| *** hugokuo has joined #openstack-dev | 00:05 | |
| *** mwhooker has joined #openstack-dev | 00:06 | |
| *** rbasak has joined #openstack-dev | 00:07 | |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: Adds network model and network info cache. https://review.openstack.org/2028 | 00:07 |
|---|---|---|
| *** jeffjapan has joined #openstack-dev | 00:15 | |
| openstackgerrit | Gabriel Hurley proposed a change to openstack/horizon: Reworks messages so they display if they have multiple tags applied. https://review.openstack.org/2162 | 00:16 |
| openstackjenkins | Project nova-docs build #1,738: SUCCESS in 3 min 7 sec: https://jenkins.openstack.org/job/nova-docs/1738/ | 00:18 |
| openstackjenkins | Brad Hall: Fix for bug 901459 | 00:19 |
| uvirtbot | Launchpad bug 901459 in nova "QuantumManager#get_all_networks uninitalized variable" [Undecided,Fix committed] https://launchpad.net/bugs/901459 | 00:19 |
| *** nati2_ has joined #openstack-dev | 00:31 | |
| *** hugokuo has left #openstack-dev | 00:33 | |
| *** nati2 has quit IRC | 00:35 | |
| *** rods has joined #openstack-dev | 00:40 | |
| *** dragondm_ has quit IRC | 00:47 | |
| *** dragondm has quit IRC | 00:47 | |
| *** jmckenty_ has quit IRC | 00:52 | |
| *** adjohn has quit IRC | 00:56 | |
| smoser | ok.. i said earlier that euca2ools was moaning at me becauase of the site-package that gets installed in devstack (due to nova's 'setup.py develop') | 01:02 |
| smoser | i'm python-newbie enough not to know how to easily tell pytnon to ignore those directories for runs of euca2oosl. | 01:02 |
| smoser | mucking with PYTHONPATH didnt seem to work. | 01:02 |
| mtaylor | smoser: which directories? | 01:03 |
| smoser | python -c 'import boto.auth_handler' | 01:03 |
| smoser | is what will fail | 01:03 |
| mtaylor | because nova installed something conflicting/ | 01:03 |
| mtaylor | ? | 01:03 |
| smoser | strace shows: http://paste.ubuntu.com/763338/ | 01:04 |
| smoser | right. | 01:04 |
| smoser | nova gets 1.9b and euca2ools needs 2.0 (which system has) | 01:04 |
| mtaylor | ok - bear with me here... system is possibly slightly vague, so I'm going to ask a few more questions | 01:04 |
| mtaylor | did you run devstack in a vm? | 01:04 |
| smoser | so default run of python gets boto from /usr/local/lib/python2.7/dist-packages/ | 01:04 |
| *** bsza has joined #openstack-dev | 01:04 | |
| smoser | "vm" == instance inside openstack | 01:05 |
| mtaylor | ok. so you ran it on its own machine | 01:05 |
| *** zaitcev has joined #openstack-dev | 01:05 | |
| smoser | yeah. | 01:05 |
| mtaylor | and then on that machine you are also trying to install euca2ools | 01:05 |
| mtaylor | I'm guessing that you're trying to install those via apt-get / | 01:06 |
| mtaylor | ? | 01:06 |
| smoser | well.. they're installed in an ubuntu image by default. | 01:06 |
| smoser | but yeah. | 01:06 |
| smoser | from apt | 01:06 |
| smoser | /usr/lib/python2.7/dist-packages/boto/ | 01:06 |
| mtaylor | ok. so - things installed in /usr/local/ are going to override things installed in /usr (which is where apt is going to install things) | 01:06 |
| smoser | right. and i want to tell python "ignore that /usr/local crap" | 01:07 |
| smoser | :) | 01:07 |
| mtaylor | you would have to do that by mucking with sys.path after python had started | 01:07 |
| mtaylor | you might be able to point python at a site.py file | 01:07 |
| smoser | devstack actually installes euca2ools also. | 01:08 |
| mtaylor | so in theory the devstack installed euca2ools | 01:08 |
| smoser | right.. i was really hopeing to do somethin with an environment variable. | 01:08 |
| mtaylor | might work where the system ones don't? | 01:08 |
| smoser | devstack installs euca2ools with apt | 01:08 |
| mtaylor | ah. funny | 01:08 |
| mtaylor | what happens if you pip install -U boto | 01:09 |
| mtaylor | does nova still work? | 01:09 |
| smoser | gets boto-2.1.1.tar.gz | 01:09 |
| mtaylor | (not saying it's the right answer overall, just curious with path holds the right answer) | 01:09 |
| smoser | we'll see if nova still works. | 01:09 |
| *** pixelbeat has quit IRC | 01:09 | |
