| eglynn | jaypipes: there? | 00:00 |
|---|---|---|
| *** milner has quit IRC | 00:00 | |
| *** milner has joined #openstack-dev | 00:00 | |
| *** jakedahn has quit IRC | 00:01 | |
| *** milner has quit IRC | 00:01 | |
| *** milner has joined #openstack-dev | 00:01 | |
| *** torgomatic has quit IRC | 00:04 | |
| *** torgomatic has joined #openstack-dev | 00:05 | |
| *** martine has joined #openstack-dev | 00:12 | |
| *** jakedahn__ is now known as jakedahn | 00:15 | |
| *** jakedahn has joined #openstack-dev | 00:15 | |
| anotherjesse2 | dtroyer / dolphm / jaypipes: http://pastie.org/3492534 <- wip by bcwaldon and myself | 00:15 |
| jakedahn | what's the diff look like | 00:15 |
| vishy | jk0: ping | 00:15 |
| anotherjesse2 | dolphm: could really use some help verifying the _validate_user_token response code ... | 00:16 |
| *** torgomatic has quit IRC | 00:16 | |
| jk0 | vishy: sup | 00:17 |
| vishy | jk0: reading through the soft-reboot stuff in xen | 00:17 |
| vishy | jk0: am I correct in my impression that a soft reboot will eventually turn into a hard reboot if enough time passes? | 00:17 |
| jk0 | vishy: in xenapi or our implementation? | 00:19 |
| vishy | in your implementation | 00:19 |
| vishy | that is my reading of the cleanup stuff | 00:19 |
| vishy | in vmops | 00:19 |
| vishy | there was a note from you so i figured you might know for sure | 00:19 |
| jk0 | ah, what line #? | 00:19 |
| jk0 | ohh, I see | 00:20 |
| jk0 | yeah, anything stuck in that state will be cleaned up if that periodic task is enabled | 00:20 |
| jk0 | I think that was put in there from a kernel issue a while back | 00:22 |
| *** dubsquared has quit IRC | 00:22 | |
| jk0 | vishy: it's disabled by default | 00:24 |
| vishy | jk0: well soft reboots don't work if the guest doesn't support them | 00:24 |
| vishy | or if they are locked up | 00:25 |
| vishy | so I'm trying to see what happens in that case | 00:25 |
| vishy | does it do a hard reboot instead? | 00:25 |
| jk0 | yeah, the reboot method catches a XenAPI exception if it fails and issues the hard_reboot call | 00:25 |
| jk0 | but in this case there was no exceptions being thrown | 00:26 |
| jk0 | *were | 00:26 |
| anotherjesse2 | jakedahn: http://pastie.org/3492603 is the current diff | 00:26 |
| jk0 | hm, maybe it doesn't | 00:27 |
| jk0 | I thought it used to | 00:27 |
| *** andrewbogott is now known as andrewbogott_afk | 00:28 | |
| jk0 | vishy: well, if it does throw the XenAPI exception, you could have it issue the hard_reboot instead :) | 00:28 |
| jk0 | could have swore it did that already *shrug* | 00:29 |
| vishy | jk0: sorry maybe I'm confused | 00:29 |
| vishy | jk0: there is a cleanup task that turns failed soft_reboots into hard reboots | 00:29 |
| vishy | I'm writing the soft reboot code for libvirt | 00:29 |
| vishy | i just want to make sure it works the same way | 00:30 |
| vishy | but I have to say i don't understand the logic of issuing a hard reboot automatically | 00:30 |
| jk0 | yep, those cases were from some one-off kernel bug where it would just hang instead of throwing an exception or being ignored alltogether | 00:30 |
| *** letterj has left #openstack-dev | 00:30 | |
| jk0 | the only way to get the vm to respond is by issuing a hard reboot | 00:30 |
| jk0 | in that case | 00:30 |
| vishy | sure | 00:30 |
| vishy | but why have two reboots exposed to the user | 00:31 |
| vishy | then | 00:31 |
| vishy | shouldn't user's always request soft and let the cloud do a hard if soft fails? | 00:31 |
| jk0 | yeah, I get what you're saying | 00:32 |
| anotherjesse2 | vishy / heckj / jaypipes / dolphm - review please - https://review.openstack.org/#change,4675,patchset=14 | 00:33 |
| jk0 | if I recall correctly, the hard reboot should only be availabe in admin api | 00:33 |
| anotherjesse2 | bcwaldon is going to update the docstrings ... | 00:34 |
| jk0 | at least it was when added originally | 00:34 |
| jk0 | vishy: hopefully that makes sense.. I'm having a hard time remembering all of the details | 00:37 |
| vishy | ok | 00:38 |
| heckj | anotherjesse2: lookin' now | 00:39 |
| jk0 | bbiab | 00:39 |
| anotherjesse2 | heckj: it is probably really testable now - hoping bcwaldon will help with that! | 00:40 |
| vishy | anotherjesse2: might want to remove your oopsy | 00:40 |
| anotherjesse2 | heh | 00:40 |
| heckj | you got logging in there! Nice!!! | 00:40 |
| anotherjesse2 | might need fixed - but the basics should work | 00:41 |
| anotherjesse2 | vishy: pushed new review with 2 lines changed (comments) | 00:42 |
| *** zzed has quit IRC | 00:44 | |
| *** Arahnoid has joined #openstack-dev | 00:49 | |
| *** Arahnoid has left #openstack-dev | 00:49 | |
| *** mdomsch has joined #openstack-dev | 00:58 | |
| *** jakedahn_ has joined #openstack-dev | 01:00 | |
| *** jakedahn_ has quit IRC | 01:00 | |
| *** rbasak has quit IRC | 01:00 | |
| *** nati_ueno has quit IRC | 01:00 | |
| *** nati_ueno has joined #openstack-dev | 01:00 | |
| *** anotherjesse2 has quit IRC | 01:01 | |
| *** rods has quit IRC | 01:01 | |
| *** jakedahn has quit IRC | 01:04 | |
| *** _adjohn has joined #openstack-dev | 01:04 | |
| *** adjohn has quit IRC | 01:08 | |
| *** _adjohn is now known as adjohn | 01:08 | |
| *** markvoelker has joined #openstack-dev | 01:09 | |
| *** littleidea has quit IRC | 01:09 | |
| *** jdg has quit IRC | 01:13 | |
| *** _adjohn has joined #openstack-dev | 01:14 | |
| *** pixelbeat has quit IRC | 01:15 | |
| *** adjohn has quit IRC | 01:17 | |
| *** _adjohn is now known as adjohn | 01:17 | |
| *** bengrue has quit IRC | 01:19 | |
| openstackgerrit | Verification of a change to openstack/quantum failed: make sure pip-requires is included in setup.py sdist https://review.openstack.org/4734 | 01:27 |
| *** littleidea has joined #openstack-dev | 01:29 | |
| *** anotherjesse1 has joined #openstack-dev | 01:29 | |
| *** flaviamissi has joined #openstack-dev | 01:36 | |
| *** anotherjesse1 has quit IRC | 01:38 | |
| zul | mtaylor: still around? | 01:38 |
| *** adjohn has quit IRC | 01:43 | |
| *** troytoman is now known as troytoman-away | 01:49 | |
| *** x86brandon has quit IRC | 01:50 | |
| danwent | mtaylor or jeblair: around? have a fairly urgent openstack-ci issue. | 01:50 |
| danwent | seems like a new gate was introduced for quantum trunk, but it fails with an infrastructure error: https://jenkins.openstack.org/job/gate-quantum-python27/lastUnsuccessfulBuild/console | 01:50 |
| danwent | preventing us from getting a fix in we need for e-4 | 01:50 |
| danwent | any help would be appreciated | 01:51 |
| *** reed has quit IRC | 01:55 | |
| *** crobinso has quit IRC | 02:00 | |
| *** nati_ueno has quit IRC | 02:05 | |
| *** vincentricci has quit IRC | 02:05 | |
| *** novas0x2a|laptop has quit IRC | 02:08 | |
| *** rbasak has joined #openstack-dev | 02:08 | |
| *** markvoelker has quit IRC | 02:09 | |
| *** stuntmachine has joined #openstack-dev | 02:12 | |
| openstackgerrit | Verification of a change to openstack/quantum failed: make sure pip-requires is included in setup.py sdist https://review.openstack.org/4734 | 02:13 |
| *** hugokuo has joined #openstack-dev | 02:15 | |
| mtaylor | danwent: hey | 02:17 |
| danwent | hey | 02:18 |
| *** jog0 has left #openstack-dev | 02:19 | |
| danwent | mtaylor: did you see email? | 02:21 |
| *** jog0_ has joined #openstack-dev | 02:24 | |
| *** jog0_ has quit IRC | 02:24 | |
| *** mattray has joined #openstack-dev | 02:25 | |
| mtaylor | danwent: yes... | 02:26 |
| mtaylor | danwent: I don't think it's essential that we land that change pre-e4 | 02:26 |
| danwent | mtaylor: ok. with change from jeblair, we can now get stuff merged for e-4 milestone. then we can work on implementing the ci enhancements. | 02:27 |
| mtaylor | danwent: sounds great to me! | 02:28 |
| danwent | great. talk to you later :) | 02:28 |
| *** jakedahn has joined #openstack-dev | 02:29 | |
| *** ncode has joined #openstack-dev | 02:36 | |
| *** ncode has joined #openstack-dev | 02:36 | |
| *** adjohn has joined #openstack-dev | 02:37 | |
| *** x86brandon has joined #openstack-dev | 02:40 | |
| *** jdurgin has quit IRC | 02:41 | |
| *** Ryan_Lane has quit IRC | 02:44 | |
| *** ncode has quit IRC | 02:45 | |
| *** rbasak has quit IRC | 02:47 | |
| *** jakedahn_ has joined #openstack-dev | 02:52 | |
| *** jakedahn has quit IRC | 02:54 | |
| *** jakedahn_ is now known as jakedahn | 02:54 | |
| *** HugoKuo_ has joined #openstack-dev | 02:56 | |
| *** mattray has quit IRC | 02:56 | |
| *** rbasak has joined #openstack-dev | 03:00 | |
| *** hugokuo has quit IRC | 03:00 | |
| *** zigo has joined #openstack-dev | 03:00 | |
| *** ncode has joined #openstack-dev | 03:03 | |
| *** ncode has joined #openstack-dev | 03:03 | |
| *** x86brandon has quit IRC | 03:07 | |
| *** ncode has quit IRC | 03:07 | |
| *** stuntmachine has quit IRC | 03:13 | |
| *** ncode has joined #openstack-dev | 03:14 | |
| *** mdomsch has quit IRC | 03:14 | |
| *** ncode has quit IRC | 03:21 | |
| jakedahn | when is rc1/how does the rc process work? | 03:21 |
| *** dubsquared has joined #openstack-dev | 03:24 | |
| *** mnewby_ has joined #openstack-dev | 03:29 | |
| *** mnewby has quit IRC | 03:33 | |
| *** anotherjesse1 has joined #openstack-dev | 03:34 | |
| *** nati_ueno has joined #openstack-dev | 03:37 | |
| ohnoimdead | termie: you around? | 03:43 |
| *** nati_ueno has quit IRC | 03:56 | |
| *** nati_ueno has joined #openstack-dev | 03:56 | |
| *** danwent has quit IRC | 04:04 | |
| *** vizsla has quit IRC | 04:21 | |
| *** andrewsmedina has quit IRC | 04:24 | |
| *** danwent has joined #openstack-dev | 04:39 | |
| *** dalang has quit IRC | 04:46 | |
| *** bengrue has joined #openstack-dev | 04:57 | |
| *** dtroyer has quit IRC | 05:04 | |
| *** dalang has joined #openstack-dev | 05:04 | |
| *** heckj has quit IRC | 05:09 | |
| danwent | jakedahn: no date set. each project will fix all critical bugs they are targeting for essex, put out an rc1, and then evaluate whether any new critical bugs are found in rc1, in which case those bugs will be fixed and there will be an rc2. Rinse and repeat, with the bar of what is considered a "critical" bug rising as time goes on. Goal is that most rc1 versions will be close to e-4 drop, and that entire rc process | 05:15 |
| danwent | wraps up well in advance of april 5th. | 05:15 |
| jakedahn | cool, thx | 05:18 |
| *** adjohn has quit IRC | 05:18 | |
| *** flaviamissi_ has joined #openstack-dev | 05:43 | |
| *** martine has quit IRC | 05:45 | |
| *** flaviamissi has quit IRC | 05:46 | |
| *** HugoKuo__ has joined #openstack-dev | 05:58 | |
| *** adjohn has joined #openstack-dev | 06:00 | |
| *** HugoKuo_ has quit IRC | 06:01 | |
| *** gkotton has quit IRC | 06:19 | |
| *** HugoKuo_ has joined #openstack-dev | 06:32 | |
| *** HugoKuo__ has quit IRC | 06:35 | |
| *** gkotton has joined #openstack-dev | 06:36 | |
| *** zigo has quit IRC | 06:47 | |
| *** HugoKuo__ has joined #openstack-dev | 06:54 | |
| *** nati_ueno has quit IRC | 06:54 | |
| *** HugoKuo_ has quit IRC | 06:57 | |
| *** bepernoot has joined #openstack-dev | 07:02 | |
| YorikSar | bcwaldon: Sorry. Wrote everything I wanted to Gerrit. | 07:03 |
| *** bepernoot has quit IRC | 07:04 | |
| *** HugoKuo_ has joined #openstack-dev | 07:08 | |
| *** hugokuo has joined #openstack-dev | 07:10 | |
| *** bepernoot has joined #openstack-dev | 07:10 | |
| *** HugoKuo__ has quit IRC | 07:11 | |
| ttx | danwent: I couldn't summarize it better :) | 07:12 |
