, the file is just informational otherwise | 11:44 |
| mordred | markmc: setup.py | 11:44 |
| *** maploin has joined #openstack-dev | 11:44 | |
| *** maploin has quit IRC | 11:44 | |
| *** maploin has joined #openstack-dev | 11:44 | |
| markmc | mordred, hmm, point | 11:44 |
| mordred | markmc: have I mentioned that the python-deps vs. distro deps combined with how pypi works continually makes me want to die | 11:44 |
| markmc | mordred, no, but it's not fun | 11:45 |
| markmc | mordred, a quick reply to the thread on the -dev list would be good | 11:45 |
| markmc | mordred, I'll poke a bit more | 11:45 |
| mordred | markmc: ok. will do (at UDS, so I've been bad about email) - but in _general_ I like the approach | 11:46 |
| *** koolhead17 has joined #openstack-dev | 11:48 | |
| *** tgall_foo has joined #openstack-dev | 11:49 | |
| markmc | mordred, people will run folsom on newer distros as well as older ones | 11:50 |
| markmc | mordred, not sure I like the pinned versions going in setup.py at all | 11:50 |
| markmc | mordred, I'm just trying to control gate instability | 11:50 |
| Daviey | well, lack of pinned versions makes life really hard for older openstack releases | 11:51 |
| *** devcamca1 has joined #openstack-dev | 11:51 | |
| Daviey | where there was never an expectation you'd be using the latest pypi | 11:51 |
| mordred | I thnk pinning the current upper bound at release time gets us a good way further | 11:53 |
| mordred | just adding a ,<=current_version | 11:53 |
| mordred | and then we can manage other version things by hand, because that will protect us from pypi randomness | 11:54 |
| markmc | I think your argument against running on older distros applies equally to running on newer distros | 11:54 |
| *** amotoki has joined #openstack-dev | 11:54 | |
| markmc | an upper bound is bad for newer distros, pinning to a newer version is bad for older distros | 11:54 |
| * markmc needs to think about the setup.py implications, was only thinking about the gating side of things | 11:55 | |
| maploin | sorry to pop in, but are you guys now talking about pip-requires or setup.py? | 11:55 |
| maploin | imho setup.py should never have hardcoded versions, that turns out really badly for distros who can't change the versions of the packages they have installed, but could patch to backport fixes to the packages they have | 11:56 |
| markmc | maploin, never even a >= version ? | 11:56 |
| Daviey | markmc: The fact that we have been unable to land fixes to older openstack releases due to changing deps is a real issue. | 11:56 |
| markmc | Daviey, example? | 11:57 |
| *** littleidea has joined #openstack-dev | 11:57 | |
| mordred | Daviey: right- but if we added an upper version pin and left the lower version open, that wouldn't break things for you, right? | 11:57 |
| mordred | or left the lower version to what it's been during hte dev cycle | 11:57 |
| Daviey | markmc: There were a bunch of them, particularly security ones which impacted diablo. | 11:57 |
| Daviey | the last few commits are examples | 11:58 |
| markmc | Daviey, ok, you're talking about gate instability - yes, that's what I'm trying to fix | 11:58 |
| Daviey | mordred: Ah, yes.. that sounds good | 11:58 |
| Daviey | markmc: right! Having known consitent versions. | 11:58 |
| Daviey | I really think deps should be pinned at every ODS, with a rigour process to adjust them | 11:59 |
| *** armaan has joined #openstack-dev | 11:59 | |
| *** armaan has left #openstack-dev | 11:59 | |
| markmc | Daviey, for the gate, yes - but we're now discussing the implications of pinned versions in setup.py | 11:59 |
| markmc | Daviey, maploin, do either the Ubuntu or SUSE packaging use the versions from setup.py? | 11:59 |
| markmc | AFAIK Fedora ignores them | 12:00 |
| mordred | I don't think that upper bound ranges should break setup.py - I do think that single-version =='s can be problematic | 12:00 |
| Daviey | no idea :( | 12:00 |
| *** roge has joined #openstack-dev | 12:00 | |
| maploin | markmc: even >= can end up badly when abused, say the only thing that you really depend on in the newer version is a rename of a module. I can easily fix that by patching an existing package with a one-line fix rather than updating the whole package to a new API. | 12:00 |
| markmc | mordred, why the difference in implications? | 12:00 |
| markmc | maploin, explain how versions in setup.py affect SUSE packaging | 12:00 |
| mordred | markmc: because the problem in distros is effectively that they cannot bump a version | 12:00 |
| mordred | markmc: but they can patch an existing thing | 12:00 |
| maploin | markmc: AFAIK you can't ignore them, when you try installing the package, python will just break | 12:00 |
| markmc | maploin, ok, thanks | 12:00 |
| mordred | so if you say <=3.0 and suse has 2.9 installed, it will work | 12:01 |
| markmc | mordred, if maploin is correct, the upper bound limit will affect newer distros | 12:01 |
| mordred | but if you say == 3.0, and suse has 2.9 with the appropriate patch backported, it will break | 12:01 |
| markmc | mordred, i.e. won't work if suse has 3.1 installed | 12:01 |
| maploin | it's about running setup.py install | 12:01 |
| *** matiu has quit IRC | 12:01 | |
| mordred | markmc: upper bound limit can raise in next dev release | 12:01 |
| mordred | _noone_ is working on support older openstack on newer distro releases when they come out | 12:02 |
| markmc | mordred, why shouldn't you be able to build folsom packages on SUSE just because it has 3.1 ? | 12:02 |
| mordred | the things people are doing are concerning themselves with latest openstack on latest disto, latest openstack on older distro, and on maintaining older openstack on the distro that was available at release time | 12:02 |
| *** danpb has quit IRC | 12:02 | |
| mordred | as a dev community, we cannot expect to forward port diablo to quantal | 12:03 |
| markmc | mordred, current stable version of OpenStack on latest distro too | 12:03 |
| mordred | but we can help make folsom available on precise | 12:03 |
| mordred | yes | 12:03 |
| mordred | but none of those are broken by upper-boundaries | 12:03 |
| markmc | mordred, the upper bound is an arbitrary restriction on where you can run the stable version | 12:03 |
| markmc | mordred, latest distro pulls in newer versions of libraries all the time | 12:03 |
| mordred | again | 12:03 |
| markmc | mordred, because they're compatible | 12:03 |
| *** devcamca1 has quit IRC | 12:04 | |
| mordred | ok - so that's also solvable | 12:04 |
| maploin | mordred: the dev community isn't expected to forward port anything, but it should allow the distros to easily do that if they need to | 12:04 |
| mordred | _if_ current stable is release is causing problem for a dev release of a distro | 12:05 |
| mordred | and they want to pull in a newer version | 12:05 |
| mordred | and it does not break the stable branch of openstack | 12:05 |
| mordred | then they can submit a patch to bump the version in stable/* just fine | 12:05 |
| mordred | these aren't immutable | 12:05 |
| markmc | nah, that's backwards - library releases should be assumed to be compatible | 12:05 |
| mordred | they are just a record of what we know to be true about the world at release time | 12:05 |
| mordred | no | 12:05 |
| mordred | that's insane | 12:05 |
| markmc | sqlalchemy is an outlier | 12:06 |
| mordred | have you ever pulled code from pypi that's compatible? | 12:06 |
| mordred | no | 12:06 |
| mordred | it's not | 12:06 |
| mordred | it's the norm | 12:06 |
| mordred | all of the libraries on pypi are run by crazy hipsters | 12:06 |
| mordred | we get broken by this ALL T?HE TIME | 12:06 |
| markmc | we push newer versions of libraries to stable Fedora all the time | 12:06 |
| mordred | those are all manual pushes | 12:06 |
| markmc | because a lot of python libraries are compatible | 12:06 |
| mordred | yes | 12:06 |
| mordred | they are | 12:06 |
| mordred | they just can't be _assumed_ to be | 12:06 |
| markmc | ok, look | 12:06 |
| markmc | we have two separate things going on here | 12:06 |
| markmc | the versions in setup.py which affects distros etc. | 12:07 |
| openstackgerrit | A change was merged to openstack/openstack-common: Don't log exceptions for GreenletExit and thread_done https://review.openstack.org/15159 | 12:07 |
| markmc | and keeping the gate running on an unchanging set of libraries | 12:07 |
| mordred | please tell me a specific example of how setting an upper boundary on our code will actually break a distro | 12:07 |
| markmc | the latter is what I want to fix in the short term | 12:07 |
| markmc | if we have e.g. Paste<1.7.5.2 | 12:08 |
| markmc | 1.7.5.2 is released with a bug fix | 12:08 |
| markmc | and is perfectly compatible | 12:08 |
| markmc | and pushed to Fedora | 12:08 |
| markmc | you can't build the folsom nova package on Fedora anymore | 12:08 |
| mordred | then someone from fedora proposes a patch to nova with Paste<1.7.5.3 | 12:09 |
| mordred | and that works, because it's actually valid | 12:09 |
| markmc | that's backwards | 12:09 |
| mordred | it's not | 12:09 |
| markmc | it is | 12:09 |
| mordred | no | 12:09 |
| markmc | again, this is a separate issue from keeping the gate stable | 12:09 |
| mordred | 1.7.5.2 has been testied | 12:09 |
| mordred | it's not | 12:09 |
| mordred | it's the same thing | 12:09 |
| mordred | we can, as a community, assert positively that we work with 1.7.5.2 because we have actually tested | 12:09 |
| mordred | it | 12:09 |
| mordred | we cannot say the same thing about 1.7.5.3 | 12:10 |
| maploin | mordred: right, but what are the benefits of asserting anything in setup.py? vs. pip-requires | 12:10 |
| markmc | library releases should be compatible | 12:10 |
| markmc | that's the norm we want to encourage | 12:10 |
| mordred | markmc: but they are not | 12:10 |
| mordred | but we have no control over tit | 12:10 |
| mordred | it | 12:10 |
| markmc | I think you're overstating how bad the python world is on this | 12:10 |
| mordred | the reason we want to assert this | 12:10 |
| mordred | I am not | 12:10 |