| mtaylor | because if that works, then we might just need to bump the boto version in nova's tools/pip-requires | 01:10 |
| smoser | euca worlks. | 01:10 |
| smoser | works | 01:10 |
| *** adjohn has joined #openstack-dev | 01:10 | |
| smoser | nova-{cpu,net,api,scheduler} do not stacktrace on startup | 01:11 |
| *** _adjohn has joined #openstack-dev | 01:11 | |
| *** lloydde has quit IRC | 01:11 | |
| *** ben_duyujie has joined #openstack-dev | 01:12 | |
| mtaylor | that's a start | 01:12 |
| mtaylor | I'm running the unit tests here | 01:13 |
| mtaylor | smoser: are you on oneiric? | 01:13 |
| smoser | instance is oneiric, yes. | 01:13 |
| mtaylor | great. | 01:13 |
| openstackgerrit | Jenkins proposed a change to openstack/nova: Imported Translations from Launchpad https://review.openstack.org/2164 | 01:14 |
| *** adjohn has quit IRC | 01:15 | |
| *** _adjohn is now known as adjohn | 01:15 | |
| smoser | mtaylor, thanks for the work around for ht e moment. | 01:16 |
| mtaylor | yay! translations imports work! | 01:16 |
| *** darthflatus has joined #openstack-dev | 01:17 | |
| smoser | vladimir3p, around? | 01:19 |
| smoser | i found out what was breaking the euca2ools... sort of | 01:19 |
| smoser | something is setting USERNAME in the screen session, then when you '. openrc' you everything gets 'ubuntu' rather than 'demo' for NOVA_USERNAME, EC2_ACCESS_KEY and the like. | 01:20 |
| smoser | so just 'unset USERNAME' | 01:20 |
| smoser | then '. openrc' | 01:20 |
| smoser | and it is working for me. | 01:20 |
| *** darthflatus has left #openstack-dev | 01:21 | |
| smoser | if anotherjesse was around i'd let him know that. | 01:22 |
| mtaylor | smoser: file bug? | 01:22 |
| smoser | yeah. | 01:22 |
| mtaylor | smoser: https://bugs.launchpad.net/devstack | 01:23 |
| smoser | yeah. | 01:23 |
| smoser | you want a bug for the nova/boto issue above? | 01:23 |
| smoser | there can still be issues, but it happens that at least euca2ools is happeri with 2.1.1 than 1.9b | 01:24 |
| mtaylor | smoser: yeah. file a bug. I'll submit a patch and we'll see what people think about it | 01:30 |
| *** rohitk has quit IRC | 01:36 | |
| smoser | mtaylor, bug 901500 | 01:36 |
| uvirtbot | Launchpad bug 901500 in nova "nova setup installs boto-1.9b, which is incompatible with euca2ools" [Undecided,New] https://launchpad.net/bugs/901500 | 01:36 |
| *** jog0 has quit IRC | 01:39 | |
| openstackgerrit | Monty Taylor proposed a change to openstack/nova: Relaxes version constraint on boto. https://review.openstack.org/2165 | 01:41 |
| *** novas0x2a|laptop has quit IRC | 01:43 | |
| *** spiffxp has quit IRC | 01:47 | |
| *** beekhof__ has joined #openstack-dev | 01:49 | |
| *** beekhof_ has quit IRC | 01:49 | |
| *** reed_ has joined #openstack-dev | 01:50 | |
| *** dragondm has joined #openstack-dev | 01:52 | |
| vishy | smoser, mtaylor: we just came across that | 01:53 |
| vishy | there is another addition needed | 01:53 |
| *** reed has quit IRC | 01:53 | |
| vishy | have to remove nova-adminclient from pip requires as well | 01:53 |
| *** dragondm_ has joined #openstack-dev | 01:54 | |
| smoser | vishy, you have a suggested work around for comment 2 in bug 897164 | 01:55 |
| uvirtbot | Launchpad bug 897164 in nova "neither EC2 api nor openstack API expose the hostname" [Wishlist,Fix committed] https://launchpad.net/bugs/897164 | 01:55 |
| smoser | should i be able to use 'nova list' in devstack ? its seemingly not configured for me. | 01:56 |
| *** vladimir3p has quit IRC | 01:56 | |
| *** beekhof_ has joined #openstack-dev | 01:57 | |
| *** beekhof__ has quit IRC | 01:57 | |
| vishy | smoser: we recently had nova list and euca break | 01:57 |
| vishy | smoser: did you see my comment above that nova-adminclient relies on boto 1.9b as well | 01:58 |
| *** jdurgin has quit IRC | 01:58 | |
| *** bsza has quit IRC | 01:58 | |
| smoser | well, thats unrelated to my comment in that bug (comment 2) i think | 01:59 |
| smoser | the tools aren't broken, they work for me, i just get a hostname which i can't resolve rather than an IP address in 'describe-instances' | 01:59 |