| *** HugoKuo_ has quit IRC | 07:12 | |
| *** HugoKuo_ has joined #openstack-dev | 07:14 | |
| *** hugokuo has quit IRC | 07:15 | |
| *** dubsquared has quit IRC | 07:16 | |
| *** mnewby has joined #openstack-dev | 07:18 | |
| *** bepernoot has quit IRC | 07:20 | |
| *** Mkenneth has joined #openstack-dev | 07:21 | |
| *** Mkenneth has left #openstack-dev | 07:22 | |
| *** dneary has joined #openstack-dev | 07:37 | |
| *** dneary has quit IRC | 07:37 | |
| *** dneary has joined #openstack-dev | 07:37 | |
| *** apevec has joined #openstack-dev | 07:44 | |
| *** bencherian has quit IRC | 07:46 | |
| *** shevek_ has joined #openstack-dev | 07:49 | |
| *** stanchan has joined #openstack-dev | 07:50 | |
| *** dolphm has quit IRC | 07:59 | |
| *** stanchan has quit IRC | 07:59 | |
| *** anotherjesse1 has quit IRC | 08:12 | |
| *** Ryan_Lane has joined #openstack-dev | 08:18 | |
| *** littleidea has quit IRC | 08:26 | |
| *** Mkenneth has joined #openstack-dev | 08:33 | |
| *** Ryan_Lane has quit IRC | 08:50 | |
| *** hashar has joined #openstack-dev | 08:57 | |
| *** hashar has joined #openstack-dev | 08:57 | |
| *** darraghb has joined #openstack-dev | 09:02 | |
| *** bengrue has quit IRC | 09:13 | |
| *** derekh has joined #openstack-dev | 09:18 | |
| *** adjohn has quit IRC | 09:20 | |
| *** pixelbeat has joined #openstack-dev | 09:24 | |
| ttx | mikal: around ? | 09:27 |
| *** Mkenneth has quit IRC | 09:32 | |
| *** Mkenneth has joined #openstack-dev | 09:33 | |
| *** dneary has quit IRC | 09:44 | |
| *** bepernoot has joined #openstack-dev | 09:50 | |
| *** Mkenneth has left #openstack-dev | 10:00 | |
| *** gooberguy has quit IRC | 10:01 | |
| *** bepernoot has quit IRC | 10:02 | |
| eglynn | ttx: what time today are you planning to cut E4 final? | 10:02 |
| ttx | eglynn: when the bug list is empty. Hopefully before the end of the day, otherwise very early tomorrow | 10:03 |
| ttx | While it's still Thursday in Hawaii. | 10:03 |
| eglynn | ttx: thanks ;) | 10:03 |
| eglynn | ttx: btw thanks for the approval on https://review.openstack.org/4704 | 10:04 |
| eglynn | ttx: I also have another fix backported from master https://review.openstack.org/4702 that I'd like to get into E4 | 10:05 |
| ttx | This review is just about checking that the bug is a E4 bug and was merged to master first | 10:05 |
| ttx | I saw that one -- wanted to make sure jay wanted it in E4 | 10:05 |
| ttx | Those are just one step away from getting in, so I'm not too concerned about them... The others in the list are more of a concern | 10:06 |
| eglynn | ttx: yep it was merged to master first https://review.openstack.org/4602 and I'm pretty sure Jay wants it in E4 | 10:06 |
| eglynn | ttx: (chatted yesterday about it on IRC) | 10:06 |
| ttx | eglynn: ok, will push it then | 10:06 |
| eglynn | ttx: thanks! | 10:06 |
| ttx | done | 10:07 |
| *** vizsla has joined #openstack-dev | 10:17 | |
| *** Mandell has quit IRC | 10:19 | |
| *** vizsla has quit IRC | 10:21 | |
| *** kyriakos has joined #openstack-dev | 10:29 | |
| *** bepernoot has joined #openstack-dev | 10:31 | |
| *** bepernoot has quit IRC | 10:40 | |
| *** oneiroi has joined #openstack-dev | 10:44 | |
| *** flaviamissi has joined #openstack-dev | 10:53 | |
| *** flaviamissi has quit IRC | 11:02 | |
| *** rods has joined #openstack-dev | 11:06 | |
| *** Mkenneth has joined #openstack-dev | 11:07 | |
| *** Mkenneth has joined #openstack-dev | 11:09 | |
| *** danwent has quit IRC | 11:09 | |
| *** dneary has joined #openstack-dev | 11:12 | |
| *** dneary has quit IRC | 11:12 | |
| *** dneary has joined #openstack-dev | 11:12 | |
| *** stuart2 has joined #openstack-dev | 11:16 | |
| stuart2 | eglynn: Hi Eoghan | 11:18 |
| eglynn | hi stuart2 | 11:18 |
| stuart2 | Imaginative, huh? | 11:18 |
| eglynn | yeah ;) | 11:19 |
| eglynn | did you have a chance to look at https://review.openstack.org/#change,4728 ? | 11:19 |
| eglynn | its identical to your patch on master, except with two changes ... | 11:19 |
| stuart2 | just having a look now, give me a sec | 11:19 |
| eglynn | stuart2: to save you comparing here's the first change I made: http://www.pastie.org/3492089 | 11:20 |
| stuart2 | thanks, having a look | 11:20 |
| eglynn | the problem was that check_auth_params() was being called prematurely from KeystoneStrategy._init__ | 11:21 |
| eglynn | whereas previously it was called later in KeystoneStraegy.authenticate() | 11:21 |
| eglynn | in a round-about way, this cuased nova image-list to fail | 11:21 |
| stuart2 | ah, ok. I've no problem with that change. Shows the value of the test setup, no? | 11:22 |
| eglynn | (as the nova image client does not provide the password credential explicitly, just the auth token IIRC) | 11:22 |
| eglynn | stuart2: yeah, wasn't cuaght in glance tests as there isn't a test case that mimics the nova image client exactly | 11:23 |
| eglynn | stuart2: so it only showed up in devstack integ tests, which do run nova image-list explicitly | 11:23 |
| eglynn | stuart2: so the other change is ... one sec | 11:24 |
| stuart2 | nice catch by Brian on the publicURL too. We should probably update the test cases to make sure we get a URL | 11:24 |
| eglynn | stuart2: https://review.openstack.org/#patch,sidebyside,4728,2,glance/common/auth.py line 208 | 11:24 |
| eglynn | i.e. return endpoint['publicURL'] instead of return endpoint | 11:25 |
| eglynn | stuart2: if the entire endpoint dict is returned, all the bin/glance CLI commands fail, again breaking devstack | 11:26 |
| stuart2 | Yes, absolutely needed. Do you think we should update the tests, or is time against us? | 11:26 |
| eglynn | stuart2: I *think* this slipped thru the glance tests as they always just fallback to v1 auth | 11:26 |
| eglynn | stuart2: in the medium term we need to up the test cpverage certainly | 11:26 |
| eglynn | stuart2: but in the interests of getting this into E4 quickly (time is agin' us as you say) | 11:27 |
| eglynn | stuart2: prolly best to just apply those two change to your patch on master #4350 | 11:27 |
| stuart2 | Right, but a change test_auth.py unit tests could catch it? (Perhaps take the code as is but put in a but to update test_auth.py?) | 11:28 |
| stuart2 | s/but/bug/ | 11:28 |
| eglynn | stuart2: sure, if you can get that in quickly, go for it! | 11:28 |
| eglynn | stuart2: E4 final is being cut later on today | 11:29 |
| stuart2 | eglynn: ok, I'll update my code with your changes and will look at a test change too. If the test change looks like it will slow us down I'll punt and just go with the code change. Sound ok? | 11:29 |
| eglynn | stuart2: so to have a realistic chance of getting the fix in, your patch #4350 on master will need to be merged with a couple of hours | 11:29 |
| eglynn | suart2: sounds like a plan | 11:30 |
| eglynn | stuart2 ^^^ | 11:30 |
| stuart2 | Should be able to make a start right now. I'll keep you posted | 11:30 |
| eglynn | stuart2: cool, thanks! | 11:30 |
| stuart2 | eglynn: you there? | 11:36 |
| *** zykes has joined #openstack-dev | 11:37 | |
| eglynn | stuart2: yep, whassup? | 11:37 |
| *** dayou has quit IRC | 11:38 | |
| stuart2 | would this be better? | 11:38 |
| stuart2 | return endpoint.get('publicURL') | 11:38 |
| stuart2 | (if publicURL isn't in the dict) | 11:38 |
| *** dayou has joined #openstack-dev | 11:39 | |
| eglynn | stuart2: normally yes, but in this case we'd be screwed later if the publicURL is missing | 11:39 |
| eglynn | stuart2: also it's part of the endpoint "contract" so I think we can rely on it being there | 11:40 |
| stuart2 | gotcha. Probably need to do something like raise an exception, but not now. | 11:40 |
| eglynn | stuart2: yep, cool | 11:41 |
| *** mikal has quit IRC | 11:56 | |
| *** bepernoot has joined #openstack-dev | 11:57 | |
| stuart2 | eglynn: ok, just pushed up a change. Your changes plus a bonus of two in the unit test. | 11:59 |
| stuart2 | two lines ^^ | 11:59 |
| *** mikal has joined #openstack-dev | 11:59 | |
| eglynn | stuart2: cool, looking now | 11:59 |
| stuart2 | passes our unit tests anyway... | 12:00 |
| *** dneary has quit IRC | 12:02 | |
| eglynn | stuart2: nice | 12:02 |
| eglynn | stuart2: one tiny suggestion self.assertEquals(plugin.management_url, 'http://localhost:9292') instead of self.assertTrue(plugin.management_url == 'http://localhost:9292') | 12:02 |
| eglynn | stuart2: gives more useful error message on failure | 12:03 |
| stuart2 | ok let me re-spin | 12:03 |
| eglynn | cool, thanks! | 12:03 |
| eglynn | stuart2: in the meantime, I'll update the test in milestone-proposed patch | 12:05 |
| eglynn | (https://review.openstack.org/#change,4728) | 12:05 |
| *** rods has quit IRC | 12:06 | |
| stuart2 | eglynn: ok, re-spin pushed | 12:08 |
| * eglynn +1'ing | 12:09 | |
| stuart2 | fingers crossed... | 12:10 |
| *** stuart2 has quit IRC | 12:11 | |
| openstackgerrit | Verification of a change to openstack/glance failed: Allow region selection when using V2 keystone https://review.openstack.org/4350 | 12:18 |
| *** hashar is now known as hasharEat | 12:19 | |
| eglynn | stuart2: devstack passes! but glance test fail now ... grrr! | 12:21 |
| eglynn | https://jenkins.openstack.org/job/gate-glance-python27/59/console | 12:22 |
| eglynn | stuart2: looks like a transient failure | 12:22 |
| *** dprince has joined #openstack-dev | 12:22 | |
| * eglynn pushing it through again | 12:22 | |
| *** dprince has quit IRC | 12:31 | |
| *** flaviamissi has joined #openstack-dev | 12:32 | |
| eglynn | stuart2: https://review.openstack.org/4350 is finally merged, yay! | 12:33 |
| * eglynn is looking to scare up some glance core reviewers for the corresponding patch on milestone-proposed branch | 12:34 | |
| eglynn | https://review.openstack.org/#change,4728 | 12:34 |
| *** bepernoot has quit IRC | 12:36 | |
| *** cjwilson has quit IRC | 12:39 | |
| *** hasharEat is now known as hashar | 12:43 | |
| *** markvoelker has joined #openstack-dev | 12:43 | |
| *** andrewsmedina has joined #openstack-dev | 12:46 | |
| *** ches has quit IRC | 12:47 | |
| *** ches has joined #openstack-dev | 12:48 | |
| *** andrewsmedina has quit IRC | 12:53 | |
| *** dtroyer has joined #openstack-dev | 13:03 | |
| *** stuntmachine has joined #openstack-dev | 13:11 | |
| *** lts has joined #openstack-dev | 13:13 | |
| ttx | eglynn: I can review milestone-proposed patches. Looking | 13:16 |
| *** andrewsmedina has joined #openstack-dev | 13:19 | |
| eglynn | ttx: thanks! | 13:20 |
| ttx | eglynn: the patch is not strictly similar (extra newline in milestone-proposed), but I guess we can live with that | 13:21 |
| eglynn | ttx: I can remove the newline | 13:21 |
| ttx | eglynn: ok, do it and I'll approve | 13:22 |
| ttx | glance/common/auth.py:119 | 13:22 |
| ttx | hrm | 13:23 |
| ttx | :120 actually. | 13:24 |
| *** dprince has joined #openstack-dev | 13:24 | |
| eglynn | ttx: newline-be-gone! | 13:26 |
| ttx | eglynn: and approved | 13:26 |
| eglynn | ttx: thanks! | 13:26 |
| ttx | eglynn: last one on list is https://review.openstack.org/#change,4609 | 13:28 |
| ttx | (for Glance) | 13:28 |
| * eglynn looking ... | 13:29 | |
| ttx | nova-core reviewers: please give the following E4 reviews some love: | 13:29 |
| ttx | https://review.openstack.org/#change,4621 | 13:29 |
| ttx | https://review.openstack.org/#change,4646 | 13:30 |
| *** dneary has joined #openstack-dev | 13:30 | |
| *** dneary has quit IRC | 13:30 | |