| markmc | sqlalchemy is fucked, but most others aren't | 12:10 |
| markmc | sure we don't have control over other projects | 12:11 |
| mordred | I am the one woh has to deal with this every time it breaks | 12:11 |
| mordred | so I promise you | 12:11 |
| mordred | it is worse than you think it is | 12:11 |
| mordred | the thing is | 12:11 |
| mordred | when someone pulls our code and runs setup.py and that pulls in dependencies | 12:11 |
| *** zehndton has joined #openstack-dev | 12:11 | |
| mordred | if that pulls in a non-working versoin of a depend because we have loose version requires | 12:11 |
| markmc | see, I have no problem with us running the stable branch tests on a fixed set of versions | 12:12 |
| *** boden has joined #openstack-dev | 12:12 | |
| *** eglynn_ has joined #openstack-dev | 12:12 | |
| markmc | I have a real problem with us telling users "don't run the stable branch on Paste 1.7.5.2!" | 12:12 |
| *** koolhead17 has quit IRC | 12:12 | |
| markmc | they are different concerns | 12:12 |
| mordred | why? | 12:12 |
| mordred | I see no reason why we should have a problem with telling users of our STABLE release | 12:12 |
| markmc | because we're assuming library releases are incompatible and they shouldn't be and often aren't | 12:12 |
| mordred | "hey, pull this version - it WORKS" | 12:12 |
| mordred | you keep saying shouldn't | 12:12 |
| mordred | which I agree with | 12:13 |
| markmc | users don't have a choice of what library they run on | 12:13 |
| mordred | they SO DO | 12:13 |
| mordred | I think you're mixing use cases | 12:13 |
| markmc | it's the latest version available in their distro | 12:13 |
| mordred | nope | 12:13 |
| mordred | you're wrong | 12:13 |
| mordred | if they are installing from PACKAGES | 12:13 |
| mordred | it's the latest distro version | 12:13 |
| mordred | if they are instlling from setup.py - it's pip | 12:13 |
| markmc | or building packages | 12:13 |
| mordred | if they are building packages, it's a totally different thing | 12:13 |
| markmc | no it's not | 12:13 |
| mordred | and that affects about 12 people | 12:13 |
| markmc | you're rrong | 12:13 |
| markmc | you're wrong | 12:14 |
| mordred | ok. | 12:14 |
| mordred | we might need to take this off of IRC | 12:14 |
| mordred | and into something more long form | 12:14 |
| markmc | yeah | 12:14 |
| *** zing has quit IRC | 12:14 | |
| mordred | because I believe we have an intractable set of assumptions about the world | 12:14 |
| markmc | but ... | 12:14 |
| markmc | we have a short term issue | 12:14 |
| mordred | we do | 12:14 |
| markmc | keeping the gate stable | 12:14 |
| markmc | can we fix that without getting into "what versions we put into setup.py" ? | 12:14 |
| mordred | can we add <= to the pip-requires for now? | 12:14 |
| *** TerryH has quit IRC | 12:14 | |
| markmc | not if it affects setup.py | 12:15 |
| mordred | ok - can I suggest something? | 12:15 |
| * markmc listening | 12:15 | |
| garyk | *x | 12:15 |
| mordred | packagers who run in to this as a propblem can toss in a 2 line patch to avoid having setup.py pull in pip-requires | 12:15 |
| *** zing has joined #openstack-dev | 12:15 | |
| mordred | or, even if they don't run in to this problem | 12:15 |
| mordred | they can put that patch into the spec or the debian/patches | 12:15 |
| *** littleidea has quit IRC | 12:15 | |
| mordred | and then our upper bounds to keep the stable branch building doesn't actually break them | 12:16 |
| mordred | then - longer term, you and I can argue until we've come to an agreement about the world which works for both ofus | 12:16 |
| markmc | that they can workaround the fact we fucked up the versions in setup.py doesn't make me happy | 12:16 |
| markmc | can we do something about the gate without fucking up setup.py? | 12:16 |
| mordred | I understand - but there is no solution short term that will make both of us happy | 12:17 |
| mordred | no | 12:17 |
| mordred | we cannot | 12:17 |
| markmc | a different *-requires file for the gate? | 12:17 |
| mordred | by your definition of fucking up | 12:17 |
| mordred | no | 12:17 |
| markmc | that tox uses? | 12:17 |
| mordred | no | 12:17 |
| markmc | why not, seems straightforward? | 12:17 |
| markmc | tools/frozen-pip-requires | 12:17 |
| mordred | it seems exactly the opposite of straight forward | 12:17 |
| *** markvoelker has joined #openstack-dev | 12:17 | |
| mordred | and it immediately wins the argument on your side of hte issue, and I do not concede your argument | 12:18 |
| mordred | I STRONGLY believe that setup.py should reflect reality | 12:18 |
| mordred | I recognize that you disagree | 12:18 |
| mordred | I think a quick change in either of our favor is unfair until we've resolved this | 12:18 |
| *** littleidea has joined #openstack-dev | 12:18 | |
| markmc | we need to discuss at length what to do about setup.py | 12:18 |
| mordred | yes | 12:18 |
| mordred | we do | 12:18 |
| markmc | we all agree what to do about the gate | 12:18 |
| markmc | I'm suggesting how to move forward on the latter while we discuss the former | 12:19 |
| *** alexpilotti has quit IRC | 12:19 | |
| mordred | forgive me if this sounds snarky - it's not intended to - but I am kind of concerned that if we do that the urge to solve this properly will go away | 12:19 |
| mordred | because it will not be a problem anymore, and when the conversation gets hard I think people will give u | 12:20 |
| mordred | up | 12:20 |
| markmc | that's not my intent - to kill the discussion about setup.py | 12:20 |
| zykes- | mordred: : ) | 12:20 |
| *** littleidea has quit IRC | 12:20 | |
| markmc | but the urgent problem in people's eyes is gate instability, sure | 12:20 |
| zykes- | mordred: can you give me the runthrough of libra ? :) | 12:20 |
| mordred | zykes-: no. not right now | 12:20 |
| zykes- | ok :) | 12:20 |
| mordred | markmc: as long as you will continue to have this discussion, I'll assent to the frozen-pip-requires | 12:21 |
| markmc | mordred, the setup.py discussion will come up again as part of the dependency management stuff too | 12:21 |
| mordred | it just needs to be understood that I thnk it's a MAJOR problem | 12:21 |
| markmc | mordred, e.g. if we're doing the upper bound thing, whether we should do it on master too | 12:21 |
| *** darjeeling has joined #openstack-dev | 12:22 | |
| mordred | and that I am not happy about it as a solution | 12:22 |
| *** tmishler has joined #openstack-dev | 12:22 | |
| markmc | mordred, you're happy with pinning the gate to specific versions, but not happy the versions we have in install_requires ... understood | 12:24 |
| mordred | markmc: yeah. I think it's _ESSENTIAL_ that we actually assert versions in setup.py | 12:24 |
| mordred | because sqlalchemy and kombu and prettytable have and do break us on incompatible version changes | 12:24 |
| *** mmagr has quit IRC | 12:24 | |
| markmc | mordred, ok, well let's continue that discussion on the list | 12:24 |
| *** timello has joined #openstack-dev | 12:27 | |
| *** alunduil has quit IRC | 12:28 | |
| *** mrodden has quit IRC | 12:28 | |
| mordred | markmc: I believe your frozen idea may need more engineering effort that we may be missing... | 12:29 |
| markmc | mordred, like? | 12:29 |
| mordred | markmc: so even to do it, we need to walk through some code paths | 12:29 |
| markmc | mordred, stop being so vague :) | 12:29 |
| *** dolphm has joined #openstack-dev | 12:30 | |
| mordred | markmc: for instance - the pypi mirror does not know to look for tools/frozen-pip-requires in bulding the list of versions it needs to mirror | 12:30 |
| *** eharney has joined #openstack-dev | 12:30 | |
| markmc | mordred, ok - got a pointer to that code? | 12:30 |
| mordred | markmc: additionally, setup.py, which gets run by tox, is still tied to pip-requires, so we'll need to verify that pip-requires isn't going to then supercede what tox installed at venv creation time | 12:30 |
| markmc | mordred, ok | 12:31 |
| mordred | markmc: also, we'll be providing broken tarballs to people | 12:31 |
| markmc | mordred, broken tarballs? | 12:31 |
| mordred | markmc: yup. anybody who pulls the tarball of the stable branch and attempts to use it | 12:31 |
| mordred | will get what setup.py pulls | 12:31 |
| mordred | but the version freeze will be in frozen-pip-requires, not pip-requires | 12:32 |
| *** devcamca1 has joined #openstack-dev | 12:32 | |
| mordred | so they'll get sqlalchemy 0.8 | 12:32 |
| mordred | and then it won't work | 12:32 |
| markmc | no, we already have the upper bound for sqlalchemy | 12:32 |
| markmc | because we know that doesn't work | 12:32 |
| mordred | ok. I thnk what I meant was ... | 12:32 |
| mordred | we _might_ be shipping broken tarballs | 12:32 |
| mordred | or tarballs that will break later | 12:32 |
| mordred | in the same way that this just broke - so we will still need to watch the situation and patch the actual pip-requires | 12:33 |
| mordred | and I don't know how to catch that error in the future | 12:33 |
| *** krtaylor has quit IRC | 12:33 | |
| markmc | yeah, fair | 12:33 |
| mordred | if we've patched the gate | 12:33 |
| markmc | we'd probably hit issues on master and just have to remember that it applies to stable too | 12:33 |
| markmc | but the flipside is, we're explicitly saying we don't want to hit these issues on stable :) | 12:34 |
| mordred | so - you know about the nightly stable gate check jobs, right? | 12:34 |
| markmc | yep | 12:34 |
| mordred | those catch these errors | 12:34 |
| mordred | so far, no one has every noticed them or cared | 12:34 |
| *** chrisfer has quit IRC | 12:35 | |
| markmc | yeah, I asked jim to start emailing stable-maint | 12:35 |
| markmc | with those errors | 12:35 |
| mordred | so in terms of systemically preventing the gate from breaking ... | 12:35 |
| *** chrisfer has joined #openstack-dev | 12:35 | |
| mordred | perhaps we should just get the stable-maint emails going and see if that helps | 12:35 |
| mordred | rather than doing the frozen file quickly | 12:35 |
| mordred | because at the moment, there is actually only one known breakage | 12:35 |