| *** dneary has joined #openstack-dev | 13:30 | |
| *** mancdaz1203 has quit IRC | 13:31 | |
| eglynn | ttx: 4609 still has whitespace errors, I'll see if I can the author to re-submit with these fixed up | 13:31 |
| *** dtroyer has quit IRC | 13:32 | |
| ttx | eglynn: or we'll push it to rc1 | 13:32 |
| *** sandywalsh has quit IRC | 13:32 | |
| *** berendt has joined #openstack-dev | 13:32 | |
| eglynn | ttx: yep, its fairly low down the priority stack | 13:32 |
| *** mancdaz1203 has joined #openstack-dev | 13:35 | |
| *** hashar has quit IRC | 13:37 | |
| *** roge has joined #openstack-dev | 13:38 | |
| *** j05h1 has joined #openstack-dev | 13:40 | |
| *** vizsla has joined #openstack-dev | 13:42 | |
| *** mattray has joined #openstack-dev | 13:44 | |
| *** sandywalsh has joined #openstack-dev | 13:44 | |
| *** bsza has joined #openstack-dev | 13:47 | |
| *** martine has joined #openstack-dev | 13:55 | |
| *** dtroyer has joined #openstack-dev | 14:06 | |
| *** littleidea has joined #openstack-dev | 14:07 | |
| *** joesavak has joined #openstack-dev | 14:23 | |
| *** rods has joined #openstack-dev | 14:23 | |
| *** littleidea has quit IRC | 14:29 | |
| smoser | vishy, here now. | 14:37 |
| *** mdomsch has joined #openstack-dev | 14:55 | |
| *** Mkenneth has quit IRC | 14:58 | |
| *** cp16net has joined #openstack-dev | 15:03 | |
| *** bencherian has joined #openstack-dev | 15:07 | |
| *** misheska has joined #openstack-dev | 15:09 | |
| *** mancdaz1203 has quit IRC | 15:10 | |
| *** mancdaz1203 has joined #openstack-dev | 15:10 | |
| *** AlanClark has joined #openstack-dev | 15:13 | |
| *** misheska has quit IRC | 15:15 | |
| *** mancdaz1203 has quit IRC | 15:17 | |
| *** kbringard has joined #openstack-dev | 15:17 | |
| ttx | jaypipes: let me know if we should wait for 908803 to do E4 | 15:19 |
| annegentle | Happy Hack Day Stackers! | 15:25 |
| *** mancdaz1203 has joined #openstack-dev | 15:26 | |
| *** danwent has joined #openstack-dev | 15:29 | |
| *** bencherian has quit IRC | 15:36 | |
| *** bencherian_ has joined #openstack-dev | 15:36 | |
| *** vincentricci has joined #openstack-dev | 15:37 | |
| *** hashar has joined #openstack-dev | 15:48 | |
| *** hub_cap has joined #openstack-dev | 15:50 | |
| smoser | ttx, bug 942965 seems like it should be on some "important list" | 15:50 |
| uvirtbot` | Launchpad bug 942965 in unity "Press Alt+F1, panel’s shadow dissapear" [Low,Triaged] https://launchpad.net/bugs/942965 | 15:50 |
| *** maplebed has quit IRC | 15:50 | |
| smoser | oops | 15:50 |
| smoser | bug 942865 | 15:51 |
| uvirtbot` | Launchpad bug 942865 in nova "upgrade from diablo leaves existing images with kernel unbootable" [Undecided,New] https://launchpad.net/bugs/942865 | 15:51 |
| smoser | but pressing alt+f1 and having panel shadow disappear is important too | 15:51 |
| smoser | :) | 15:51 |
| vishy | smoser: is there any way to get cirrus to support acpi? | 15:52 |
| smoser | not that i know of. i'd have to poke. | 15:53 |
| smoser | what is it that you're wanting ? | 15:53 |
| smoser | to respond cleanly to shutdown ? | 15:53 |
| smoser | and i'm surprised i'm even resonding to you when you spelled its name wrong :)~ cirros is an OS for the clouds, cirrus *is* a cloud | 15:54 |
| *** zzed has joined #openstack-dev | 15:55 | |
| *** kbringard has quit IRC | 15:57 | |
| ttx | vishy: could we look into the E4 buglist now ? | 15:57 |
| *** hhoover has joined #openstack-dev | 15:59 | |
| vishy | ttx: No i'm off to the dr. | 16:01 |
| vishy | In an hour? | 16:01 |
| ttx | vishy: ok | 16:01 |
| vishy | smoser: hehe, sorry. I would like cirros to respond to the signal | 16:02 |
| vishy | virsh shutdown has no effect | 16:02 |
| ttx | sounds like cirrhosis | 16:02 |
| smoser | vishy, open a bug, please. i agree that would be nice. | 16:03 |
| smoser | https://launchpad.net/cirros | 16:03 |
| *** crobinso has joined #openstack-dev | 16:03 | |
| jaypipes | ttx: did you approve the two branches that were in glance's milestone-proposed branch? | 16:06 |
| ttx | yes | 16:06 |
| jaypipes | ttx: I've moved the other three bugs to the RC | 16:06 |
| *** dolphm has joined #openstack-dev | 16:06 | |
| jaypipes | ttx: ok, good to ship, AFAIAC | 16:07 |
| jaypipes | ttx: thx for help | 16:07 |
| *** gyee has joined #openstack-dev | 16:09 | |
| zykes- | anyone here from Mirantis ? | 16:13 |
| ttx | ok, glance E4 done | 16:15 |
| *** Mkenneth has joined #openstack-dev | 16:16 | |
| *** martine has quit IRC | 16:20 | |
| ttx | quantum E4 done | 16:20 |
| ttx | zns, heckj, termie: keystone E4 looks ready to publish according to https://launchpad.net/keystone/+milestone/essex-4 | 16:21 |
| ttx | will publish in a few, unless you scream | 16:22 |
| ttx | devcamcar: horizon E4 looks ready to publish according to https://launchpad.net/horizon/+milestone/essex-4 | 16:22 |
| ttx | will also publish in a few minutes, unless you scream | 16:23 |
| *** vizsla has quit IRC | 16:24 | |
| *** heckj has joined #openstack-dev | 16:26 | |
| *** andrewsben has joined #openstack-dev | 16:30 | |
| *** nati_ueno has joined #openstack-dev | 16:33 | |
| ttx | keystone E4 done | 16:35 |
| joesavak | yay! | 16:35 |
| heckj | WOOT! | 16:35 |
| *** dubsquared has joined #openstack-dev | 16:35 | |
| dolphm | ttx: :D | 16:35 |
| ttx | only nova and horizon left | 16:36 |
| * dolphm furiously writes migration docs | 16:36 | |
| *** maplebed has joined #openstack-dev | 16:36 | |
| heckj | dolphm: thank yoU! | 16:36 |
| ttx | (melange E4 is done too) | 16:36 |
| joesavak | good job heckj, dolphm, termie, and anotherjesse!! | 16:36 |
| joesavak | heckj: I don't see any BP on AD integration. Ok if I open one, or will your flamethrower torch it? | 16:39 |
| heckj | You can open it if you're going to advocate for it :-) (i.e. assign it to yourself). I think we've got some use cases around that in the wiki, but nothing yet in the blueprint list. | 16:41 |
| jaypipes | QA meeting in 19 minutes on #openstack-meeting! | 16:41 |
| *** andrewbogott_afk is now known as andrewbogott | 16:43 | |
| *** andrewbogott is now known as andrewbogott_afk | 16:43 | |
| *** andrewbogott_afk is now known as andrewbogott | 16:43 | |
| *** jsavak has joined #openstack-dev | 16:46 | |
| andrewbogott | is essex a done deal, or will there be yet another release cut at some point? | 16:47 |
| heckj | andrewbogott: Ther will be release candidates coming for Essex | 16:48 |
| heckj | andrewbogott: view E4 as a final feature freeze window - the intention is to focus on QA, testing, bug fixing (and in my case) docs until we cut a final Essex release. | 16:49 |
| andrewbogott | In that case... can I get a +2 from someone on change 4209? bcwaldon perchance? | 16:49 |
| *** joesavak has quit IRC | 16:49 | |
| *** andrewbogott is now known as andrewbogott_afk | 16:52 | |
| *** berendt has quit IRC | 16:52 | |
| *** jdg has joined #openstack-dev | 16:53 | |
| ttx | Horizon E4 done | 16:55 |
| *** andrewbogott_afk is now known as andrewbogott | 16:55 | |
| *** berendt has joined #openstack-dev | 16:56 | |
| ohnoimdead | ttx: thanks! | 16:57 |
| ttx | Only Nova left, waiting for vishy to trim the E4 blockers. | 16:57 |
| *** armaan has joined #openstack-dev | 16:58 | |
| *** dwalleck has joined #openstack-dev | 16:59 | |
| *** spiffxp has joined #openstack-dev | 17:01 | |
| vishy | ttx: ready | 17:02 |
| ttx | vishy: looking at https://launchpad.net/nova/+milestone/essex-4 | 17:02 |
| ttx | vishy: looks like none of those bugs are ready to land except mine, proposed @ https://review.openstack.org/4744 | 17:02 |
| ttx | vishy: do you want to delay E4 for any of the others ? | 17:03 |
| vishy | i don't think it is necessary to delay | 17:03 |
| vishy | i think they can go rc1 | 17:03 |
| vishy | https://review.openstack.org/#change,4621 | 17:03 |
| vishy | that one just needs another core approve | 17:03 |
| vishy | i just sent in yours | 17:04 |
| ttx | vishy: we can wait for mine + this one to hit, if you can get it reviewed and backport it | 17:04 |
| ttx | I can certainly give you a couple of hours, I'm about to pause for dinner | 17:05 |
| vishy | pvo, johan_-_, tr3buchet, can someone review this: https://review.openstack.org/#change,4621 | 17:05 |
| vishy | should be a quicky | 17:05 |
| ttx | I push the others to rc1 | 17:05 |
| vishy | i got a soft-reboot review in | 17:05 |
| tr3buchet | vishy: on it | 17:05 |
| vishy | although i haven't been able to functionally test it yet with a vm that supports soft reboot. It does fallback to hard reboot successfully | 17:06 |
| ttx | vishy: I think that can wait rc1 | 17:06 |
| ttx | but I'll take whatever you can sneak in in the next two hours :) | 17:06 |
| ttx | vishy: deal ? | 17:06 |
| heckj | dtroyer: ping | 17:07 |
| armaan | hi folks, when i give the command euca-describe-availability-zones verbose, i got an error:500 server has either erred or is incapable of performing the request... please help | 17:07 |
| *** littleidea has joined #openstack-dev | 17:07 | |
| *** johngarbutt has joined #openstack-dev | 17:08 | |
| dtroyer | heckj: dtroyer is alive | 17:08 |
| ttx | Pushed bug 942156 to RC1 | 17:08 |
| uvirtbot` | Launchpad bug 942156 in nova "floating-ip-delete disassociates and doesn't de-allocate when ip is associated." [Critical,Triaged] https://launchpad.net/bugs/942156 | 17:08 |
| heckj | dtroyer: reading up on https://bugs.launchpad.net/devstack/+bug/942979, and I think you do have it covered with https://review.openstack.org/4668. Going to mark it fix committed, but wanted to check in with you first and see if there was anything you saw as missing | 17:09 |
| uvirtbot` | Launchpad bug 942979 in devstack "devstack should create admin tenant, users for each service, and associated role for each" [Undecided,Confirmed] | 17:09 |
| dtroyer | heckj: I'm good with Tres' answer, I think that was it | 17:10 |
| heckj | dtroyer: fix committed then! | 17:11 |
| *** ayoung-afk is now known as ayoung | 17:11 | |
| *** vizsla has joined #openstack-dev | 17:13 | |
| *** Gordonz has joined #openstack-dev | 17:15 | |
| zul | jaypipes: ping have you seen this before? http://paste.ubuntu.com/863745/ | 17:16 |
| *** andrewsben has quit IRC | 17:16 | |
| *** PotHix has joined #openstack-dev | 17:17 | |
| *** Gordonz has quit IRC | 17:18 | |
| *** Gordonz has joined #openstack-dev | 17:19 | |
| ayoung | termie, just posted a response on the LDAP get_user_by_name review. I think the patch is ready to push as is. | 17:22 |
| termie | ayoung: i'll take a look when i get to the office in a bit | 17:23 |
| ayoung | thanks | 17:23 |
| armaan | hi :) , when i give the command euca-describe-availability-zones verbose, it gives an error 500: server has either erred or is incapable of performing the request... please help | 17:24 |
| heckj | dtroyer: er, looks like we need to backtrack and add in a horizon service user after all - password changing in keystone is Admin only (right now), so it'll need a service account. I'll go make a patch though. | 17:28 |
| dtroyer | heckj: roger | 17:29 |
| *** x86brandon has joined #openstack-dev | 17:31 | |
| *** x86brandon has quit IRC | 17:31 | |
| *** x86brandon has joined #openstack-dev | 17:31 | |
| zul | vishy: ping have you seen this before? http://paste.ubuntu.com/863739/ | 17:32 |
| *** armaan has left #openstack-dev | 17:32 | |
| *** jsavak has quit IRC | 17:32 | |
| *** shevek_ has quit IRC | 17:35 | |
| *** oneiroi has quit IRC | 17:37 | |