| markmc | ok, yeah - I think that makes sense | 12:35 |
| mordred | and I think we're both worried about collateral damage (different sets of it) | 12:35 |
| mordred | of solutions | 12:36 |
| mordred | whee! | 12:36 |
| mordred | markmc: cool. I think that's a great next step -and give us the space to have further fun conversations :) | 12:37 |
| mordred | and I'm sitting next to jeblair, who said he'll get the emails going | 12:37 |
| markmc | mordred, funny that we already agreed to do that at the summit :) | 12:37 |
| mordred | markmc: yeah, he looked sheepish when I brought it up | 12:38 |
| markmc | mordred, nah, it's not that it's not done is funny - just that we forgot that it was the solution to this we'd already agreed on | 12:38 |
| mordred | markmc: :) | 12:41 |
| *** bing_bu has quit IRC | 12:41 | |
| *** dabo has joined #openstack-dev | 12:41 | |
| markmc | mordred, sent a summary to the list | 12:41 |
| *** maurosr has joined #openstack-dev | 12:42 | |
| *** devcamca1 has quit IRC | 12:43 | |
| zykes- | annegentle: png | 12:43 |
| *** dolphm has quit IRC | 12:45 | |
| *** adalbas has joined #openstack-dev | 12:45 | |
| *** mrodden has joined #openstack-dev | 12:46 | |
| *** tgall_foo has quit IRC | 12:47 | |
| mordred | markmc: in terms of notification ... since there isn't actually a stable-maint list at the moment... | 12:49 |
| mordred | markmc: would a build failure tweet to an #openstack-stable-maint work? or do you want a mailing list and emails? | 12:49 |
| *** itooon has joined #openstack-dev | 12:50 | |
| markmc | mordred, heh, I've enough problems in life without *having* to follow stuff on twitter | 12:50 |
| markmc | mordred, any form of email suits me fine | 12:50 |
| markmc | mordred, if we have to set up a mailing list, I'm ok with that | 12:50 |
| mordred | markmc: ok. | 12:50 |
| jeblair | markmc: openstack-stable-maint@lists.openstack.org ? | 12:51 |
| mordred | markmc: and here I thought I was the only hacker around here who didn't live on twitter all the time | 12:51 |
| markmc | jeblair, sounds good | 12:51 |
| jeblair | mordred: hardly :) | 12:51 |
| *** shang has quit IRC | 12:53 | |
| *** maurosr has quit IRC | 12:54 | |
| *** maurosr has joined #openstack-dev | 12:54 | |
| *** kbringard has joined #openstack-dev | 12:55 | |
| *** pcrews has joined #openstack-dev | 12:59 | |
| *** ck has joined #openstack-dev | 13:01 | |
| *** eglynn_ has quit IRC | 13:01 | |
| *** sandywalsh has joined #openstack-dev | 13:04 | |
| Kiall | sandywalsh, ping | 13:04 |
| sandywalsh | Kiall, pong | 13:05 |
| Kiall | Heya - Was looking at your StachTach project yesterday, and ran into some issues | 13:05 |
| Kiall | (Trying to get the details back up!) | 13:06 |
| sandywalsh | cool | 13:06 |
| Kiall | Everything appeared to work - events would show in the list, but clicking on anything would produce an error | 13:06 |
| sandywalsh | hmm, ok, do you have the output from django perhaps? | 13:06 |
| *** jergerber has joined #openstack-dev | 13:07 | |
| Kiall | http://paste.openstack.org/show/23450/ | 13:07 |
| zykes- | using stacky for Kiall ? ;) | 13:08 |
| Kiall | I've never used Django, so not sure how else to debug it ;) | 13:08 |
| *** ayoung has joined #openstack-dev | 13:09 | |
| sandywalsh | Kiall, ah, yup, that's a bug. Try selecting the "All" deployment from the main page instead ... I'll work on a patch | 13:09 |
| Kiall | no clue why I didn't think of that already -_- | 13:09 |
| zykes- | does it work on others then nova as well sandywalsh ? | 13:10 |
| Kiall | worked a charm :) | 13:10 |
| *** koolhead17 has joined #openstack-dev | 13:10 | |
| sandywalsh | Kiall, excellent! | 13:10 |
| Kiall | zykes-, 2 reasons.. 1 Always nice to have this kind of data available... | 13:10 |
| sandywalsh | zykes-, it should accept notifications from any of the enabled systems, but I haven't looked at their payloads to ensure they conform with Nova's | 13:10 |
| Kiall | and 2) Peeking inside the notification payloads | 13:11 |
| sandywalsh | Kiall, I added the "watch" command to Stacky yesterday, you should try it ... it's very cool | 13:11 |
| *** strife25 has joined #openstack-dev | 13:11 | |
| Kiall | Thats the CLI component? Havent looked at it yet! | 13:11 |
| zykes- | ooh :) | 13:11 |
| zykes- | hmm | 13:12 |
| zykes- | how can I make a hook to update docs on ReadTheDocs upon a merge in gerrit mordred ? | 13:12 |
| Kiall | sagar_nikam, A config var for durable=False would be nice BTW ;) One of my environments is not using durable queues for some unknown reason -_- | 13:12 |
| Kiall | gah | 13:12 |
| Kiall | sorry sagar_nikam, I mean sandywalsh | 13:12 |
| *** devcamca1 has joined #openstack-dev | 13:12 | |
| sandywalsh | Kiall, ah, interesting ... I'll see what I can do there. Can you file an issue on github please? | 13:12 |
| sandywalsh | Kiall, I'm glad to hear you got it working so quickly. | 13:13 |
| Kiall | 15 mins, other than that bug ;) | 13:14 |
| sagar_nikam | Kiall: no problem | 13:14 |
| sandywalsh | excellent | 13:14 |
| Kiall | Most of that was waiting on pip too | 13:14 |
| sandywalsh | yeah, no surprise there :) | 13:14 |
| *** adalbas has quit IRC | 13:15 | |
| Kiall | issue #1 filed ;) | 13:15 |
| *** krtaylor has joined #openstack-dev | 13:16 | |
| *** bmeshjason has joined #openstack-dev | 13:16 | |
| sandywalsh | thanks! | 13:16 |
| *** mmagr has joined #openstack-dev | 13:17 | |
| *** arata has quit IRC | 13:17 | |
| maurosr | annegentle: just FYI, I didn't forget about the api site integration with the api samples, just had a tough day yesterday, but I'm writing that email about the idea with submodules to the dev list right now | 13:22 |
| *** shang has joined #openstack-dev | 13:22 | |
| *** devcamca1 has quit IRC | 13:23 | |
| *** rkukura has joined #openstack-dev | 13:24 | |
| *** rnirmal has joined #openstack-dev | 13:27 | |
| zykes- | annegentle: or anyone that knows about docs: can we autogenerate these ?https://review.openstack.org/#/c/15194/1/doc/source/configuration.rst | 13:30 |
| zykes- | see the comment from Kiall | 13:30 |
| mordred | zykes-: adding readthedocs support for our project has been on our todo list | 13:31 |
| *** adalbas has joined #openstack-dev | 13:31 | |
| zykes- | how you mean by that mordred ? | 13:32 |
| *** littleidea has joined #openstack-dev | 13:34 | |
| *** eharney has quit IRC | 13:36 | |
| Kiall | zykes-, looks like I was wrong about nova generating them .. Can't find it! | 13:37 |
| *** eglynn_ has joined #openstack-dev | 13:38 | |
| *** ewindisch has joined #openstack-dev | 13:43 | |
| *** dprince has joined #openstack-dev | 13:43 | |
| *** alunduil has joined #openstack-dev | 13:44 | |
| *** aeperezt has joined #openstack-dev | 13:44 | |
| *** alunduil has quit IRC | 13:45 | |
| *** tgall_foo has joined #openstack-dev | 13:45 | |
| *** tgall_foo has quit IRC | 13:45 | |
| *** tgall_foo has joined #openstack-dev | 13:45 | |
| *** alunduil has joined #openstack-dev | 13:45 | |
| *** mtreinish has joined #openstack-dev | 13:47 | |
| *** tmishler__ has joined #openstack-dev | 13:48 | |
| *** cloudvirt has joined #openstack-dev | 13:48 | |
| *** tmishler__ has quit IRC | 13:48 | |
| *** cloudvirt has left #openstack-dev | 13:48 | |
| *** devcamca1 has joined #openstack-dev | 13:50 | |
| *** tmishler has quit IRC | 13:51 | |
| *** zodiak has joined #openstack-dev | 13:51 | |
| *** henrynash has quit IRC | 13:53 | |
| *** henrynash has joined #openstack-dev | 13:55 | |
| ndipanov | pixelbeat, hey - mind taking a look at https://review.openstack.org/#/c/15125/ and re-bless :) | 13:57 |
| *** shang has quit IRC | 13:58 | |
| *** tmishler has joined #openstack-dev | 13:58 | |
| pixelbeat | ndipanov, I the bug qpidd specific? | 13:59 |
| ndipanov | probably not - but I only tested it with qpid... | 14:00 |
| ndipanov | highly likely sending an sqlalchemy model instance to rabbit will confuse it too | 14:01 |
| ndipanov | so I put qpid in the bug title to be specific | 14:01 |
| *** vipuls has joined #openstack-dev | 14:02 | |
| pixelbeat | I'd edit the bug title so, lest rabbit users think they don't need to worry about it. | 14:03 |
| *** devcamca1 has quit IRC | 14:03 | |
| ndipanov | let me try it with rabbit then just in case | 14:03 |
| *** eharney has joined #openstack-dev | 14:03 | |
| mikal | eglynn_: you awake / around? | 14:04 |
| *** radez_g0n3 is now known as radez | 14:06 | |
| *** andrewbogott_afk is now known as andrewbogott | 14:07 | |
| jeblair | markmc: https://review.openstack.org/15193 | 14:08 |
| markmc | jeblair, yep, saw it - didn't realize you wanted my +1 | 14:09 |
| markmc | jeblair, oh, that's notifications for the gate jobs too? | 14:10 |
| markmc | jeblair, I presumed just bitrot jobs? | 14:10 |
| markmc | jeblair, i.e. we only want email when the bitrot jobs fail, otherwise we'll learn to ignore the mails | 14:10 |
| *** markmcclain has joined #openstack-dev | 14:11 | |
| mordred | markmc: he walked away | 14:14 |
| markmc | mordred, k, I commented in the review | 14:15 |
| mordred | markmc: apparently there are monster cars next door or something | 14:15 |
| *** shang has joined #openstack-dev | 14:17 | |
| *** ck has quit IRC | 14:18 | |
| *** radez is now known as radez_g0n3 | 14:19 | |
| eglynn_ | mikal: on a call there, missed your ping | 14:19 |
| eglynn_ | mikal: 'sup? | 14:19 |
| *** derekh has joined #openstack-dev | 14:20 | |
| *** jimfehlig has joined #openstack-dev | 14:22 | |
| mikal | eglynn_: I wanted to trick you into reviewing | 14:24 |
| eglynn_ | mikal: np, what's the review? | 14:24 |
| mikal | https://review.openstack.org/#/c/14924/ | 14:24 |
| * eglynn_ looking ... | 14:24 | |
| mikal | (Sorry, I hit return too early) | 14:24 |
| *** salv-orlando has joined #openstack-dev | 14:25 | |
| *** annegentle__ has joined #openstack-dev | 14:27 | |
| *** Hitesh has quit IRC | 14:28 | |
| jeblair | markmc: that _should_ just be the bitrot jobs (whose names are actually "periodic-{name of normal gate job}" | 14:28 |
| *** sandywalsh has quit IRC | 14:28 | |
| markmc | jeblair, cool, sorry about that | 14:29 |