| ohnoimdead | dtroyer, heckj: are we really putting super user creds back in local_settings.py? | 17:37 |
| heckj | ohnoimdead: if you want to be able to change a user's password, where should the credentials get stored for the service account? | 17:39 |
| *** nati_ueno has quit IRC | 17:40 | |
| *** ss7pro has joined #openstack-dev | 17:40 | |
| ohnoimdead | heckj: nowhere because a user should be able to update their own password? | 17:40 |
| ss7pro | Hi, there's a small bug with nova-api-metadata withtin POSTROUTING chain | 17:40 |
| ohnoimdead | heckj: (you knew i had to say that) | 17:41 |
| ss7pro | chain name is limited to 28 chars | 17:41 |
| devcamcar | ttx: e4 can drop for horizon | 17:41 |
| ss7pro | so when nova-api-metadata-POSTROUTING | 17:41 |
| ss7pro | is to long | 17:41 |
| ss7pro | and it breaks this component from working | 17:41 |
| devcamcar | ttx: and it's already been done, so even better :) | 17:41 |
| ss7pro | is't known bug ? | 17:41 |
| ttx | devcamcar: I read your mind through the milestone page.. and heckj. | 17:42 |
| devcamcar | ttx: awesome :) | 17:42 |
| *** dneary has quit IRC | 17:43 | |
| ss7pro | "iptables-restore v1.4.10: Invalid ch"iptables-restore v1.4.10: Invalid chain name `nova-api-metadata-POSTROUTING' (28 chars max)\nError occurred at line: 16\nTry `iptables-restore -h' or 'iptables-restore --help' for more information.\n" ain name `nova-api-metadata-POSTROUTING' (28 chars max)\nError occurred at line: 16\nTry `iptables-restore -h' or 'iptables-restore --help' for more information.\n" | 17:43 |
| heckj | dtroyer: (https://review.openstack.org/4752) needs testing, and I need to figure out with horizon team how to put in the creds while they cry - so WIP for now | 17:43 |
| ohnoimdead | heckj: there's really no place to put it other than local_settings.py. horizon has no data store. it's only backend is openstack. | 17:44 |
| *** dwalleck has quit IRC | 17:45 | |
| *** johngarbutt has quit IRC | 17:45 | |
| ohnoimdead | heckj: you got me so flustered i used the wrong its. | 17:47 |
| *** zigo has joined #openstack-dev | 17:47 | |
| ss7pro | anyone ? | 17:47 |
| *** AlanClark has quit IRC | 17:49 | |
| *** ches has quit IRC | 17:50 | |
| *** ches has joined #openstack-dev | 17:50 | |
| *** Mkenneth has quit IRC | 17:51 | |
| *** Mkenneth has joined #openstack-dev | 17:52 | |
| *** AlanClark has joined #openstack-dev | 17:52 | |
| jaypipes | ttx: I goofed and set https://bugs.launchpad.net/glance/+bug/936798 to RC milestone instead of E4 :( didn't realize you and eglynn had gotten it thourhg :( | 17:53 |
| uvirtbot` | Launchpad bug 936798 in glance "Glance always picks the first endpoint from the service catalog" [High,Fix released] | 17:53 |
| jaypipes | ttx: same for https://bugs.launchpad.net/bugs/920386 :( | 17:54 |
| uvirtbot` | Launchpad bug 920386 in glance "Client side SSL related variables" [Medium,Fix released] | 17:54 |
| *** jdurgin has joined #openstack-dev | 17:57 | |
| dtroyer | heckj: looks good, tested ok | 17:58 |
| *** bencherian_ has quit IRC | 17:59 | |
| heckj | dtroyer: thanks - I've got to add something into configs, and then we'll be ready to roll with it | 17:59 |
| *** bencherian has joined #openstack-dev | 17:59 | |
| jaypipes | jeblair: mind checking into why Daryl Walleck doesn't have core permissions in the stable/diablo Tempest branch? See his comment here: https://review.openstack.org/#change,4722 | 17:59 |
| ss7pro | https://bugs.launchpad.net/nova/+bug/944184 | 18:02 |
| uvirtbot` | Launchpad bug 944184 in nova "Iptables chaine name is to long for nova-api-metadata" [Undecided,New] | 18:02 |
| jeblair | jaypipes: https://launchpad.net/~openstack-stable-maint/+members | 18:02 |
| jeblair | jaypipes: only openstack-stable-maint can +2 stable/* | 18:03 |
| *** hhoover has left #openstack-dev | 18:03 | |
| jeblair | (openstack-release is a member of openstack-stable-maint, so they can too) | 18:03 |
| *** hashar has quit IRC | 18:03 | |
| *** Mandell has joined #openstack-dev | 18:04 | |
| *** dolphm has quit IRC | 18:05 | |
| *** Mandell has quit IRC | 18:05 | |
| jaypipes | jeblair: ok, thx for the explanation. I'd forgotten about that... | 18:06 |
| *** vizsla has quit IRC | 18:07 | |
| *** vizsla has joined #openstack-dev | 18:07 | |
| *** derekh has quit IRC | 18:08 | |
| danwent | anyone seeing an error with devstack when adding the cirros image? Following command is returning an error it seems: zcat --force /home/nicira/devstack/files/images/cirros-0.3.0-x86_64-uec/cirros-0.3.0-x86_64-blank.img | 18:08 |
| danwent | result is that nova image-list only includes ramdisk & kernel | 18:08 |
| *** jog0 has joined #openstack-dev | 18:09 | |
| *** dolphm has joined #openstack-dev | 18:10 | |
| *** rnirmal has joined #openstack-dev | 18:13 | |
| *** oneiroi has joined #openstack-dev | 18:14 | |
| *** oneiroi has joined #openstack-dev | 18:14 | |
| *** hhoover has joined #openstack-dev | 18:15 | |
| *** hhoover has left #openstack-dev | 18:15 | |
| *** Gordonz has quit IRC | 18:17 | |
| *** maplebed has quit IRC | 18:26 | |
| *** Mandell has joined #openstack-dev | 18:30 | |
| *** Gordonz has joined #openstack-dev | 18:31 | |
| *** mdomsch has quit IRC | 18:31 | |
| *** pixelbeat has quit IRC | 18:31 | |
| *** nati_ueno has joined #openstack-dev | 18:35 | |
| *** andrewbogott is now known as andrewbogott_afk | 18:35 | |
| *** zaitcev has joined #openstack-dev | 18:35 | |
| *** ss7pro has quit IRC | 18:36 | |
| *** mdomsch has joined #openstack-dev | 18:36 | |
| *** mnewby has joined #openstack-dev | 18:36 | |
| *** rnirmal has quit IRC | 18:38 | |
| *** dolphm has quit IRC | 18:39 | |
| openstackgerrit | Verification of a change to openstack/horizon failed: Added package_data to horizon https://review.openstack.org/4745 | 18:40 |
| *** darraghb has left #openstack-dev | 18:41 | |
| termie | hrmph, openstack hackathon at dreamhost today, but i feel like crap and should maybe sit alone at the office :/ | 18:43 |
| gyee | anyone from rackspace? | 18:44 |
| *** hub_cap has quit IRC | 18:45 | |
| termie | gyee: lots | 18:46 |
| *** Ryan_Lane has joined #openstack-dev | 18:46 | |
| *** mnewby has quit IRC | 18:52 | |
| *** adjohn has joined #openstack-dev | 18:55 | |
| markwash_ | littleidea: around? | 18:57 |
| littleidea | markwash_: I am | 18:57 |
| markwash_ | just wondering if you are working on 928521 already | 18:58 |
| markwash_ | my team was looking at it to, but we can just go hands off if you prefer and help with the review process | 18:59 |
| *** adjohn has quit IRC | 18:59 | |
| *** adjohn has joined #openstack-dev | 19:00 | |
| markwash_ | littleidea: ^^ | 19:00 |
| *** vizsla has quit IRC | 19:02 | |
| littleidea | markwash_: I can make a 'fix' that changes the vm state back to ACTIVE fairly trivially, vishy and bcwaldon suggested it would be better to change it so API returns a 500, which I think I have pretty close, but I've struggled to see how to test it. | 19:02 |
| littleidea | markwash_: have you looked at it enough to have a plan of attack? | 19:03 |
| *** nati_ueno has quit IRC | 19:03 | |
| markwash_ | littleidea: I think my plan is just "set it back to active" :-) | 19:03 |
| markwash_ | and to clean up some of the nastiness in that vicinity | 19:03 |
| littleidea | the concern was that is a silent failure | 19:03 |
| markwash_ | littleidea: but I'm confused about the 500 because at that point the message has been cast off and the api has returned 200 already | 19:04 |
| markwash_ | littleidea: agreed, but the api docs say its okay I believe | 19:04 |
| littleidea | change the cast to a call | 19:04 |
| markwash_ | littleidea: since the transitions go from ACTIVE -> ACTIVE (even on error) | 19:04 |
| littleidea | that's what was advocated when we discussed it | 19:04 |
| westmaas | lp928521 | 19:05 |
| westmaas | aww thought that would link it :( | 19:05 |
| markwash_ | littleidea: I think we should keep the api as async as possible here | 19:06 |
| markwash_ | the scheduler could take a while or be physically distributed | 19:06 |
| markwash_ | the silent failure issue can be addressed later when we have better instance fault reporting | 19:07 |
| littleidea | markwash_: I was deferring to vishy there, since I didn't feel I had enough context to advocate something | 19:07 |
| littleidea | markwash_: in the discussion, I brought up that I thought a per instance event log seemed like the best way to do it, and everyone seemed to agree, but that would be hard to plumb for essex | 19:08 |
| westmaas | littleidea: sounds like silent failure was maybe bcwaldon and vishy's concern and not yours, but can you help me understand what you mean there? | 19:08 |
| westmaas | if it was yours, even better :) | 19:09 |
| littleidea | westmaas: if you try to resize and instance, and the scheduler can't find an suitable host, it leaves the vm state as RESIZING, but the vm is unchanged | 19:10 |
| jdg | clayg: ping? | 19:10 |
| markwash_ | westmaas: and just switching it back to ACTIVE is even silenter | 19:10 |
| littleidea | westmaas: so if you just change it back to ACTIVE, a user may not realize the resize/migration failed | 19:11 |
| *** andrewsben has joined #openstack-dev | 19:11 | |
| markwash_ | littleidea: but wouldn't they know because it never landed in CONFIRM_RESIZE ? | 19:11 |
| westmaas | littleidea: gotcha. seems like nirmal's suggestion then of doing an async fault may be sufficient? | 19:11 |
| *** markwash_ is now known as markwash | 19:11 | |
| *** blamar is now known as markwash_ | 19:11 | |
| *** markwash_ is now known as blamar | 19:11 | |
| *** aweiss has joined #openstack-dev | 19:12 | |
| littleidea | markwash_: the speculation was someone might assume overlook/space off/think someone else dealt with the confirmation | 19:12 |
| markwash | littleidea: that makes sense, I guess I just think synchronous behavior here is more costly | 19:12 |
| markwash | littleidea: it would destroy cells/zones | 19:13 |
| markwash | littleidea: maybe there should be a way for us to expose some faults even when in ACTIVE for some defined period of time | 19:13 |
| littleidea | westmaas: the async fault is a little more involved that that suggestion, because there is a conditional to bail if the call isn't 'run_instance' | 19:13 |
| littleidea | the code is already technically trying to set the fault from my understanding | 19:14 |
| westmaas | a conditional to not set async fault if it isn't run_instance? | 19:14 |
| markwash | westmaas: yeah there is a tremendously dumb conditional in there | 19:14 |
| westmaas | I guess we have a problem of not being able to clear the async fault | 19:15 |
| markwash | maybe we should lure vishy or bcwaldon into this chat | 19:15 |
| markwash | vishy: ^^ | 19:15 |
| markwash | bcwaldon: ^^ | 19:15 |
| westmaas | bcwaldon: vv | 19:15 |
| westmaas | hi | 19:15 |
| littleidea | if you look in nova/scheduler/manager.py, you can see it calls _set_instance_error on any exception | 19:15 |
| westmaas | gotcha | 19:15 |
| westmaas | what other cases go through the scheduler that could error? | 19:17 |
| markwash | westmaas: I'm pretty sure it is only run_instance and prep_resize | 19:18 |
| westmaas | resize, obviously | 19:18 |
| markwash | which is why I thought it was dumb | 19:18 |
| markwash | "we need to be so general to handle these two cases" | 19:18 |
| ttx | vishy: ping | 19:20 |
| littleidea | if there isn't a blue print for per instance event logs, seems like there should be | 19:21 |