| jeblair | mordred: and there's totally a bugatti veyron over there. :) | 14:30 |
| *** shakhat has joined #openstack-dev | 14:30 | |
| *** devcamca1 has joined #openstack-dev | 14:31 | |
| mikal | jeblair: oh, you gate crashed? | 14:31 |
| *** danpb has joined #openstack-dev | 14:32 | |
| *** samkottler|afk is now known as samkottler | 14:32 | |
| jeblair | mikal: yeah, and you can get to the sky bridge from the far side of the second floor. | 14:32 |
| *** dolphm has joined #openstack-dev | 14:32 | |
| *** redbeard2 has joined #openstack-dev | 14:33 | |
| *** radez_g0n3 is now known as radez | 14:34 | |
| *** jaypipes has quit IRC | 14:38 | |
| *** alexpilotti has joined #openstack-dev | 14:38 | |
| eglynn_ | mikal: one minor suggestion on #14924 to consider ... | 14:40 |
| *** egallen has joined #openstack-dev | 14:42 | |
| *** radez is now known as radez_g0n3 | 14:42 | |
| *** devcamca1 has quit IRC | 14:42 | |
| mikal | eglynn_: ta | 14:43 |
| mikal | jeblair: sky bridge? | 14:43 |
| *** sandywalsh has joined #openstack-dev | 14:44 | |
| *** dhellmann has joined #openstack-dev | 14:45 | |
| *** Mandell has joined #openstack-dev | 14:45 | |
| *** reed has joined #openstack-dev | 14:47 | |
| markmc | jeblair, should all stable-maint members be subscribed to openstack-stable-maint? | 14:49 |
| markmc | jeblair, e.g. https://launchpad.net/~openstack-stable-maint shows me subscribed | 14:49 |
| markmc | jeblair, but a mail I just sent got moderated | 14:49 |
| *** spiffxp has joined #openstack-dev | 14:51 | |
| *** shang has quit IRC | 14:53 | |
| *** otherwiseguy has joined #openstack-dev | 14:56 | |
| *** Mandell has quit IRC | 14:56 | |
| *** samkottler is now known as samkottler|bbl | 14:57 | |
| *** anniec has joined #openstack-dev | 14:57 | |
| jeblair | markmc: i have not subscribed anyone; i did set you as the owner, so i think once you subscribe you may be able to batch subscribe/invite others. | 14:58 |
| markmc | jeblair, ok, thanks | 14:58 |
| *** sandywalsh has quit IRC | 14:59 | |
| *** eglynn_ is now known as eglynn | 15:02 | |
| *** Mandell has joined #openstack-dev | 15:02 | |
| *** Mandell has quit IRC | 15:03 | |
| ewindisch | ndipanov: this should really affect all rpc libraries. We only support primitives. The zeromq driver explicitly sends all input (and responses) into json.dumps :-/ | 15:03 |
| ndipanov | ewindisch, my thoughts exactly... but didn't really check... I will update the bug (if I can) | 15:05 |
| *** cp16net is now known as cp16net|away | 15:05 | |
| *** cp16net|away is now known as cp16net | 15:05 | |
| openstackgerrit | A change was merged to openstack/quantum: Use openstack common log to do logging. https://review.openstack.org/15034 | 15:05 |
| *** datsun180b has joined #openstack-dev | 15:06 | |
| ndipanov | ewindisch, did you try to delete an instance with a volume attached (or started with one) with the current trunk usin zmq by any chance? | 15:06 |
| ewindisch | ndipanov: no, but if it passes an object, it won't work. | 15:07 |
| ndipanov | ewindisch, it does | 15:07 |
| *** jaypipes has joined #openstack-dev | 15:08 | |
| ewindisch | then it won't work. There was a point where it would fallback to Pickle if someone passed an object, but consensus was that if someone did it, it was a bug, and we'd fix it. | 15:08 |
| ndipanov | ewindisch, I was thinking that some drivers might pickle stuff | 15:09 |
| russellb | it'll break qpid, too. rabbitmq will work because kombu anyjson internally, and we set that up to automatically use to_primitive() as needed | 15:09 |
| ndipanov | that's why I stuck to qpid | 15:09 |
| *** devcamca1 has joined #openstack-dev | 15:09 | |
| russellb | if we all used anyjson, it would always work | 15:10 |
| russellb | but i don't really like that because we may end up doing conversions when we don't realize it or intend to | 15:10 |
| russellb | and FWIW, impl_qpid doesn't use json at all right now | 15:10 |
| russellb | it serializes using AMQP types | 15:11 |
| *** mmagr has quit IRC | 15:11 | |
| russellb | not because that's necessarily any better ... just letting the qpid lib serialize how it feels like, which is what we do with kombu (and internally it uses json) | 15:12 |
| *** sandywalsh has joined #openstack-dev | 15:13 | |
| *** ondergetekende has quit IRC | 15:15 | |
| *** dolphm has quit IRC | 15:16 | |
| *** dolphm has joined #openstack-dev | 15:16 | |
| *** mjfork has joined #openstack-dev | 15:16 | |
| *** dolphm has quit IRC | 15:18 | |
| *** samkottler|bbl is now known as samkottler | 15:18 | |
| ndipanov | russellb, so which approach is better in your opinion then - let the lib deal with it, serialize it in the *_imp or say I need json/dict/whatever and serialize when calling rpc? | 15:20 |
| *** rkukura has quit IRC | 15:20 | |
| russellb | ndipanov: the patch like you did it is best for now IMO | 15:20 |
| ndipanov | russellb, since this was introduced when moving db access out of compute... might be worth mentioning it in the etherpad as a gotcha | 15:21 |
| ndipanov | I can see it happening again if developing on rabbit | 15:21 |
| *** devcamca1 has quit IRC | 15:21 | |
| ndipanov | for example | 15:21 |
| russellb | yep, my bad for missing it in the review | 15:22 |
| *** koolhead17 has quit IRC | 15:23 | |
| *** rkukura has joined #openstack-dev | 15:25 | |
| *** troytoman-away is now known as troytoman | 15:30 | |
| *** imsplitbit has joined #openstack-dev | 15:34 | |
| *** shang has joined #openstack-dev | 15:36 | |
| dansmith | so, | 15:38 |
| dansmith | it's really _my_ bad, of course | 15:38 |
| dansmith | however, I think that was the first of those that I had done, | 15:38 |
| dansmith | and IIRC, while I was trying to figure out whether I needed it, I tested it on devstack | 15:39 |
| dansmith | but that would have been rabbit, so I guess that's why it worked and I convinced myself not to do it... | 15:39 |
| *** mdomsch has joined #openstack-dev | 15:41 | |
| ndipanov | dansmith, the issue is that we currently have no *working* tests that can catch this | 15:42 |
| openstackgerrit | A change was merged to openstack/nova: Make bdms primitive in rpcapi.terminate_instance https://review.openstack.org/15125 | 15:42 |
| *** colinmcnamara has joined #openstack-dev | 15:42 | |
| dansmith | ndipanov: well, that's a nice excuse for me, but still.. :D | 15:42 |
| ndipanov | dansmith, see my email on the dev list from yesterday - it has an explanation of how it didn;t get caugh... | 15:42 |
| *** dolphm has joined #openstack-dev | 15:46 | |
| *** hemna has joined #openstack-dev | 15:47 | |
| *** devcamca1 has joined #openstack-dev | 15:49 | |
| *** hub_cap has joined #openstack-dev | 15:50 | |
| *** markwash has quit IRC | 15:52 | |
| *** dolphm has quit IRC | 15:55 | |
| *** zehndton has quit IRC | 15:57 | |
| *** koolhead17 has joined #openstack-dev | 15:57 | |
| *** heckj has joined #openstack-dev | 15:58 | |
| *** dolphm has joined #openstack-dev | 15:58 | |
| *** dhellmann has quit IRC | 16:00 | |
| Kiall | sandywalsh, did I see some mention in the ceilo meeting there about you having some code/a BP for a os.common notification consumer? (I missed most of the convo ;)) | 16:00 |
| ayoung | heckj, care to take a look at https://review.openstack.org/#/c/15114/ | 16:01 |
| ayoung | needed for https://review.openstack.org/#/c/15116/ | 16:01 |
| heckj | ayoung: yep - doing a few reviews right now. I'll toss it on the list | 16:01 |
| ayoung | heckj, thanks | 16:01 |
| heckj | Odd - I was looking for those last night before I left, but didn't spot them. | 16:02 |
| ayoung | heckj, need me to look at anything specific? Thought I cleared my queue, but I find I often overlook things | 16:02 |
| *** devcamca1 has quit IRC | 16:02 | |
| sandywalsh | Kiall, I'm working on a proposal for merging all of these competing efforts to a common set of libraries | 16:02 |
| heckj | ayoung: I don't think so - at the moment anyway. | 16:02 |
| *** troytoman is now known as troytoman-away | 16:02 | |
| heckj | ayoung: just approved henrynash for the CLA, FYI | 16:02 |
| sandywalsh | Kiall, common in the low-level stuff and common in the high-level stuff | 16:02 |
| heckj | ayoung: I'll pester you later - did have some questions about optimal way to determine if a PKI token was expired/not in keystoneclient per your note | 16:03 |
| *** markmcclain has quit IRC | 16:04 | |
| ayoung | heckj, good news on henrynash | 16:05 |
| Kiall | sandywalsh, great, because I'm in need of a notification consumer ;) | 16:05 |
| sandywalsh | heh, aren't we all :) | 16:06 |
| openstackgerrit | A change was merged to openstack/quantum: support 'send_arp_for_ha' option in l3_agent https://review.openstack.org/12037 | 16:07 |
| dansmith | .names | 16:09 |
| dansmith | oops | 16:09 |
| ndipanov | dansmith, just don't type passwords here and ur ok :) | 16:10 |
| dansmith | heh | 16:10 |
| *** vipuls has quit IRC | 16:11 | |
| dansmith | pixelbeat: can you re-approve this one: https://review.openstack.org/#/c/14977/ | 16:12 |
| dansmith | pixelbeat: had to rebase and gerrit didn't re-add your approval +2, only the review one | 16:12 |
| *** jcmartin has joined #openstack-dev | 16:12 | |
| pixelbeat | looking | 16:12 |
| pixelbeat | dansmith, done | 16:14 |
| dansmith | pixelbeat: thanks! | 16:15 |
| *** jtran has joined #openstack-dev | 16:16 | |
| Kiall | sandywalsh, I assume you have done a review of the current implementations? Is there one in particular that you think might be standardized on? Hopefully save me at least some work :) | 16:17 |
| Kiall | (Sorry for the delays.. all over the place today!) | 16:19 |
| *** aeperezt has quit IRC | 16:19 | |
| sandywalsh | Kiall, yeah, I'm doing a summary in the report. The YAGI one is solid, the StackTach one should be, but has gone through a lot of changes and the Ceilometer one is too big (imho) | 16:20 |
| Kiall | Heh - I was liking the Ceilometer implementation :) | 16:20 |
| heckj | ayoung: ping | 16:20 |
| ayoung | heckj, i'm here. I'm, like, always here. | 16:20 |
| heckj | ayoung: heh | 16:20 |
| heckj | ayoung: re: https://review.openstack.org/#/c/15116/3/keystone/middleware/auth_token.py | 16:20 |
| sandywalsh | Kiall, perhaps I'm missing something there, looking forward to the feedback on it | 16:21 |
| *** danwent has joined #openstack-dev | 16:21 | |
| heckj | ayoung: is the full PKI token getting returned at all now, or are we just floating around the hash of that token as an ID? | 16:21 |
| Kiall | sandywalsh, I'll probably not notice the BP when you publish it, if you remember, ping me :) | 16:21 |
| ayoung | heckj, still full PKI. But the hash will work in its place | 16:22 |
| ayoung | hash is being used as the key. | 16:22 |
| ayoung | heckj, so if the webui passes the hash back and forth to the web browser, they can use the hash to fetch the PKI token out of memcached | 16:23 |
| ayoung | heckj, getting pulled away to a team lunch, back in a bout an hour | 16:23 |
| *** aeperezt has joined #openstack-dev | 16:23 | |
| *** ayoung is now known as ayoung-afk | 16:23 | |
| heckj | ayoung: got it - OK - had to read back to cms_hash_token to get it, but makes sense - normalizes the lookup to the hash if it's the full PKI token | 16:23 |
| heckj | ayoung-afk: cool, thank | 16:23 |
| *** mnewby has joined #openstack-dev | 16:24 | |
| openstackgerrit | A change was merged to openstack/keystone: don't modify the passed in dict to from_dict https://review.openstack.org/14472 | 16:27 |
| *** anniec has quit IRC | 16:27 | |
| openstackgerrit | A change was merged to openstack/keystone: don't import filter_user name, use it from the identity module https://review.openstack.org/15190 | 16:28 |
| *** devcamca1 has joined #openstack-dev | 16:28 | |
| *** lloydde has joined #openstack-dev | 16:29 | |
| *** jcmartin has quit IRC | 16:33 | |
| *** dubsquared has joined #openstack-dev | 16:34 | |
| *** chrisfer has quit IRC | 16:36 | |
| *** sagar_nikam has quit IRC | 16:39 | |
| *** ck has joined #openstack-dev | 16:40 | |
| *** SlickNik has joined #openstack-dev | 16:41 | |
| *** SlickNik has joined #openstack-dev | 16:42 | |
| *** devcamca1 has quit IRC | 16:42 | |
| *** Mandell has joined #openstack-dev | 16:43 | |
| *** andrewbogott is now known as andrewbogott_afk | 16:43 | |
| *** thingee has joined #openstack-dev | 16:44 | |
| *** egallen has left #openstack-dev | 16:45 | |
| *** winston-d has joined #openstack-dev | 16:46 | |
| *** reidrac has left #openstack-dev | 16:47 | |
| *** egallen has joined #openstack-dev | 16:47 | |
| *** egallen has left #openstack-dev | 16:47 | |
| *** egallen has joined #openstack-dev | 16:48 | |
| *** egallen has left #openstack-dev | 16:49 | |
| *** spn has quit IRC | 16:54 | |
| *** spn has joined #openstack-dev | 16:55 | |
| *** mindpixel has quit IRC | 16:56 | |
| winston-d | russellb, around? | 16:58 |
| *** anniec has joined #openstack-dev | 16:58 | |
| *** anniec has quit IRC | 16:58 | |
| *** anniec has joined #openstack-dev | 16:58 | |
| *** jog0 has joined #openstack-dev | 16:58 | |
| *** anniec_ has joined #openstack-dev | 16:59 | |
| *** gongysh has quit IRC | 17:00 | |
| *** dhellmann has joined #openstack-dev | 17:01 | |
| *** mdomsch has quit IRC | 17:01 | |
| openstackgerrit | A change was merged to openstack/cinder: Return volume type name on volume create https://review.openstack.org/15145 | 17:02 |
| *** imsplitbit has quit IRC | 17:02 | |
| *** andrewbogott_afk is now known as andrewbogott | 17:02 | |
| *** anniec has quit IRC | 17:03 | |
| *** anniec_ is now known as anniec | 17:03 | |
| *** markmcclain has joined #openstack-dev | 17:04 | |
| *** devcamca1 has joined #openstack-dev | 17:08 | |
| *** afazekas has quit IRC | 17:13 | |
| *** tmishler has quit IRC | 17:14 | |
| *** metral has quit IRC | 17:15 | |
| *** metral has joined #openstack-dev | 17:15 | |
| *** cp16net is now known as cp16net|away | 17:18 | |
| *** Hitesh_ has joined #openstack-dev | 17:18 | |
| Hitesh_ | Hello, | 17:18 |
| Hitesh_ | any idea why this has been occuring, error: Installed distribution python-keystoneclient 2012.1 conflicts with requirement python-keystoneclient>=0.1.2,<0.2 | 17:19 |
| Hitesh_ | Processing dependencies for horizon==2013.1 error: Installed distribution python-keystoneclient 2012.1 conflicts with requirement python-keystoneclient>=0.1.2,<0.2 | 17:19 |
| *** dolphm has quit IRC | 17:20 | |
| *** devcamca1 has quit IRC | 17:21 | |
| vishy | danpb: ping | 17:21 |
| vishy | danwent: ping | 17:21 |
| danwent | vishy: hey | 17:21 |
| * vishy sends simultaneous pings | 17:21 | |
| danpb | vishy: you called ? | 17:21 |
| vishy | danwent: about the esx driver | 17:22 |
| vishy | danwent: can I get a blueprint for it? | 17:22 |
| vishy | danpb: I actually have two things | 17:22 |
| maploin | Hitesh_: because 2012 > 0.2? | 17:22 |
| vishy | danpb: I think i mentioned to you before about having to specify specific flags to get kvm in kvm | 17:23 |
| danwent | vishy: yes, will ping the dev. | 17:23 |
| vishy | danpb: this is the kind of thing we have to add to the libvirt.xml | 17:23 |
| vishy | danpb: http://paste.openstack.org/show/23506/ | 17:23 |
| danpb | vishy: yeah, I never really got a satisfactory conclusion on that | 17:23 |
| *** dolphm has joined #openstack-dev | 17:24 | |
| danpb | its kind of a known bug in libvirt, and i don't think its possible to work around it easily | 17:24 |
| vishy | danpb: still not really sure why we can't just use host_model | 17:24 |
| Hitesh_ | maploin: are you sure? | 17:24 |
| Hitesh_ | maploin: whats the solution? | 17:24 |
| Hitesh_ | maploin: do I need to update any? | 17:25 |
| danpb | vishy: yeah you ought to be able todo, but libvirt gets upset in the guest due to kvm masking out certain cpu flags | 17:25 |
| *** vipuls has joined #openstack-dev | 17:25 | |
| danpb | i reluctantly think the best thing todo is disable CPU model specification if running kvm-inside-kvm | 17:25 |
| *** mdomsch has joined #openstack-dev | 17:25 | |
| *** corXi has quit IRC | 17:26 | |
| vishy | danpb: ok that is good to know. Do you think we should add a new type to libvirt_cpu_mode ? | 17:26 |
| vishy | danpb: how does one disable it? We've only been able to get it to work by explicitly adding those flags. | 17:26 |
| danpb | you can set libvirt_cpu_mode="none" IIRC | 17:26 |
| danpb | and we should detect when we're running nested and make sure to default to that | 17:27 |
| vishy | danpb: that is the default, i think, but that doesn't seem to be sufficient | 17:27 |
| openstackgerrit | A change was merged to openstack/nova: Move to a more canonicalized output from qemu-img info. https://review.openstack.org/13166 | 17:27 |
| vishy | danpb: we have to explicitly specify those on the host machine or vt is not enabled in the guest | 17:27 |
| danpb | one sec, while i check the code | 17:28 |
| *** tmishler has joined #openstack-dev | 17:30 | |
| danpb | vishy: currently we do | 17:30 |
| danpb | mode = FLAGS.libvirt_cpu_mode | 17:31 |
| danpb | if mode is None: | 17:31 |
| danpb | if FLAGS.libvirt_type == "kvm" or FLAGS.libvirt_type == "qemu": | 17:31 |
| danpb | mode = "host-model" | 17:31 |
| danpb | else: | 17:31 |
| danpb | mode = "none" | 17:31 |
| danpb | we need to change that to | 17:31 |
| danpb | if (FLAGS.libvirt_type == "kvm" or FLAGS.libvirt_type == "qemu") and not running_inside_kvm(): | 17:31 |
| danpb | ...where "running_inside_kvm()" is a function someone clever needs to write :-) | 17:31 |
| *** tmishler has quit IRC | 17:32 | |
| *** itooon has quit IRC | 17:32 | |
| *** tmishler has joined #openstack-dev | 17:33 | |
| *** ayoung-afk is now known as ayoung | 17:34 | |
| *** dachary has quit IRC | 17:36 | |
| *** melwitt has joined #openstack-dev | 17:37 | |
| winston-d | russellb, could you review https://review.openstack.org/#/c/14119/7 and https://review.openstack.org/#/c/14131/? | 17:38 |
| ayoung | heckj, can you give https://review.openstack.org/#/c/14823/ a review, and I will deal with the test failure? I don't think it is anything other than an intermittant right now | 17:39 |
| vishy | danpb: i guess I'm confused | 17:42 |
| vishy | danpb: we have to set the flags in the libvirt.xml on the host | 17:42 |
| *** vipuls has quit IRC | 17:43 | |
| vishy | danpb: imagine nested scenario host > guest > nested_guest | 17:43 |
| danpb | the 1st level guest should work fine | 17:43 |
| vishy | when we write the libvirt.xml for the guest (which is running on the host) | 17:43 |
| danpb | its only when you're running nova in the guest that you need to disable the cpu model | 17:43 |
| danpb | so that the 2nd level guest does not have cpu model set | 17:44 |
| vishy | danpb: that's what i'm saying unless we put those flags in the libvirt.xml on the host | 17:44 |
| vishy | the guest does not have vt extensions or /dev/kvm | 17:44 |
| *** cp16net|away is now known as cp16net | 17:44 | |
| danpb | hmm, that should be happening automatically | 17:44 |
| vishy | danpb: using host-model doesn't work | 17:44 |
| danpb | ah, ok, so that's a 2nd bug | 17:45 |
| vishy | danpb: that is what i thought. but I couldn't get it to work without explicitly putting in the flags | 17:45 |
| *** eglynn has quit IRC | 17:45 | |
| danpb | i'll have to take alook at that then | 17:45 |
| *** adjohn has joined #openstack-dev | 17:45 | |
| jaypipes | bcwaldon: any reason why I would be running into https://bugs.launchpad.net/glance/+bug/1054183 on a fresh checkout of devstack? :( | 17:47 |
| uvirtbot | Launchpad bug 1054183 in glance "durability of rabbit exchange/queue should be configurable" [High,Fix released] | 17:47 |
| *** devcamca1 has joined #openstack-dev | 17:47 | |
| *** maploin has quit IRC | 17:48 | |
| boden | ayoung -- you around? | 17:48 |
| ayoung | boden, yep | 17:50 |
| ayoung | trying to determine why REMOTE_USER patch failed unit tests | 17:50 |
| boden | ayoung -- anything you want me to do to keep https://review.openstack.org/#/c/14823/ moving? | 17:50 |
| ayoung | boden, I think it is a spurious problem. | 17:50 |
| ayoung | boden, nah, need heckj or dolphm to look at it to approve. Other than feedback from Joe, it should be good to go | 17:51 |
| *** zbitter has joined #openstack-dev | 17:51 | |
| boden | ayoung -- ok thanks... let me know if you need anything | 17:52 |