| westmaas | littleidea: yeah, we need some better features around that, I agree. | 19:22 |
| *** nati_ueno has joined #openstack-dev | 19:22 | |
| littleidea | not sure where/how that should be stored | 19:22 |
| ttx | vishy: ready to defer the last 3 bugs and release milestone-proposed at e65045855920aeaa60032ea5f4e823a9a2e856a6 | 19:22 |
| vishy | westmaas: there is no way to display aync fault unless a vm is in error | 19:22 |
| vishy | ttx: yep lets do it | 19:22 |
| *** hub_cap has joined #openstack-dev | 19:22 | |
| littleidea | the db makes me shudder | 19:22 |
| vishy | westmaas: and the vm isn't really broken so putting it into an error state seems wrong | 19:23 |
| westmaas | error is wrong. agreed. | 19:23 |
| westmaas | and its rough to show async error on active, cause there could be a really old error in there | 19:23 |
| *** hub_cap has quit IRC | 19:23 | |
| vishy | westmaas: so returning to active seems ok, but i think bcwaldon was suggesting that it might be nice to let a user know that it failed | 19:24 |
| littleidea | vishy: what creates that display limitation? | 19:24 |
| vishy | westmaas: by returning an error using a call vs. cast | 19:24 |
| westmaas | probably the instance controller | 19:24 |
| *** hub_cap has joined #openstack-dev | 19:24 | |
| vishy | littleidea, westmaas: the spec is the limitation | 19:24 |
| markwash | vishy: I wish we let the know of the failure, but the docs do seem to provide cover for _not_ letting them know explicitly --> http://docs.openstack.org/api/openstack-compute/2/content/Resize_Server-d1e3707.html | 19:24 |
| westmaas | vishy: yeah call just isn't going to work I think. especially once we move to cells. but really even before then. | 19:24 |
| dprince | westmaas: what about using task_state? | 19:25 |
| dprince | westmaas: That would keep the instance in an ACTIVE state but allow you to query for the failure via an extension? | 19:25 |
| littleidea | vishy: what does it take to change a spec? | 19:25 |
| *** utlemming has quit IRC | 19:26 | |
| dprince | westmaas: And avoid the use of a call. | 19:26 |
| westmaas | dprince: possible, worried about exposing too much there | 19:26 |
| *** jdg has quit IRC | 19:26 | |
| *** utlemming has joined #openstack-dev | 19:26 | |
| westmaas | though I guess if a deployer wanted to have that it would be their choice | 19:26 |
| ttx | nova E4 done | 19:26 |
| westmaas | worried about depending on that too much and not being able to change internal task states as freely | 19:27 |
| dprince | westmaas: how is that exposing too much? | 19:27 |
| westmaas | but maybe we go back to active, as the spec says | 19:27 |
| markwash | dprince: task_states are private | 19:27 |
| vishy | littleidea: well the spec is finalized so we can't really | 19:27 |
| markwash | or at least feel private to me | 19:27 |
| vishy | littleidea: another version of the api? | 19:27 |
| vishy | littleidea: or we could create an extension that shows faults on active instances | 19:28 |
| markwash | vishy: I'm pretty sure the current spec allows for the silent failure we're worried about | 19:28 |
| littleidea | vishy: this is the api for eternity? | 19:28 |
| westmaas | vishy: or dprince's ideas of using the existing task state extension | 19:28 |
| markwash | and its not completely silent, as you should expect it to go to confirm_resize, not ACTIVE | 19:28 |
| *** aweiss has quit IRC | 19:28 | |
| westmaas | and then add the event log in the next release | 19:28 |
| vishy | dprince: task_state does show up in dash through the extened state extension | 19:28 |
| vishy | dprince, so that is reasonable i guess | 19:29 |
| westmaas | event log extension I suppose | 19:29 |
| dprince | vishy: sure. Control panels and the like can get at it already. | 19:29 |
| westmaas | and see if that makes it in to core | 19:29 |
| vishy | dprince: westmaas does make a good point about causing potential side-effects though | 19:29 |
| vishy | westmaas: +1 to event log extension | 19:29 |
| westmaas | vishy: I'd just say active for now, task extension in the interim, event log extension for long term | 19:30 |
| vishy | so what do we set the task state to? | 19:30 |
| westmaas | that'd require setting the task state though.. | 19:30 |
| westmaas | and that affects 409s on other actions | 19:30 |
| westmaas | failing_resize seems odd heh | 19:31 |
| dprince | ACTIVE. resize_failed | 19:32 |
| *** Mandell has quit IRC | 19:32 | |
| westmaas | we could also set an async fault so that operations teams can easily see faults | 19:33 |
| westmaas | not great for end user though | 19:33 |
| *** dwalleck has joined #openstack-dev | 19:33 | |
| markwash | any task_state is going to break some api functions | 19:33 |
| westmaas | dprince: that'd be fine just have to go make sure other api actions work | 19:33 |
| *** mikeyp has joined #openstack-dev | 19:34 | |
| markwash | but I suppose we can add the task state to the problematic check_instance_state usages | 19:34 |
| markwash | e.g. compute.api.API::stop() would break with a new task state for this | 19:34 |
| westmaas | littleidea: have we sufficiently scared you away from this bug? :) | 19:35 |
| *** aweiss has joined #openstack-dev | 19:35 | |
| vishy | ttx: where is the best graph for nova bugs? http://webnumbr.com/open-nova-bugs ? | 19:35 |
| littleidea | the bug isn't scary, though it seems like a lot is working around hand cuffs to a spec. | 19:36 |
| ttx | vishy: depends on what you're after ;) | 19:37 |
| ttx | http://wiki.openstack.org/releasestatus/nova.html | 19:37 |
| littleidea | westmaas: if the api doesn't change to a call, I even think I understand how to test changing the state | 19:37 |
| ttx | it might make sense to substract the wishlist bugs from that opn bugs figure | 19:38 |
| clayg | jdg: pong? | 19:39 |
| danwent | blamar: ping? have a chance to quickly comment on style issues? I'm fine changing them if you prefer | 19:39 |
| littleidea | vishy, markwash, westmaas: is just setting the vm_state to ACTIVE and logging enough for now? | 19:39 |
| littleidea | let the chips fall | 19:40 |
| danwent | blamar: context is https://review.openstack.org/#change,4754 | 19:40 |
| markwash | littleidea: +1 from me. . might want to ask comstud or sandywalsh about any possible notification need | 19:40 |
| blamar | danwent: yeah, I saw Monty's updates and didn't want to heavily conflict + didn't have bugs/blueprints or anything | 19:40 |
| danwent | blamar: sorry, was talking about nova change | 19:40 |
| blamar | danwent: haha hold please | 19:40 |
| markwash | littleidea: but those could be added later so I wouldn't worry about it | 19:40 |
| *** novas0x2a|laptop has joined #openstack-dev | 19:41 | |
| littleidea | ok, I'll make that change and push for review then | 19:41 |
| markwash | littleidea: sweet | 19:41 |
| *** aweiss has quit IRC | 19:42 | |
| sandywalsh | littleidea, what's the question? | 19:43 |
| blamar | danwent: replied | 19:44 |
| *** mattray has quit IRC | 19:45 | |
| littleidea | sandywalsh: right now if a resize fails to find a host in the scheduler, it leaves the vm in RESIZING, the proposal is just to change that state back to ACTIVE | 19:45 |
| markwash | sandywalsh: I was just wondering about whether or not a schedule_prep_resize failure needed a notification like schedule_run_instance currently has? | 19:45 |
| *** ewanmellor has joined #openstack-dev | 19:46 | |
| mtaylor | termie: I have no idea why that didn't merge cleanly - I'm going to rebase and push again | 19:46 |
| sandywalsh | I think it should, otherwise the user will continue to be billed | 19:46 |
| *** jog0 has left #openstack-dev | 19:47 | |
| danwent | blamar: ok, so i will switch log statement to include the exception, and remove the NOTE? Agree with comment that unit tests are right away to enforce. Already have an internal todo there. | 19:47 |
| sandywalsh | littleidea, is the original instance still running? | 19:47 |
| littleidea | sandywalsh: yes | 19:47 |
| littleidea | sandywalsh: unchanged | 19:47 |
| *** dwalleck has quit IRC | 19:47 | |
| sandywalsh | littleidea, cool ... should be fine to set to active, the notification is a nice-to-have | 19:47 |
| ewanmellor | Hi all -- Lloyd promoted a separate #openstack-event IRC channel for today's global hack-in. There's hardly anyone there though. Are you all using this channel? | 19:48 |
| sandywalsh | littleidea, so long some error is reported along the way | 19:48 |
| *** jog0_ has joined #openstack-dev | 19:48 | |
| blamar | danwent: many thanks, sounds great | 19:48 |
| *** dwalleck has joined #openstack-dev | 19:48 | |
| *** adjohn has quit IRC | 19:49 | |
| littleidea | markwash, sandywalsh: is this the notifier code then? just swap the run_instance to resize? https://gist.github.com/1952611 | 19:49 |
| sandywalsh | littleidea, yup | 19:50 |
| danwent | blamar: seems like LOG.exception() automatically logs the last exception without it neededing to be passed in explicitly? | 19:50 |
| danwent | is that correct? | 19:50 |
| *** mattray has joined #openstack-dev | 19:50 | |
| blamar | danwent: correct | 19:50 |
| blamar | danwent: You can just pass the existing message s/warn/exception | 19:50 |
| danwent | blamar: yup, thanks. | 19:50 |
| termie | mtaylor: restamp | 19:50 |
| termie | 'd | 19:50 |
| *** jog0_ has quit IRC | 19:52 | |
| *** novas0x2a|laptop has quit IRC | 19:53 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:53 | |
| *** dolphm has joined #openstack-dev | 19:54 | |
| eglynn | dwalleck: hola | 19:55 |
| sandywalsh | justinsb, ping? | 19:56 |
| dwalleck | eglynn: howdy! | 19:56 |
| eglynn | dwalleck: I've got a couple of patches I'd like to get into tempest if you have a moment to review them? | 19:56 |
| dwalleck | Sorry, bouncing between meetings | 19:56 |
| eglynn | https://review.openstack.org/4746 & https://review.openstack.org/4755 | 19:56 |
| eglynn | dwalleck: no rush | 19:56 |
| dwalleck | eglynn: Sure, let me look | 19:56 |
| eglynn | dwalleck: I'm interested in including tempest in the Fedora17/openstack test day next week | 19:57 |
| eglynn | dwalleck: so obviously eager to have it working smoothly on Fedora in advance | 19:57 |
| dwalleck | eglynn: Awesome! I'll do what I can to make that possible | 19:57 |
| eglynn | dwalleck: cool, thanks! | 19:58 |
| *** jog0 has joined #openstack-dev | 19:59 | |
| *** bengrue has joined #openstack-dev | 19:59 | |
| *** dwalleck has quit IRC | 20:01 | |
| *** jog0 has left #openstack-dev | 20:01 | |
| *** aweiss has joined #openstack-dev | 20:01 | |
| *** jog0 has joined #openstack-dev | 20:01 | |
| *** jdg has joined #openstack-dev | 20:02 | |
| vishy | if you are participating in the hack-in join #openstack-event !!!! | 20:02 |
| *** hub_cap has quit IRC | 20:06 | |
| *** Mandell has joined #openstack-dev | 20:06 | |
| *** Ryan_Lane has quit IRC | 20:07 | |
| *** anotherjesse1 has joined #openstack-dev | 20:09 | |
| *** hub_cap has joined #openstack-dev | 20:09 | |
| *** mdomsch_ has joined #openstack-dev | 20:15 | |
| *** dwalleck has joined #openstack-dev | 20:15 | |
| *** mdomsch has quit IRC | 20:18 | |
| *** novas0x2a|laptop has quit IRC | 20:19 | |
| *** nati_ueno has quit IRC | 20:21 | |
| *** Mandell has quit IRC | 20:21 | |
| *** asdfasdf has joined #openstack-dev | 20:21 | |
| *** dubsquared has quit IRC | 20:21 | |
| *** novas0x2a|laptop has joined #openstack-dev | 20:21 | |
| *** Ryan_Lane has joined #openstack-dev | 20:23 | |
| *** bencherian has quit IRC | 20:27 | |
| *** dwalleck has quit IRC | 20:29 | |
| *** bencherian has joined #openstack-dev | 20:30 | |
| *** Mandell has joined #openstack-dev | 20:31 | |
| *** vincentricci has left #openstack-dev | 20:32 | |
| *** dwalleck has joined #openstack-dev | 20:34 | |
| *** dwalleck has quit IRC | 20:39 | |
| anotherjesse1 | bcwaldon: thanks for pointing out all of my little failures | 20:42 |
| westmaas | markwash: vv | 20:42 |
| bcwaldon | anotherjesse1: that's why I'm here | 20:42 |
| anotherjesse1 | bcwaldon does code review and life review… and votes −2 | 20:42 |
| bcwaldon | anotherjesse1: -2 for you | 20:42 |
| westmaas | markwash: https://review.openstack.org/#change,4675 | 20:44 |
| *** adjohn has joined #openstack-dev | 20:44 | |
| *** dprince has quit IRC | 20:49 | |
| bcwaldon | anotherjesse1: review time! | 20:49 |
| *** adiantum has joined #openstack-dev | 20:50 | |
| adam_g | is all of this service token / service account work targetted toward essex or folsom? | 20:51 |
| anotherjesse1 | adam_g: essex - since it is already supported in the middleware - just painfully (eg, lots of bugs) | 20:52 |
| anotherjesse1 | bcwaldon: I think termie's comment still holds - although if token validation fails the app shouldn't be called — it would be good to remove auth headers just in case? | 20:52 |
| bcwaldon | anotherjesse1: sure, we can do that | 20:54 |
| bcwaldon | anotherjesse1: do you want to do it, or shall I? | 20:54 |
| *** aweiss has quit IRC | 20:55 | |
| anotherjesse1 | bcwaldon: take it! | 20:56 |
| bcwaldon | anotherjesse1: kk | 20:56 |
| anotherjesse1 | bcwaldon: I have at least 4 patches you will need to make | 20:56 |
| bcwaldon | anotherjesse1: :( | 20:56 |
| bcwaldon | anotherjesse1: I'm not sure termie's comment is true anymore | 20:57 |
| bcwaldon | anotherjesse1: we *always* set all of the auth headers to a specific value | 20:57 |
| termie | bcwaldon: y'all forgot a lot of periods | 20:58 |
| termie | like, all of themn | 20:59 |
| anotherjesse1 | bcwaldon: are you sure? what if delayed auth is set to true, and the user sends an invalid token but HTTP_X_ROLE is set | 20:59 |
| bcwaldon | anotherjesse1: good point | 20:59 |
| termie | BOOM | 20:59 |
| anotherjesse1 | bcwaldon: −2 for you today | 20:59 |
| bcwaldon | termie: I dont want to add periods in log lines | 21:00 |
| termie | bcwaldon: the docs | 21:00 |
| termie | tings | 21:00 |
| termie | string | 21:00 |
| bcwaldon | termie: yes, on it | 21:00 |
| *** troytoman-away has quit IRC | 21:02 | |
| *** zzed has quit IRC | 21:02 | |
| *** troytoman-away has joined #openstack-dev | 21:02 | |
| *** clayg has quit IRC | 21:02 | |
| *** 13WAAGXMU is now known as clayg | 21:02 | |
| termie | hrmph, if you push a new thing then my comments on this one will be hidden :/ | 21:03 |
| anotherjesse1 | termie: yes, but the usually folks go back and look at comments on older versions | 21:04 |
| blamar | bcwaldon: | 21:05 |
| bcwaldon | blamar: | 21:05 |
| vishy | mtaylor / jeblair : either of you have access to the thing generating the json for http://wiki.openstack.org/bugstats/ | 21:05 |
| vishy | i would like to switch it to today if possible | 21:05 |
| blamar | bcwaldon: should 'glance add id="asdf" location="http://asdf"' be valid? | 21:05 |
| bcwaldon | blamar: 'id' will be interpreted as a property | 21:05 |
| bcwaldon | blamar: so yes | 21:05 |
| blamar | bcwaldon: why can't it just set the id? | 21:06 |
| bcwaldon | blamar: it could | 21:06 |
| bcwaldon | blamar: we just dont support it | 21:06 |
| bcwaldon | blamar: I wouldn't be against it | 21:06 |
| *** Mandell has quit IRC | 21:06 | |
| blamar | bcwaldon: k, the comments of image_add() indicate that id= will set the image id, unless I'm reading them wrong | 21:07 |
| *** sandywalsh has quit IRC | 21:07 | |
| blamar | gonna update it to work | 21:08 |
| *** mnewby has joined #openstack-dev | 21:08 | |
| *** Mandell has joined #openstack-dev | 21:08 | |
| *** AlanClark has quit IRC | 21:09 | |
| termie | ayoung: did you see my review of your patch? | 21:11 |
| ayoung | termie, yes. I've made the changes, but have not resubmitted yet | 21:12 |
| termie | ayoung: groovy | 21:12 |
| termie | i think if you rebase on my change you can just make it as a dependeny | 21:12 |
| termie | you can do git review -d 4659 to get my change as a branch | 21:12 |
| termie | and then rebase your change on it, when you git review again it will make it as a dependency | 21:13 |
| ayoung | termie, probably, or I can wait until yours get submitted...I can bug someone to review it | 21:13 |
| ayoung | I made my change on a branch that was already on top of your changes | 21:13 |
| *** stuntmachine has quit IRC | 21:14 | |
| *** vincentricci has joined #openstack-dev | 21:14 | |
| ayoung | termie, I also submitted a patch on top of your https://review.openstack.org/#change,4762 which fixed the live version of the LDAP tests | 21:14 |
| ayoung | the unit tests are OK, but the fakeldap implementation doesn't make me trust code the way running it against a live server does.... | 21:15 |
| markwash | anotherjesse1: hey, I was just taking a look at the new auth_token middleware you're propping | 21:15 |
| termie | ayoung: i don't see any other patches by you | 21:15 |
| markwash | anotherjesse1: btw this code looks about 1 million times better than the previous code | 21:15 |
| termie | ayoung: https://review.openstack.org/#q,status:open+project:openstack/keystone,n,z | 21:15 |
| ayoung | https://review.openstack.org/#change,4762 patch set 3? | 21:16 |
| ayoung | termie, no, to your review | 21:16 |
| markwash | anotherjesse1: one issue that I will end up having with this change (100% on my side to deal with though!) is | 21:16 |
| termie | ayoung: oh, you're not supposed to do that | 21:16 |
| ayoung | I didn't change any of the files you modified, just added to the commit so it didn't break the live tests | 21:16 |
| termie | ayoung: i did it giving you a day's warning before on the ldap stuff | 21:16 |
| ayoung | ah...sorry | 21:17 |
| termie | just make a new branch based on the review and add things with it as a dependency | 21:17 |
| termie | or ask the person first | 21:17 |
| markwash | anotherjesse1: currently rackspace auth doesn't scope tokens the same way keystone does, so I have a custom patch against it I will probably have to rewrite | 21:18 |
| markwash | anotherjesse1: I should really look at seeing if some form of my custom patch could land in trunk as a measure of mercy :-) | 21:18 |
| ayoung | termie, if you would rather revert it, go ahead, and I will submit it as a separate patch. | 21:19 |
| termie | ayoung: i don't think it is worth the effort in this case, but please don't do that again unless i'm expecting it | 21:20 |
| *** zzed has joined #openstack-dev | 21:20 | |
| ayoung | termie, Sure. I'm still learning the work flow patterns here. Its less of an issue when you are attaching a patch to an email on a mailing list. | 21:21 |
| anotherjesse1 | markwash: now that the auth_token is cleaner perhaps ... | 21:21 |
| *** aweiss has joined #openstack-dev | 21:22 | |
| anotherjesse1 | markwash: the spec does say how roles are scoped - just that they are returned when you validate a token - surprised they had to do something different | 21:22 |
| *** aweiss has left #openstack-dev | 21:23 | |
| termie | s/does/does not/ | 21:23 |
| termie | ? | 21:23 |
| markwash | anotherjesse1: yeah I think that once rax-auth adds tenant-ids to roles, we might have convergence again (at least if I understand correctly) | 21:23 |
| markwash | anotherjesse1: but until then I have to patch it to pass in ?belongsTo=<tenant-id> in the validate request | 21:24 |
| termie | markwash: we don't return tenant-ids with roles | 21:24 |
| anotherjesse1 | markwash: I think belongsTo is part of core - but we don't do it because then the middleware would have to know too much about url structure | 21:24 |
| markwash | anotherjesse1: really you guys just have no need for it--you can auth without specifying a tenant because your token is only associated with one tenant id | 21:25 |
| termie | markwash: not exactly, no | 21:25 |
| * markwash is now sure he's confused :-) | 21:25 | |
| termie | markwash: or rather, the token is yes, but you get that from authing with a tenant | 21:26 |
| termie | markwash: do you mean validate? | 21:26 |
| termie | markwash: i think you mean validate | 21:26 |
| markwash | termie: yes, validating | 21:26 |
| termie | auth == authenticate | 21:26 |
| termie | kk, then yes | 21:26 |
| anotherjesse1 | markwash: for instance glance doesn't put the tenant in the path iirc | 21:26 |
| markwash | anotherjesse1: precisely, that's what my patch is geared towards | 21:27 |
| markwash | anotherjesse1: I guess I also have a patch to glance to pass the tenant-id around even when its using token authentication | 21:27 |
| markwash | so maybe I'm a hopeless case | 21:27 |
| * markwash curses the irrelevance of a divergent rackspace auth | 21:27 | |
| markwash | well, hopefully auth_token will settle back down after this rewrite so I can revisit the patch less frequently | 21:28 |
| anotherjesse1 | markwash: hopefully the code makes it easier to patch even if you need to | 21:29 |
| anotherjesse1 | my understanding is that even using the middleware is temporary perhaps - until glance + repose works | 21:29 |
| markwash | anotherjesse1: hmm, interesting. . . | 21:29 |
| markwash | anotherjesse1: maybe I can help push that along then | 21:30 |
| markwash | anotherjesse1: I think the only thing stopping glance and repose from working together is the fact that glance client doesn't pass around the tenant id | 21:30 |
| markwash | anotherjesse1: well and any issues that repose might have with working with keystone itself | 21:30 |
| anotherjesse1 | markwash: feels like too late in the essex cycle to fix that :( | 21:31 |
| *** AZ_ has joined #openstack-dev | 21:31 | |
| markwash | anotherjesse1: sure, but if its possible in folsom, that's still better than I had hoped | 21:31 |
| anotherjesse1 | don't know if I like tenant_id in the url - but I don't like that services are different | 21:31 |
| markwash | anotherjesse1: right now I'm getting it from headers | 21:32 |
| AZ_ | -? | 21:32 |
| anotherjesse1 | markwash / westmaas - a show of support (review +1 or +2) from your team would be great before merge | 21:32 |
| markwash | anotherjesse1: I'll jump on | 21:33 |
| anotherjesse1 | heckj: do you want to update the middleware_architecture.rst | 21:33 |
| *** hub_cap has joined #openstack-dev | 21:34 | |
| *** hub_cap has quit IRC | 21:35 | |
| *** hub_cap has joined #openstack-dev | 21:36 | |
| heckj | anotherjesse1: yep, I've got it. | 21:37 |
| openstackgerrit | Verification of a change to openstack/nova failed: Adds soft-reboot support to libvirt https://review.openstack.org/4747 | 21:37 |
| mtaylor | vishy: I do not - but now that you're pointing it out - I'd really like to have access to it. :) | 21:37 |
| *** ncode has joined #openstack-dev | 21:37 | |
| *** zigo has quit IRC | 21:37 | |
| *** ncode has quit IRC | 21:37 | |
| *** ncode has joined #openstack-dev | 21:37 | |
| westmaas | thx for checking in anotherjesse1 | 21:38 |
| vishy | mtaylor: oh, i thought you had access to the wiki server | 21:39 |
| vishy | mtaylor: soren seems to | 21:40 |