| bcwaldon | jaypipes: no idear | 17:52 |
| *** Ryan_Lane has joined #openstack-dev | 17:53 | |
| *** Tross1 has joined #openstack-dev | 17:53 | |
| *** shang has quit IRC | 17:53 | |
| *** oneiroi has quit IRC | 17:53 | |
| *** vipuls has joined #openstack-dev | 17:54 | |
| *** zaneb has quit IRC | 17:55 | |
| *** heckj has quit IRC | 17:56 | |
| *** garyk has quit IRC | 17:56 | |
| *** Tross has quit IRC | 17:56 | |
| *** heckj has joined #openstack-dev | 17:56 | |
| *** Tross has joined #openstack-dev | 17:56 | |
| *** oneiroi has joined #openstack-dev | 17:56 | |
| *** winston-d has quit IRC | 17:58 | |
| *** Hitesh_ has quit IRC | 17:58 | |
| *** Tross1 has quit IRC | 17:59 | |
| *** devcamca1 has quit IRC | 18:00 | |
| *** zing has quit IRC | 18:01 | |
| *** shang has joined #openstack-dev | 18:02 | |
| *** dims has joined #openstack-dev | 18:03 | |
| heckj | boden: looking at in a few min | 18:05 |
| jaypipes | bcwaldon: please see last comment on https://bugs.launchpad.net/glance/+bug/1054183. This is a serious issue... causes glance API log to grow at about 20M every 5 seconds. :( | 18:06 |
| uvirtbot | Launchpad bug 1054183 in glance "durability of rabbit exchange/queue should be configurable" [High,Fix released] | 18:06 |
| *** aeperezt has quit IRC | 18:06 | |
| *** danpb has quit IRC | 18:06 | |
| *** dims has quit IRC | 18:08 | |
| *** dims has joined #openstack-dev | 18:09 | |
| *** aeperezt has joined #openstack-dev | 18:09 | |
| bcwaldon | jaypipes: ok, I'll look into it | 18:09 |
| bcwaldon | jaypipes: are you using master or stable/folsom? | 18:11 |
| jaypipes | bcwaldon: master.. | 18:12 |
| *** shang has quit IRC | 18:12 | |
| *** armaan has joined #openstack-dev | 18:12 | |
| *** gatuus has joined #openstack-dev | 18:13 | |
| *** aeperezt has quit IRC | 18:14 | |
| bcwaldon | jaypipes: what OS? | 18:16 |
| *** dolphm has quit IRC | 18:17 | |
| bcwaldon | jaypipes: asking because a completely fresh devstack on a vanilla precise box seems unaffected | 18:17 |
| jaypipes | bcwaldon: xubuntu 12.04/kvm/libvirt/rabbitmq | 18:17 |
| *** Tross1 has joined #openstack-dev | 18:17 | |
| *** aeperezt has joined #openstack-dev | 18:18 | |
| *** reed has quit IRC | 18:19 | |
| *** Tross has quit IRC | 18:20 | |
| *** PaulM has joined #openstack-dev | 18:22 | |
| *** garyk has joined #openstack-dev | 18:23 | |
| *** derekh has quit IRC | 18:23 | |
| ayoung | heckj, dolphm_zzz can we please put https://review.openstack.org/#/c/14823/ out of my misery? | 18:24 |
| *** devcamca1 has joined #openstack-dev | 18:27 | |
| sdague | davidkranz: for qa blueprints do you want to be using https://blueprints.launchpad.net/openstack-qa or https://blueprints.launchpad.net/tempest ? | 18:27 |
| PaulM | ayoung: heh. Sorry for making that worse for you | 18:27 |
| sdague | https://blueprints.launchpad.net/openstack-qa seems in some ways more appropriate, but it also seems to have a lot of old stuff in it | 18:27 |
| *** aeperezt has quit IRC | 18:28 | |
| *** rkukura has quit IRC | 18:29 | |
| *** rkukura has joined #openstack-dev | 18:30 | |
| *** aeperezt has joined #openstack-dev | 18:31 | |
| *** reed has joined #openstack-dev | 18:31 | |
| *** dachary has joined #openstack-dev | 18:32 | |
| *** dolphm has joined #openstack-dev | 18:32 | |
| ayoung | PaulM, make what worse? | 18:36 |
| *** novas0x2a|laptop has joined #openstack-dev | 18:36 | |
| PaulM | ayoung: oh, it had a bunch of +1s and looked ready to go, then I picked another security nit at it | 18:36 |
| ayoung | PaulM, yes, but I asked for that. That makes it better, not worse | 18:37 |
| PaulM | true :) | 18:37 |
| *** darraghb has quit IRC | 18:38 | |
| *** dachary has quit IRC | 18:39 | |
| *** dolphm_ has joined #openstack-dev | 18:39 | |
| *** dolphm has quit IRC | 18:39 | |
| *** devcamca1 has quit IRC | 18:39 | |
| *** joearnold has quit IRC | 18:39 | |
| ayoung | PaulM, though going back through and putting a +1 next to it now would not be remiss.... | 18:40 |
| PaulM | oh, I hadn't hit "submit" yet on that. Oops. Done. | 18:40 |
| *** Ryan_Lane has quit IRC | 18:41 | |
| *** johnpostlethwait has joined #openstack-dev | 18:42 | |
| *** eharney has quit IRC | 18:42 | |
| ayoung | PaulM, heckj and with that, I think I can claim it is well enough reviewed to merge. | 18:42 |
| PaulM | excellent | 18:42 |
| *** joearnold has joined #openstack-dev | 18:42 | |
| heckj | ayoung, PaulM - looking now | 18:42 |
| *** cloudvirt has joined #openstack-dev | 18:42 | |
| *** cloudvirt has left #openstack-dev | 18:43 | |
| heckj | erp, you already approved. Cool | 18:43 |
| *** gyee has joined #openstack-dev | 18:44 | |
| ayoung | heckj, yeah, it had been addressed enough times, but still better if you know what we are doing here | 18:44 |
| ayoung | and it was really PaulM 's review that I was waiting on, since that was the security review this patch warranted | 18:44 |
| *** eharney has joined #openstack-dev | 18:45 | |
| *** eharney has quit IRC | 18:45 | |
| *** eharney has joined #openstack-dev | 18:45 | |
| zykes- | markmc: how can i update my openstack-common ? :) | 18:45 |
| *** cp16net is now known as cp16net|away | 18:45 | |
| jk0 | zykes-: http://lmgtfy.com/?q=update+openstack-common | 18:47 |
| *** bmeshjason has quit IRC | 18:48 | |
| ayoung | heckj, who needs to approve https://review.openstack.org/#/c/14858/ for stable? | 18:50 |
| *** imsplitbit has joined #openstack-dev | 18:51 | |
| *** Ryan_Lane has joined #openstack-dev | 18:52 | |
| zykes- | funny jk0 :p | 18:52 |
| *** aeperezt has quit IRC | 18:53 | |
| heckj | ayoung: I was expecting either you or dolph to hit approve since I backported. I can approve now - there's enough review on it | 18:53 |
| ayoung | heckj, I don't have approval on stable. | 18:54 |
| ayoung | BTW, is it OK if I approve https://review.openstack.org/#/c/15114/ now? | 18:54 |
| dolphm_ | heckj: i can't approve stable | 18:54 |
| heckj | really? hrm - ok- not sure who that set is then. Stable-maint I guess. I'll keep closer eye | 18:54 |
| *** ewindisch_ has joined #openstack-dev | 18:54 | |
| heckj | ayoung: yep - good for me on 15114! | 18:55 |
| ayoung | dolphm_, heckj, should I wait on one more core on https://review.openstack.org/#/c/15116/ | 18:56 |
| ayoung | or are we saying written by core +1 from community +one core is sufficient? | 18:56 |
| heckj | ayoung: yeah - dolph, when you have a sec, please review: https://review.openstack.org/#/c/15116/ | 18:56 |
| dolphm_ | heckj: which review is that? i'm on the slowest wifi in the world | 18:57 |
| *** ewindisch has quit IRC | 18:57 | |
| *** ewindisch_ is now known as ewindisch | 18:57 | |
| heckj | I'd like to keep core +2 or core+1, community+2 - we're still building the community, but it's closer | 18:57 |
| heckj | dolphm_: airplane? :-) | 18:57 |
| dolphm_ | heckj: rackspace | 18:57 |
| ayoung | heckj, dolphm_ OK. I'm headed to the home office. ping me then or email me if there is something I can do to keep things moving along | 18:57 |
| *** cloudvirt has joined #openstack-dev | 18:59 | |
| openstackgerrit | A change was merged to openstack/python-cinderclient: Remove attach/detach code from cinderclient https://review.openstack.org/15042 | 18:59 |
| *** ewindisch has quit IRC | 18:59 | |
| *** cloudvirt has left #openstack-dev | 18:59 | |
| *** ewindisch has joined #openstack-dev | 19:00 | |
| *** jog0 has quit IRC | 19:00 | |
| garyk | markmcclain: you there? | 19:01 |
| *** littleidea has quit IRC | 19:01 | |
| *** jgriffith has quit IRC | 19:01 | |
| *** Aaton is now known as entropyworks | 19:02 | |
| *** ayoung has quit IRC | 19:02 | |
| *** jgriffith has joined #openstack-dev | 19:03 | |
| markmcclain | garyk: yes.. what's up? | 19:03 |
| garyk | going over your comments. not sure what to do for the hostname. i'll try update the regex. | 19:04 |
| openstackgerrit | A change was merged to openstack/python-cinderclient: Remove extra-specs from types-list command output https://review.openstack.org/15162 | 19:04 |
| *** devcamca1 has joined #openstack-dev | 19:06 | |
| markmcclain | ok.. yeah that one is much tougher because I'm not sure we can fully validate other than it might be plausible | 19:07 |
| *** thovden_ has joined #openstack-dev | 19:09 | |
| garyk | markmcclain: originally i was thinking of just validating a string but was under the impression that that is just pointless | 19:10 |
| *** littleidea has joined #openstack-dev | 19:13 | |
| *** littleidea has joined #openstack-dev | 19:14 | |
| *** mdomsch has quit IRC | 19:14 | |
| *** littleidea has quit IRC | 19:14 | |
| *** littleidea has joined #openstack-dev | 19:15 | |
| *** larsks is now known as larsks|away | 19:16 | |
| *** thovden_ is now known as thovden | 19:18 | |
| *** devcamca1 has quit IRC | 19:19 | |
| *** edmund has joined #openstack-dev | 19:20 | |
| *** littleidea has joined #openstack-dev | 19:23 | |
| *** littleidea has quit IRC | 19:24 | |
| *** maurosr has quit IRC | 19:26 | |
| *** maurosr has joined #openstack-dev | 19:27 | |
| *** patelna has joined #openstack-dev | 19:28 | |
| *** epim has joined #openstack-dev | 19:28 | |
| *** epim_ has joined #openstack-dev | 19:29 | |
| *** japage has quit IRC | 19:30 | |
| *** dolphm_ has quit IRC | 19:32 | |
| *** epim has quit IRC | 19:33 | |
| *** epim_ is now known as epim | 19:33 | |
| *** Ryan_Lane has quit IRC | 19:33 | |
| *** cp16net|away is now known as cp16net | 19:33 | |
| *** thovden has quit IRC | 19:36 | |
| *** colinmcnamara has left #openstack-dev | 19:36 | |
| *** nunosantos has joined #openstack-dev | 19:36 | |
| *** samkottler is now known as samkottler|lunch | 19:38 | |
| *** zaitcev has joined #openstack-dev | 19:38 | |
| *** ayoung has joined #openstack-dev | 19:40 | |
| devananda | vishy: your review 15215, is there currently a problem in n-cpu where it takes >1 iteration of update_service_capabilities (and therefor > 1 minute) before the scheduler becomes aware of changes to n-cpu's capabilities? | 19:44 |
| *** devcamca1 has joined #openstack-dev | 19:44 | |
| *** lifeless has quit IRC | 19:48 | |