| *** jmckenty has joined #openstack-dev | 21:40 | |
| vishy | mtaylor: or maybe that is the docs server | 21:40 |
| dolphm | bug 944373 | 21:40 |
| uvirtbot` | Launchpad bug 944373 in devstack "devstack should use nova-rootwrap instead of sudoers" [Undecided,New] https://launchpad.net/bugs/944373 | 21:40 |
| blamar | so I'm not the only one who gets openstack launchpad emails delayed like 30 minutes right? | 21:42 |
| Kiall | blamar nope .. not the only one at all | 21:42 |
| blamar | good god it's impossible to converse via carrier pigeon like this | 21:43 |
| novas0x2a|laptop | does rajalakshmi ganesan hang out on irc? | 21:44 |
| *** berendt has quit IRC | 21:47 | |
| annegentle | vishy: I have access to the wiki server, what do you need? | 21:49 |
| anotherjesse1 | mtaylor: do we know why the email is so slow? | 21:52 |
| *** shevek_ has joined #openstack-dev | 21:53 | |
| mtaylor | vishy: oh, I have access to the wiki server | 21:53 |
| mtaylor | I just don'tknow what creates that json | 21:53 |
| mtaylor | anotherjesse1: the email from? | 21:53 |
| heckj | zul, adam_g: ping | 21:54 |
| anotherjesse1 | mtaylor: se complaints about slowness of openstack email list above | 21:55 |
| anotherjesse1 | do we know who to talk to? | 21:55 |
| *** clayg_ has joined #openstack-dev | 21:56 | |
| mtaylor | anotherjesse1: oh, that. that's launchpad | 21:56 |
| vishy | smoser: where would i file a bug about installing acpid by default in the cloud images | 21:56 |
| anotherjesse1 | as in - it is expected for email to take 30 minutes - no biggie? | 21:56 |
| *** sandywalsh has joined #openstack-dev | 21:56 | |
| smoser | ubuntu. | 21:56 |
| vishy | smoser: just on the general ubuntu project? | 21:56 |
| smoser | vishy, just against ubuntu. subscribe me. | 21:56 |
| smoser | yeah. tag with cloud-images, ec2-iamges. | 21:57 |
| *** AlanClark has joined #openstack-dev | 21:57 | |
| vishy | smoser: cool | 21:57 |
| mtaylor | anotherjesse1: well... no, I think that's a pile of ass ... :) | 21:57 |
| mtaylor | anotherjesse1: lemme see if I can find someone to punch | 21:58 |
| smoser | vishy, acpid is in the images | 21:58 |
| smoser | at least precise | 21:58 |
| *** anotherjesse has quit IRC | 21:58 | |
| vishy | smoser: oh? i just tried oneiric and i had to apt-get it | 21:58 |
| vishy | if it is in precise i guess that is ok | 21:58 |
| smoser | yeah. by happenstance it seems to be there in precise. | 21:58 |
| *** ncode has quit IRC | 21:59 | |
| anotherjesse1 | heckj: any clue why http://keystone.openstack.org/sourcecode/keystone.middleware.auth_token.html ended up empty? | 22:00 |
| mtaylor | anotherjesse1: there's a currently known issue that they're apparently really close to solving. the words "quadratic archive page writes" were used | 22:01 |
| *** AlanClark has quit IRC | 22:01 | |
| *** mdomsch_ has quit IRC | 22:01 | |
| *** flaviamissi has quit IRC | 22:02 | |
| anotherjesse1 | mtaylor: let them know we have quantum technology … "quantum archive page writes" might be faster | 22:02 |
| heckj | anotherjesse1: nope, hadn't seen that before. All the core bits are empty - I'll see if I can reproduce | 22:02 |
| *** maplebed has joined #openstack-dev | 22:02 | |
| heckj | (by core bits, I mean module entries) | 22:02 |
| jeblair | anotherjesse1: only if nobody looks at the archive page | 22:02 |
| anotherjesse1 | heckj: also - what do you think about a bug that our sphinx builds should have buildtime on the copyright footer | 22:02 |
| *** vincentricci has joined #openstack-dev | 22:02 | |
| anotherjesse1 | heckj: probably a bug for all docs | 22:03 |
| heckj | anotherjesse1: a build for all the docs - that would be excellent to have in all of them | 22:03 |
| heckj | probably linked to all the projects, single bug | 22:03 |
| *** hub-cap has joined #openstack-dev | 22:04 | |
| *** hub_cap has quit IRC | 22:04 | |
| *** hub-cap is now known as hub_cap | 22:04 | |
| mtaylor | anotherjesse1: ahahaha | 22:04 |
| *** Mkenneth has quit IRC | 22:05 | |
| *** Mkenneth has joined #openstack-dev | 22:06 | |
| anotherjesse1 | annegentle: would it be you that we work with to get the footers of keystone.openstack.org, glance.openstack.org, … to have buildtime / git rev in the footer? | 22:07 |
| annegentle | anotherjesse1: I do know how to do it (have to look again) | 22:08 |
| annegentle | anotherjesse1: actually sleepsonthefloor knows more than I | 22:08 |
| anotherjesse1 | annegentle: looks like layout.html | 22:10 |
| *** cjwilson has joined #openstack-dev | 22:11 | |
| *** dtroyer has quit IRC | 22:12 | |
| annegentle | anotherjesse1: yeah that sounds right, though theme.conf extends layout.html | 22:15 |
| openstackgerrit | Verification of a change to openstack/nova failed: Adds soft-reboot support to libvirt https://review.openstack.org/4747 | 22:17 |
| *** utlemming has quit IRC | 22:20 | |
| *** ncode has joined #openstack-dev | 22:21 | |
| anotherjesse1 | heckj / westmaas / markwash / jaypipes / termie: I just gave https://review.openstack.org/#change,4675 +2 - others want one last review? | 22:21 |
| anotherjesse1 | I see small tweaks to docs, but we can continue those afterward | 22:21 |
| termie | also +2d | 22:22 |
| *** asdfasdf has quit IRC | 22:22 | |
| * heckj lookin | 22:22 | |
| anotherjesse1 | heckj - nothing landed in keystone since patch 17 (or wherever you last reviewed) - so using the base patch feature is nice | 22:23 |
| heckj | anotherjesse1: what's the "base patch feature"? | 22:24 |
| *** Mandell has quit IRC | 22:24 | |
| *** andrewsben has quit IRC | 22:25 | |
| anotherjesse1 | heckj: click the "patch history" at the top of https://review.openstack.org/#patch,sidebyside,4675,20,keystone/middleware/auth_token.py | 22:25 |
| heckj | OOoohh! | 22:25 |
| anotherjesse1 | then you can see old version -> change from base to 18 or whatever | 22:25 |
| heckj | never noticed the disclosure triangle there! | 22:25 |
| *** andrewsmedina has quit IRC | 22:25 | |
| anotherjesse1 | you can on the main page too - "old version history" - I wish it put that in the url | 22:25 |
| *** Mandell has joined #openstack-dev | 22:27 | |
| heckj | yeah, no kidding! | 22:28 |
| openstackgerrit | Verification of a change to openstack/nova failed: Raise 409 when rescuing instance in RESCUE mode https://review.openstack.org/4599 | 22:28 |
| heckj | anotherjesse1, termie: +2'd | 22:29 |
| anotherjesse1 | approve it | 22:30 |
| heckj | bit flipped! | 22:30 |
| vishy | mtaylor: so you can get to the wiki? So does that mean you can change the bug display to show today? | 22:31 |
| vishy | mtaylor: or at least create a new page that shows current nova bug graph? | 22:31 |
| heckj | Wow - I really wish it enabled those between patch-set diffs as a bookmarkable item | 22:32 |
| termie | auth_token_light | 22:32 |
| termie | heckj: yeah | 22:32 |
| termie | heckj: i use the key commands and it keeps the patch set base between moves | 22:32 |
| heckj | that's what I've done as well, but I didn't know you could even look between diffs. I'd been going back and forth between two open windows to do ti | 22:33 |
| termie | DOUBLE PRODUCTIVITY | 22:33 |
| termie | i'll be expecting twice as many reviews out of you now | 22:33 |
| heckj | oh yeah | 22:33 |
| *** lts has quit IRC | 22:33 | |
| mnewby | termie: Are you the keystone guru? | 22:34 |
| termie | mnewby: depends if you are asking me questions i don't want to answer | 22:35 |
| mnewby | heh | 22:35 |
| termie | mnewby: in those cases heckj is | 22:35 |
| termie | mnewby: we let anotherjess hang out with us too | 22:35 |
| mnewby | Keystone uses POSTED json for auth calls? | 22:35 |
| termie | mnewby: yes, now try a harder one | 22:36 |
| anotherjesse1 | w00h - it just landed | 22:36 |
| mnewby | Is there any interest in supporting legacy header-based auth, ala Swift? | 22:36 |
| anotherjesse1 | time to re-stack all my boxes | 22:36 |
| mnewby | Or is the idea to just update all clients to use Keystone? | 22:36 |
| heckj | termie <-- definitely the guru, especiallty when he doesn't want to answer | 22:36 |
| termie | mnewby: i think everybody is supposed to use keystone, there is appropriate middelware for everything to do that now | 22:36 |
| termie | mnewby: swift probably still accepts things via headers and makes keystone requests | 22:36 |
| mnewby | Hmmm... | 22:37 |
| mnewby | If there is swift middleware that does that, I haven't seen it. | 22:37 |
| mnewby | Is it part of swift or keystone? | 22:37 |
| termie | mnewby: look in the keystone's middleware | 22:37 |
| mnewby | I know there is swift-supporting middleware in keystone. | 22:37 |
| heckj | mnewby: it's been a while since I looked at the swift auth pieces, but I know they were doing some things locally and acting as a proxy to enable higher volume of requests - i.e. permission and token lookups. I think most of that moved into reasonable middleware with caching support though. | 22:37 |
| mnewby | auth_token and swift_auth | 22:37 |
| anotherjesse1 | mnewby: chmouel knows a lot about swift auth | 22:37 |
| mnewby | So I should pester him? | 22:38 |
| termie | mnewby: well, i'd probably read the middleware for swift in keystone first | 22:39 |
| termie | mnewby: if you havne't | 22:39 |
| mnewby | I guess that it makes sense to just add middleware support in swift to translate a header-based call to keystone's json call. | 22:39 |
| mnewby | auth_token handles already-granted tokens, and swift_auth performs authorization on keystone-authenticated request. I just hadn't seen anything that actually granted tokens and presumed that a call to keystone was required. | 22:40 |
| anotherjesse1 | mnewby: if chmouel isn't around now email the list as he's been focused on making that all work | 22:40 |
| mnewby | Ok | 22:40 |
| mnewby | danke | 22:40 |
| *** ncode has quit IRC | 22:42 | |
| vishy | mnewby: the swift client has support for new protocol | 22:43 |
| mnewby | vishy: I know the openstack stuff supports it. No guarantees that all client software in use does, though. | 22:44 |
| mnewby | I think there's a good case for continuing to support the header-based auth via translation middleware. | 22:44 |
| mnewby | Not necessarily in openstack proper, but I think that's the approach we'll take at internap/voxel. | 22:45 |
| anotherjesse1 | mnewby: so - the auth_token middleware does the following: | 22:46 |
| mnewby | Plus, it makes keystone adoption easier if the the switchover only has to be made server-side. | 22:46 |
| vishy | mnewby: yes makes sense. The auth_token middleware once supported v1 auth | 22:46 |
| *** dolphm has quit IRC | 22:47 | |
| anotherjesse1 | mnewby: gets the token from herehttps://github.com/openstack/keystone/blob/master/keystone/middleware/auth_token.py#L212 | 22:47 |
| anotherjesse1 | mnewby: so it looks for both X-Auth-Token and X-Storage-Token | 22:47 |
| mnewby | Right. | 22:47 |
| mnewby | But there's nothing in there about _granting_ a token. | 22:47 |
| anotherjesse1 | mnewby: the question should probably be: can swift clients get a X-Storage-Token - … exactly | 22:48 |
| anotherjesse1 | and for that I don't think keystone supports that | 22:48 |
| mnewby | At least not granting a token via the header-based 1.0 call. | 22:48 |
| *** cp16net has quit IRC | 22:48 | |
| anotherjesse1 | mnewby: what clients exist that talk swift? would it be easier to us to go update them rather than support a deprecated prototcol? | 22:48 |
| mnewby | cyberduck? | 22:49 |
| zaitcev | There's a ton on obscure crap nobody uses, too. Like iwhd's CloudFiles back-end. | 22:49 |
| mnewby | And forcing clients to upgrade vs simply maintaining support for an old protocol seems like a no-brainer to me. | 22:49 |
| mnewby | But anyway, my question was whether there was or should be support for v1 header-based auth in keystone. | 22:50 |
| mnewby | I'm hearing that the answer is no, which is fine. I can always write the middleware myself. | 22:50 |
| vishy | yes they can | 22:51 |
| vishy | well there is currently middleware in swift to give the token back | 22:51 |
| mnewby | But not to issue the token. | 22:51 |
| mnewby | So it's a bit piecemeal. | 22:51 |
| vishy | yes to issue as well | 22:51 |
| vishy | that is how swauth works | 22:51 |
| mnewby | Oh, sorry. | 22:51 |
| mnewby | I meant for keystone. | 22:52 |
| vishy | mnewby: but legacy clients don't talk to keystone | 22:52 |
| mnewby | Yes, of course swift supports v1.0 auth. | 22:52 |
| mnewby | But if I were to want to standardize on keystone, I wouldn't want a separate auth mechanism for swift. | 22:52 |
| vishy | mnewby: well... | 22:52 |
| vishy | all of the instructions for third party software | 22:52 |
| vishy | (which by the way is almost nothing) | 22:53 |
| anotherjesse1 | mnewby: there was never a pre-2.0 auth in openstack - it was rax auth | 22:53 |
| vishy | have you use swift as the auth endpoint and use swauth | 22:53 |
| anotherjesse1 | mnewby: I agree with you that it is painful | 22:53 |
| anotherjesse1 | it is painful both ways | 22:53 |
| vishy | most clients are hard-coded to use the rax auth system | 22:53 |
| mnewby | Ah, ok. | 22:53 |
| vishy | so if you are going to put in compatibility for old clients | 22:53 |
| *** markvoelker has quit IRC | 22:54 | |
| vishy | you 'could' put it in swift as well | 22:54 |
| vishy | smoser: ping | 22:57 |
| vishy | mnewby: but i suppose a compatibility middleware in keystone would also make sense | 22:57 |
| bcwaldon | jeblair: is horizon gated on devstack? | 22:58 |
| mnewby | vishy: I'd be happy to contribute compatibility middleware if you think others would find it useful. | 22:59 |
| anotherjesse1 | bcwaldon: there aren't any tests to run :( | 22:59 |
| bcwaldon | anotherjesse1: doesn't matter | 22:59 |
| bcwaldon | anotherjesse1: we need to make sure devstack works with horizon | 22:59 |
| jeblair | bcwaldon: no, is it ready to be? | 22:59 |
| bcwaldon | anotherjesse1: we broke devstack the other day with horizon changes | 22:59 |
| anotherjesse1 | bcwaldon: we run this https://github.com/anotherjesse/potato/blob/master/launch.py against our daily deploy | 22:59 |
| bcwaldon | which broke all gating | 22:59 |
| anotherjesse1 | to figure out when we are able to login | 23:00 |
| anotherjesse1 | note - the test is REALLY ghetto | 23:00 |
| anotherjesse1 | err | 23:00 |
| *** Mkenneth has quit IRC | 23:00 | |
| mnewby | Another swift/keystone question - swift presumably uses a token-based auth system to minimize load on the auth system. | 23:00 |
| jeblair | bcwaldon: how did gating break if horizon isn't used? | 23:00 |
| *** Mkenneth has joined #openstack-dev | 23:00 | |
| vishy | jeblair: yeah we might want to run it | 23:01 |
| bcwaldon | horizon is set up in devstack | 23:01 |
| vishy | jeblair: it broke the install | 23:01 |
| anotherjesse1 | bcwaldon: actually we test https://github.com/cloudbuilders/stillalive/blob/master/check.py | 23:01 |
| bcwaldon | maybe we should just disable horizon in our gating hob? | 23:01 |
| bcwaldon | job* | 23:01 |
| *** bsza has quit IRC | 23:01 | |
| jeblair | ENABLED_SERVICES=g-api,g-reg,key,n-api,n-crt,n-obj,n-cpu,n-net,n-vol,n-sch,mysql,rabbit | 23:02 |
| bcwaldon | problem solved | 23:02 |
| jeblair | bcwaldon: that's the current state | 23:02 |
| jeblair | so i'm not sure how something was broken | 23:02 |
| bcwaldon | I assumed we were gating on it when I saw 'devstack is broken' the other day | 23:02 |
| bcwaldon | I guess gating never broke | 23:02 |
| bcwaldon | but it would still be good to get coverage on making sure devstack and horizon work together | 23:03 |
| notmyname | mnewby: swift uses a token based auth system because it was written at rackspace and rax auth is token based :-) | 23:03 |
| jeblair | ok. good; just wanted to make sure there wasn't something in the gating test design that was borked. absolutely agree with you, we should test horizon to make sure that doesn't happen in the future | 23:03 |
| anotherjesse1 | bcwaldon: the test to clone/setup horizon seems to be correct https://github.com/openstack-dev/devstack/blob/master/stack.sh#L587 | 23:03 |
| *** x86brandon has quit IRC | 23:03 | |
| bcwaldon | it is now | 23:03 |
| bcwaldon | it wasnt yesterday | 23:04 |
| anotherjesse1 | ah | 23:04 |
| *** apevec has quit IRC | 23:04 | |
| bcwaldon | I'm asking to add devstack gating to horizon | 23:04 |
| jeblair | bcwaldon: do you think if we just add it to enabled services, it will work now? | 23:04 |
| bcwaldon | yes | 23:04 |
| jeblair | is there an exercise test for it? | 23:05 |
| bcwaldon | no, but setting it up is the minimum exercise | 23:05 |
| bcwaldon | we had path issues the last time it broke | 23:05 |
| anotherjesse1 | bcwaldon: https://github.com/cloudbuilders/stillalive/blob/master/check.py is the closest thing we have | 23:05 |
| anotherjesse1 | jeblair: ^ | 23:05 |
| bcwaldon | sure, we could use that as our 'exercise' | 23:05 |
| jeblair | will stack.sh barf if it fails to set it up before launching it in screen? | 23:05 |
| bcwaldon | yes | 23:05 |
| bcwaldon | well, for certain things | 23:05 |
| *** torgomatic has joined #openstack-dev | 23:06 | |
| jeblair | ok, good, because i don't think a failure to start in screen is detected by stack.sh, which is needed in order for the job to fail | 23:06 |
| *** hub_cap has quit IRC | 23:06 | |
| jeblair | most failures are caught by exercise | 23:06 |
| jeblair | so, iow, we don't actually notice that, say, keystone doesn't start. we just notice that the exercise tests don't work. | 23:06 |
| jeblair | bcwaldon, anotherjesse1: anything other than adding "horizon" to ENABLED_SERVICES needed? | 23:08 |
| termie | jeblair: stack.sh checks for thigns starting | 23:09 |
| vishy | jeblair, anotherjesse1, bcwaldon: but that 'exercise' isn't in sh! | 23:09 |
| termie | jeblair: if keystone doesn't start it will fail before the exercises | 23:09 |
| jeblair | termie: cool | 23:09 |
| vishy | who wants to rewrite it with curl? | 23:09 |
| *** anotherjesse1 has quit IRC | 23:10 | |
| mtaylor | vishy: I can totally change the page - I just don't know what creates the graph - do you have that? | 23:11 |
| jeblair | bcwaldon: https://review.openstack.org/4777 | 23:11 |
| vishy | mtaylor: I don't know where it comes from, but it appears to be getting a local json file. I assumed there is a cron job that is generating it ? | 23:11 |
| torgomatic | any admins around who can help with an openstack-cla membership? | 23:11 |
| *** Ryan_Lane has quit IRC | 23:12 | |
| bcwaldon | torgomatic: sure, what's the name you signed the cla with? | 23:12 |
| *** Ryan_Lane has joined #openstack-dev | 23:12 | |
| mtaylor | vishy: lemme look | 23:13 |
| *** dolphm has joined #openstack-dev | 23:13 | |
| bcwaldon | torgomatic: have you proposed to add yourself to ~openstack-cla? | 23:13 |
| mtaylor | vishy: ooh - I may have found how to refresh the data | 23:14 |
| *** danwent has quit IRC | 23:15 | |
| bcwaldon | torgomatic: you need to request to join https://launchpad.net/~openstack-cla | 23:15 |
| bcwaldon | torgomatic: then I can approve you | 23:15 |
| mtaylor | vishy: I re-ran the refresh data command - the date is hardcoded into the header - do the graphs look more appropriate now? | 23:17 |
| torgomatic | bcwaldon: I'm in already; my coworker darrellb has an open request | 23:18 |
| bcwaldon | torgomatic: ah, ok | 23:19 |
| bcwaldon | torgomatic: looking | 23:19 |
| bcwaldon | torgomatic: approved | 23:19 |
| *** danwent has joined #openstack-dev | 23:19 | |
| *** flaviamissi has joined #openstack-dev | 23:20 | |
| *** roge has quit IRC | 23:20 | |
| *** Gordonz has quit IRC | 23:24 | |
| jeblair | mtaylor: https://review.openstack.org/4780 | 23:24 |
| jeblair | I believe that change will make it so that we are _fully_ gating the gate script itself | 23:25 |
| *** dolphm has quit IRC | 23:26 | |
| jeblair | then after that, let's add horizon: https://review.openstack.org/4781 | 23:26 |
| mtaylor | looking | 23:27 |
| mtaylor | jeblair: oh, good catch. | 23:28 |
| jeblair | mtaylor: do you think bash might have a problem with the running script changing under itself? | 23:28 |
| jeblair | mtaylor: i've seen that happen but don't fully understand the circumstances | 23:28 |
| mtaylor | jeblair: bash? no. jenkins execing of bash? don't know | 23:28 |
| mtaylor | I have no idea what happens to java process handles in case of exec | 23:28 |
| *** maplebed is now known as maplebedZzzzzz | 23:29 | |
| jeblair | oh, that will be fine, we're execing over bash, not over java | 23:29 |
| mtaylor | oh right. cool | 23:29 |
| mtaylor | looks good then | 23:29 |
| *** torgomatic has quit IRC | 23:29 | |
| *** maplebedZzzzzz is now known as maplebed | 23:29 | |
| *** torgomatic has joined #openstack-dev | 23:30 | |
| mtaylor | jeblair: have I mentioned that I think that the jenkins slave scripts should be in their own repo so that changes that have nothing to do with jenkins scripts don't gate on devstack ? | 23:30 |
| torgomatic | bcwaldon: he's in now; thanks | 23:30 |
| *** adiantum has quit IRC | 23:30 | |
| jeblair | mtaylor: reasonable! | 23:30 |
| *** pixelbeat has joined #openstack-dev | 23:32 | |
| jeblair | bcwaldon, termie: https://review.openstack.org/4781 | 23:33 |
| *** littleidea has quit IRC | 23:35 | |
| termie | jeblair: +1 | 23:36 |
| *** littleidea has joined #openstack-dev | 23:37 | |
| *** madhav-puri has joined #openstack-dev | 23:38 | |
| madhav-puri | leave | 23:39 |
| devcamcar | i'm about confused about the changes to console/proxy stuff in e4 | 23:39 |
| *** madhav-puri has quit IRC | 23:39 | |
| devcamcar | specifically, whats the role of nova-console? | 23:39 |
| *** madhav-puri has joined #openstack-dev | 23:40 | |
| *** madhav-puri is now known as madhav | 23:42 | |
| mnewby | devcamcar: Can I work on this? https://bugs.launchpad.net/nova/+bug/943029 | 23:45 |
| uvirtbot` | Launchpad bug 943029 in nova "fixed_ip_associate or FixedIP tables looks broken" [Undecided,In progress] | 23:45 |
| devcamcar | mnewby: yep, doesn't look like anyone is actively working on it | 23:46 |
| *** dtroyer has joined #openstack-dev | 23:46 | |
| *** x86brandon has joined #openstack-dev | 23:47 | |
| *** ncode has joined #openstack-dev | 23:50 | |
| *** ncode has joined #openstack-dev | 23:50 | |
| jeblair | bcwaldon, termie, heckj, mtaylor: the change merged, so the devstack vm gate now tests horizon as well, and changes to horizon are gated on the test. | 23:51 |
| bcwaldon | jakedahn: take note ^ | 23:51 |
| termie | jakedahn: NEVER AGAIN | 23:51 |
| jeblair | however, more exercising of horizon specifically may still be in order. :) | 23:52 |
| heckj | jeblair: awesomesauce! | 23:54 |
| ohnoimdead | nice! | 23:55 |
| ohnoimdead | now we just need some selenium-ish end-to-end tests and we'll be aces | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!