| *** dubsquared has quit IRC | 19:49 | |
| *** garyk has quit IRC | 19:51 | |
| *** alexpilotti has quit IRC | 19:52 | |
| *** henrynash_ has joined #openstack-dev | 19:55 | |
| *** adalbas has quit IRC | 19:56 | |
| *** devcamca1 has quit IRC | 19:57 | |
| *** henrynash has quit IRC | 19:57 | |
| *** henrynash_ is now known as henrynash | 19:57 | |
| *** lifeless has joined #openstack-dev | 20:01 | |
| openstackgerrit | A change was merged to openstack/keystone: ignore .tox directory for pep8 in runtests https://review.openstack.org/15163 | 20:02 |
| openstackgerrit | A change was merged to openstack/keystone: key all backends off of hash of pki token. https://review.openstack.org/15114 | 20:02 |
| *** eglynn has joined #openstack-dev | 20:03 | |
| vishy | devananda: I don't believe so | 20:03 |
| devananda | vishy: k. i believe it's just an issue with baremetal patch, but review 15215 looks pretty similar to what i've been debugging... | 20:04 |
| *** datsun180b has quit IRC | 20:05 | |
| vishy | devananda: someone put in a patch to force get of capabilities on start | 20:06 |
| *** jog0 has joined #openstack-dev | 20:09 | |
| *** hub_cap has quit IRC | 20:09 | |
| *** tmishler has quit IRC | 20:09 | |
| devananda | vishy: yes, that was related to baremetal | 20:09 |
| *** jog0 has quit IRC | 20:09 | |
| *** jog0 has joined #openstack-dev | 20:09 | |
| *** maurosr has quit IRC | 20:10 | |
| devananda | vishy: the issue i'm seeing is that, when capabilities change (eg, when i inform n-cpu baremetal hv of a new hardware node) that it takes ~2 minutes before scheduler knows about it. | 20:10 |
| vishy | devananda: that is strange | 20:10 |
| openstackgerrit | A change was merged to openstack/keystone: Implements REMOTE_USER authentication support. https://review.openstack.org/14823 | 20:10 |
| *** matiu has joined #openstack-dev | 20:10 | |
| *** matiu has quit IRC | 20:10 | |
| *** matiu has joined #openstack-dev | 20:10 | |
| vishy | devananda: it should get it in the next state report which should only be 60 seconds | 20:11 |
| *** tmishler has joined #openstack-dev | 20:11 | |
| *** thovden has joined #openstack-dev | 20:11 | |
| devananda | first cycle of update_service_capabilities will populate nova.compute_nodes table | 20:11 |
| devananda | right | 20:11 |
| devananda | however that doesn't include the node id / nodename | 20:11 |
| lifeless | vishy: 60 seconds is a long time to wait :) | 20:11 |
| devananda | so if I call "nova boot" at that point, the scheduler will pass a request to n-cpu without the "node" parameter, and n-cpu will bail with an error | 20:11 |
| *** tmishler has quit IRC | 20:12 | |
| vishy | devananda: oh that is definitely bm specific | 20:12 |
| vishy | since we don't have a node param yet :) | 20:12 |
| devananda | yep, that's what i thought :) | 20:12 |
| *** thovden has quit IRC | 20:12 | |
| devananda | still, it'd be great if the new capabilities got pushed up immediately, instead of waiting <= 60s for the next update | 20:13 |
| *** dubsquared has joined #openstack-dev | 20:14 | |
| vishy | devananda: i'm sure that could be added | 20:15 |
| devananda | i haven't grok'd the scheduler/compute interworkings yet enough to know where to put that. any tips? :) | 20:15 |
| *** samkottler|lunch is now known as samkottler | 20:15 | |
| *** dubsquared has quit IRC | 20:15 | |
| *** dubsquared has joined #openstack-dev | 20:16 | |
| *** eglynn has quit IRC | 20:16 | |
| *** dubsquared has quit IRC | 20:19 | |
| *** dubsquared1 has joined #openstack-dev | 20:19 | |
| *** johnpostlethwait has quit IRC | 20:20 | |
| *** devcamca1 has joined #openstack-dev | 20:24 | |
| *** thovden has joined #openstack-dev | 20:27 | |
| *** thovden has quit IRC | 20:27 | |
| *** torandu has quit IRC | 20:27 | |
| *** nati_ueno has joined #openstack-dev | 20:28 | |
| *** jcmartin has joined #openstack-dev | 20:30 | |
| *** dprince has quit IRC | 20:30 | |
| *** torandu has joined #openstack-dev | 20:31 | |
| *** nati_uen_ has joined #openstack-dev | 20:31 | |
| openstackgerrit | A change was merged to openstack/openstack-common: Move utils.execute to its own module. https://review.openstack.org/15197 | 20:32 |
| *** annegentle__ has quit IRC | 20:34 | |
| *** hub_cap has joined #openstack-dev | 20:34 | |
| *** nati_ueno has quit IRC | 20:34 | |
| *** anniec has quit IRC | 20:35 | |
| *** jcmartin has quit IRC | 20:35 | |
| *** jcmartin has joined #openstack-dev | 20:36 | |
| *** devcamca1 has quit IRC | 20:36 | |
| *** littleidea has joined #openstack-dev | 20:38 | |
| *** pixelbeat has quit IRC | 20:38 | |
| mnewby | quantum devs: *ping* | 20:39 |
| *** dolphm has joined #openstack-dev | 20:40 | |
| *** littleidea has left #openstack-dev | 20:40 | |
| *** mdomsch has joined #openstack-dev | 20:45 | |
| *** thovden has joined #openstack-dev | 20:45 | |
| *** Ryan_Lane has joined #openstack-dev | 20:47 | |
| *** thovden has quit IRC | 20:47 | |
| *** jcru has joined #openstack-dev | 20:48 | |
| *** eharney has quit IRC | 20:49 | |
| *** mrodden1 has joined #openstack-dev | 20:50 | |
| *** dolphm has quit IRC | 20:51 | |
| *** mrodden has quit IRC | 20:51 | |
| *** otherwiseguy has quit IRC | 20:55 | |
| *** thovden has joined #openstack-dev | 20:56 | |
| *** dripton__ is now known as dripton | 20:59 | |
| *** dubsquared1 has quit IRC | 21:01 | |
| *** dubsquared has joined #openstack-dev | 21:03 | |
| *** devcamca1 has joined #openstack-dev | 21:03 | |
| *** dims has quit IRC | 21:05 | |
| *** jakedahn_zz is now known as jakedahn | 21:06 | |
| *** thovden has joined #openstack-dev | 21:06 | |
| *** boden has quit IRC | 21:06 | |
| *** thovden has quit IRC | 21:06 | |
| *** mrunge has quit IRC | 21:08 | |
| openstackgerrit | A change was merged to openstack/nova: Fix backwards compat of rpc to compute manager https://review.openstack.org/15217 | 21:08 |
| *** strife25 has quit IRC | 21:08 | |
| *** thovden has joined #openstack-dev | 21:08 | |
| *** datsun180b has joined #openstack-dev | 21:09 | |
| *** markvoelker has quit IRC | 21:10 | |
| *** hub_cap has quit IRC | 21:11 | |
| *** thovden has quit IRC | 21:13 | |
| *** anniec has joined #openstack-dev | 21:14 | |
| *** devcamca1 has quit IRC | 21:15 | |
| *** anniec_ has joined #openstack-dev | 21:15 | |
| *** anniec_ has quit IRC | 21:17 | |
| *** anniec_ has joined #openstack-dev | 21:17 | |
| *** mdomsch has quit IRC | 21:17 | |
| *** zing has joined #openstack-dev | 21:18 | |
| *** anniec has quit IRC | 21:19 | |
| *** anniec_ is now known as anniec | 21:19 | |
| *** tgall_foo has quit IRC | 21:20 | |
| *** mrodden has joined #openstack-dev | 21:20 | |
| *** dhellmann has quit IRC | 21:22 | |
| *** armaan has left #openstack-dev | 21:22 | |
| *** mrodden1 has quit IRC | 21:23 | |
| *** thovden has joined #openstack-dev | 21:25 | |
| *** kgriffs has quit IRC | 21:30 | |
| *** thovden has quit IRC | 21:33 | |
| *** mrodden has quit IRC | 21:34 | |
| *** jcru has quit IRC | 21:34 | |
| *** mrodden has joined #openstack-dev | 21:34 | |
| *** dubsquared has quit IRC | 21:36 | |
| *** maurosr has joined #openstack-dev | 21:37 | |
| *** spiffxp has quit IRC | 21:39 | |
| *** mrodden has quit IRC | 21:40 | |
| *** mrodden has joined #openstack-dev | 21:40 | |
| *** maurosr has quit IRC | 21:41 | |
| *** otherwiseguy has joined #openstack-dev | 21:41 | |
| *** devcamca1 has joined #openstack-dev | 21:42 | |
| *** maurosr has joined #openstack-dev | 21:42 | |
| openstackgerrit | A change was merged to openstack/openstack-common: kombu's fanout_cast_to_server was calling wrong method https://review.openstack.org/15222 | 21:42 |
| *** spiffxp has joined #openstack-dev | 21:42 | |
| *** dubsquared has joined #openstack-dev | 21:44 | |
| *** dubsquared has quit IRC | 21:44 | |
| *** dubsquared has joined #openstack-dev | 21:44 | |
| *** devananda has quit IRC | 21:46 | |
| *** strife25 has joined #openstack-dev | 21:46 | |
| *** thovden has joined #openstack-dev | 21:46 | |
| *** maurosr has quit IRC | 21:47 | |
| *** eglynn has joined #openstack-dev | 21:49 | |
| *** jaypipes has quit IRC | 21:49 | |
| *** tgall_foo has joined #openstack-dev | 21:51 | |
| *** tgall_foo has quit IRC | 21:51 | |
| *** tgall_foo has joined #openstack-dev | 21:51 | |
| *** alobbs has quit IRC | 21:52 | |
| *** devananda has joined #openstack-dev | 21:52 | |
| *** devcamca1 has quit IRC | 21:53 | |
| *** srn has joined #openstack-dev | 21:54 | |
| *** jergerber has quit IRC | 21:54 | |
| *** datsun180b has quit IRC | 21:55 | |
| *** jergerber has joined #openstack-dev | 21:57 | |
| vishy | ayoung: ping | 22:01 |
| *** jergerber has quit IRC | 22:01 | |
| *** alunduil has quit IRC | 22:03 | |
| *** eglynn has quit IRC | 22:05 | |
| *** markmc has quit IRC | 22:06 | |
| *** mtreinish has quit IRC | 22:07 | |
| *** kbringard has quit IRC | 22:08 | |
| *** tgall_foo has quit IRC | 22:08 | |
| *** hattwick has quit IRC | 22:10 | |
| *** colinmcnamara has joined #openstack-dev | 22:10 | |
| *** redbeard2 has quit IRC | 22:11 | |
| *** hattwick has joined #openstack-dev | 22:14 | |
| *** devananda has quit IRC | 22:15 | |
| *** devcamca1 has joined #openstack-dev | 22:20 | |
| *** strife25 has quit IRC | 22:20 | |
| *** entropyworks is now known as Aaton | 22:22 | |
| *** tmishler has joined #openstack-dev | 22:23 | |
| *** spiffxp has quit IRC | 22:27 | |
| *** jcmartin has quit IRC | 22:28 | |
| *** jcmartin has joined #openstack-dev | 22:28 | |
| *** tmishler has quit IRC | 22:29 | |
| asalkeld | russellb, you about? | 22:29 |
| asalkeld | just wondering if you managed to get qpid.messing (client) logging working | 22:30 |
| asalkeld | we did in heat but are not full using common logging | 22:30 |
| *** thovden has quit IRC | 22:30 | |
| *** devcamca1 has quit IRC | 22:31 | |
| heckj | ayoung: just assigned https://bugs.launchpad.net/keystone/+bug/1074172 to you for a quick vet - I think vishy's suggested patch will do the trick, noodling on how to test | 22:32 |
| uvirtbot | Launchpad bug 1074172 in keystone "PKI tokens are broken after 24 hours" [High,Triaged] | 22:32 |
| *** dachary has joined #openstack-dev | 22:35 | |
| *** sandywalsh has quit IRC | 22:36 | |
| *** armaan has joined #openstack-dev | 22:39 | |
| *** eglynn has joined #openstack-dev | 22:41 | |
| *** roge has quit IRC | 22:42 | |
| *** imsplitbit has quit IRC | 22:42 | |
| *** rpedde is now known as rpedde_away | 22:43 | |
| *** Tross1 has quit IRC | 22:43 | |
| *** markmcclain has quit IRC | 22:44 | |
| *** ewindisch has quit IRC | 22:45 | |
| *** armaan has left #openstack-dev | 22:45 | |
| *** eglynn has quit IRC | 22:48 | |
| notmyname | mordred: ping | 22:49 |
| *** almaisan-away is now known as al-maisan | 22:52 | |
| *** dubsquared has quit IRC | 22:52 | |
| *** ck has quit IRC | 22:55 | |
| *** koolhead17 has quit IRC | 22:56 | |
| *** gatuus has quit IRC | 22:56 | |
| *** devcamca1 has joined #openstack-dev | 22:57 | |
| *** dims has joined #openstack-dev | 23:02 | |
| *** Guest12051 has quit IRC | 23:08 | |
| *** devcamca1 has quit IRC | 23:09 | |
| *** al-maisan is now known as almaisan-away | 23:10 | |
| *** devananda has joined #openstack-dev | 23:10 | |
| *** dhellmann has joined #openstack-dev | 23:12 | |
| *** sandywalsh has joined #openstack-dev | 23:13 | |
| *** KurtMartin has joined #openstack-dev | 23:13 | |
| *** KurtMartin has quit IRC | 23:14 | |
| clarkb | FYI the huge influx of changes submitted to gerrit recently has slowed down jenkins a bit (nova cells etc seem to have all come in at once) | 23:15 |
| clarkb | notmyname: he is travelling/in an odd timezone. something the CI team can help with? | 23:15 |
| notmyname | clarkb: ya, thanks | 23:15 |
| notmyname | clarkb: a question about gerrit specifically, actually | 23:15 |
| notmyname | clarkb: first, who is allowed to submit a patch over an existing one with the same patch id? second, what happens in this situation when it is merged? who gets "credit" in git? | 23:16 |
| *** KurtMartin has joined #openstack-dev | 23:17 | |
| *** kmartin has quit IRC | 23:17 | |
| clarkb | notmyname: an example of that is https://review.openstack.org/#/c/15219/ I believe anyone can submit a patchset to an existing change. | 23:17 |
| clarkb | the author is the person that first submitted it and the committer the person that updated the change | 23:17 |
| * clarkb looks for an example in the git logs | 23:17 | |
| *** brucer has joined #openstack-dev | 23:17 | |
| notmyname | clarkb: ok. what the proper way to do this and maintain the separation of committer and submitter? rebase/squashing a patch-to-a-patch would overwrite the original author | 23:18 |
| *** patelna has quit IRC | 23:19 | |
| *** zing has quit IRC | 23:20 | |
| clarkb | Can you get around it when squashing? when you squash you are saying throw that history away right? | 23:20 |
| notmyname | ya | 23:21 |
| swifterdarrell | looks like git commit takes a "--author=<author>" flag to override the author | 23:22 |
| *** edmund has quit IRC | 23:22 | |
| notmyname | I'm sure it can be done, but the question is what's the best way? my natural behavior in git when confronted with the gerrit requirement of one-commit-per-patch is to squash | 23:22 |
| clarkb | notmyname: gerrit supports multiple dependent changes so you don't always have to squash | 23:23 |
| *** pixelbeat has joined #openstack-dev | 23:23 | |
| clarkb | notmyname: I think they only time that squashing is necessary is when multiple patches cannot pass the gate unless they are run together | 23:23 |
| clarkb | and arguably that is a bad thing... | 23:23 |
| clarkb | so I think my recommendation would be to keep patches distinct and let gerrit sort out dependencies that way you can preserve the author data | 23:24 |
| notmyname | clarkb: yes, I know /gerrit/ supports it, but the way I've seen it work in practice is that if I have 5 commits in a branch (with separate commit-ids because that's what gets added when I commit), I end up with 5 patches on the review page | 23:24 |
| clarkb | notmyname: correct | 23:24 |
| notmyname | in practice == openstack's particular config of gerrit | 23:24 |
| clarkb | I guess I don't understand the issue with that. Is the problem reviewing 5 semi distinct things? | 23:25 |
| clarkb | note that is what is currently going on with the cells stuff | 23:25 |
| notmyname | well, they normaly aren't distinct :-). for example, in the case at hand, it's adding/fixing a test case on a proposed code change | 23:26 |
| clarkb | it might make jenkins slow to a crawl but is otherwise fine | 23:26 |
| *** stevebake has quit IRC | 23:26 | |
| *** tgall_foo has joined #openstack-dev | 23:26 | |
| *** tgall_foo has quit IRC | 23:26 | |
| *** tgall_foo has joined #openstack-dev | 23:26 | |
| clarkb | notmyname: ok, that would be the case where working around the gate means they have to be one change | 23:27 |
| notmyname | and then you lose the original author. so how did you not lose the author on the example you linked? | 23:27 |
| clarkb | notmyname: I wrote a patchset and pushed it to gerrit which created the change. then someone else corrected an issue with it and pushed to the same change id | 23:28 |
| *** spiffxp has joined #openstack-dev | 23:29 | |
| swifterdarrell | clarkb: notmyname and I have each done that once and in both cases that new patch set commit had us as the Author | 23:29 |
| *** gongysh has joined #openstack-dev | 23:29 | |
| swifterdarrell | clarkb: do you know if Jeremy Stanley used --author on his commit or something? | 23:29 |
| clarkb | I don't think he did. | 23:30 |
| clarkb | fungi: ^ | 23:30 |
| *** otherwiseguy has quit IRC | 23:30 | |
| clarkb | swifterdarrell: did you `git review -d` the old change? I wonder if that does some magic | 23:30 |
| fungi | hrm | 23:30 |
| swifterdarrell | clarkb: don't think I did; I just created a new branch, committed orig changes + mine, copied in the old commit message (w/the changeid), edited it a little, and committed; then ran "git review" | 23:31 |
| fungi | all i did was check out clarkb's change, make a local modification, used commit --all --amend and then did another git review | 23:31 |
| fungi | nothing special | 23:32 |
| notmyname | one difference may be how we downloaded the change. we both used the copy/paste checkout link provided by gerrit | 23:32 |
| swifterdarrell | fungi: that's what I initially tried, but gerrit rejected it with "No changes between HEAD and gerrit/master. Submitting it for review would be pointless" | 23:32 |
| fungi | swifterdarrell: odd | 23:32 |
| clarkb | https://review-dev.openstack.org/500 we can play with this live | 23:32 |
| swifterdarrell | fungi: maybe the --all is necessary? (I didn't use that, or is that just to auto-add?) | 23:33 |
| fungi | notmyname: i too used the cut-n-paste checkout command gerrit provided | 23:33 |
| fungi | swifterdarrell: yeah, --all just auto adds | 23:33 |
| swifterdarrell | fungi: hrm... pretty sure that's what I tried, then | 23:33 |
| fungi | made sure i was on master and it was up to date, fetched the change and switched to it, then checked that out into a new topic branch | 23:34 |
| fungi | made my alterations, amended the last commit and reviewed it | 23:34 |
| swifterdarrell | fungi: ah... I was somewhere else, pasted in the gerrit git command, became on a detached HEAD, then attempted the above--that must have been the prob. | 23:35 |
| *** colinmcnamara has quit IRC | 23:35 | |
| fungi | swifterdarrell: yeah, i think you may need to checkout -b before you start messing with it | 23:35 |
| fungi | after the fetch | 23:35 |
| swifterdarrell | fungi: after "fetched the change and switched to it", you were detached HEAD state, right? | 23:35 |
| fungi | swifterdarrell: yep, and then i did a checkout -b mytopicbranch | 23:36 |
| swifterdarrell | fungi: what did you u se to check that out into a a new topic branch? git checkout -b some-branch? | 23:36 |
| fungi | exactly | 23:36 |
| *** colinmcnamara has joined #openstack-dev | 23:36 | |
| swifterdarrell | ah, cool | 23:36 |
| swifterdarrell | notmyname: in the meantime, I bet we can use "git commit --amend --author <author>" to fix up our two guys | 23:36 |
| *** colinmcnamara has quit IRC | 23:37 | |
| clarkb | fungi: https://review-dev.openstack.org/#/c/500/ want to test there? | 23:37 |
| fungi | mainly because i'm in the habit of relying on git review's topic detection, so i use a topic branch with the same name as the original change topic | 23:37 |
| fungi | clarkb: sure. gimme a sec | 23:37 |
| *** anniec has quit IRC | 23:38 | |
| *** vipuls has quit IRC | 23:38 | |
| *** littleidea has joined #openstack-dev | 23:38 | |
| *** stevebake has joined #openstack-dev | 23:38 | |
| *** jk0 has left #openstack-dev | 23:39 | |
| *** jk0 has joined #openstack-dev | 23:39 | |
| *** ChanServ sets mode: +v jk0 | 23:39 | |
| *** devcamca1 has joined #openstack-dev | 23:41 | |
| *** littleidea has quit IRC | 23:43 | |
| fungi | followed my same workflow from before and uploaded a modified patchset to https://review-dev.openstack.org/#/c/500/ | 23:43 |
| swifterdarrell | fungi: cool. So the "git checkout -b mytopic" is what I missed | 23:44 |
| fungi | seems that way | 23:45 |
| *** devcamca1 has quit IRC | 23:48 | |
| clarkb | git review -d should do that for you as well | 23:48 |
| fungi | i keep meaning to use that, and then i keep forgetting about it when i need to do it | 23:49 |
| *** rnirmal has quit IRC | 23:49 | |
| *** salv-orlando has quit IRC | 23:49 | |
| *** salv-orlando has joined #openstack-dev | 23:50 | |
| *** salv-orlando has quit IRC | 23:50 | |
| *** littleidea has joined #openstack-dev | 23:51 | |
| *** salv-orlando has joined #openstack-dev | 23:51 | |
| *** Ryan_Lane has quit IRC | 23:52 | |
| *** lloydde has quit IRC | 23:52 | |
| *** maurosr has joined #openstack-dev | 23:55 | |
| *** salv-orlando has quit IRC | 23:55 | |
| rootard | Are there any open blueprints relating to the configuration of OpenStack? | 23:58 |
| notmyname | clarkb: fungi: thanks for your time and help | 23:58 |
| devananda | clarkb: you guys just reminded me, is there a preferred way to propose >1 review for the same bp at a time? | 23:59 |
| *** vipuls has joined #openstack-dev | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!