| vishy | sure I just wasn't sure if you saw it | 02:00 |
| vishy | it was about the pip dependencies that you and mtaylor found | 02:00 |
| vishy | so workaround: patch euca2ools? | 02:00 |
| vishy | the api returns ip address as well, but apparently euca2ools doesn't display it | 02:01 |
| vishy | :( | 02:01 |
| vishy | we might have to back out that patch i guess | 02:01 |
| vishy | smoser: ^^ | 02:01 |
| smoser | hm.. | 02:01 |
| *** beekhof_ has quit IRC | 02:01 | |
| smoser | patch euca2ools isn't really an option | 02:02 |
| *** beekhof has joined #openstack-dev | 02:02 | |
| smoser | as that == "break euca2ools" | 02:02 |
| *** rohitk has joined #openstack-dev | 02:02 | |
| *** dragondm_ has quit IRC | 02:02 | |
| *** dragondm has quit IRC | 02:02 | |
| smoser | if there were a way to get at the dns server that could resolve that for me, i'd live with that. | 02:03 |
| smoser | but i'll also consider trying to get a 'euca-describe-instances --ip' flag upstream. | 02:03 |
| *** lloydde has joined #openstack-dev | 02:05 | |
| openstackgerrit | A change to openstack/nova has been rejected: Remove FixedIP<->Network FK https://review.openstack.org/2163 | 02:06 |
| openstackgerrit | A change to openstack/nova has been rejected: Relaxes version constraint on boto. https://review.openstack.org/2165 | 02:10 |
| *** maplebed has quit IRC | 02:13 | |
| *** lorin1 has joined #openstack-dev | 02:17 | |
| vishy | smoser: there is some effort going into an automatic dns server | 02:20 |
| *** adjohn has quit IRC | 02:20 | |
| vishy | smoser: dnsmasq can resolve it! | 02:20 |
| smoser | where is it running ? | 02:20 |
| vishy | you should be able to dig against the network ip | 02:20 |
| smoser | can i just 'host server-1 <...>' | 02:21 |
| smoser | (i tried that with 127.0.0.1 no luck) | 02:21 |
| vishy | are you running devstack | 02:22 |
| vishy | do it against the ip on br100 | 02:22 |
| vishy | ps ax | grep dnsmasq will tell you | 02:22 |
| vishy | which ip it is listening on | 02:22 |
| *** lorin1 has quit IRC | 02:23 | |
| *** zns has joined #openstack-dev | 02:23 | |
| *** sleepsonthefloo has quit IRC | 02:23 | |
| *** novas0x2a|laptop has joined #openstack-dev | 02:23 | |
| *** adjohn has joined #openstack-dev | 02:28 | |
| smoser | vishy, hm.. i thought it worked once, but now not so. | 02:28 |
| vishy | darn | 02:28 |
| vishy | :( | 02:28 |
| vishy | it works from inside the guests :( | 02:29 |
| vishy | or at least it used to | 02:29 |
| smoser | http://paste.ubuntu.com/763379/ | 02:29 |
| smoser | ah. no it is working. | 02:30 |
| vishy | oh? | 02:30 |
| smoser | server-2 was just not good. server-3 is. | 02:30 |
| vishy | what did you do differently? | 02:30 |
| vishy | haha | 02:30 |
| smoser | http://paste.ubuntu.com/763381/ | 02:31 |
| smoser | so, i could even put that in my /etc/resolv.conf... | 02:31 |
| *** adjohn has quit IRC | 02:31 | |
| smoser | except that this is in an openstack instance, and it keeps getting wiped every 120 seconds i think | 02:31 |
| vishy | haven't looked what happened to nova list yet | 02:31 |
| vishy | but it broke sometime in the past couple of days | 02:32 |
| smoser | well, vishy thanks for your time. | 02:33 |
| *** zns has quit IRC | 02:33 | |
| vishy | yw | 02:33 |
| vishy | hopefully we get stuff fixed tomorrow | 02:34 |
| vishy | in meetings today and tomorrow | 02:34 |
| *** tryggvil has quit IRC | 02:36 | |
| *** lloydde has quit IRC | 02:37 | |
| *** rohitk has quit IRC | 02:44 | |
| *** andrewbogott has quit IRC | 02:46 | |
| *** rkukura has joined #openstack-dev | 02:51 | |
| *** rkukura has left #openstack-dev | 02:55 | |
| *** rkukura has joined #openstack-dev | 02:56 | |
| *** reed_ has quit IRC | 03:02 | |
| *** vladimir3p has joined #openstack-dev | 03:03 | |
| *** rods has quit IRC | 03:06 | |
| *** nati2_ has quit IRC | 03:08 | |
| *** rkukura has quit IRC | 03:09 | |
| *** rkukura has joined #openstack-dev | 03:09 | |
| smoser | vishy, https://code.launchpad.net/~smoser/euca2ools/describe-ipv4/+merge/84881 | 03:14 |
| smoser | maybe euca2ools will get '--ipv4' | 03:14 |
| *** lloydde has joined #openstack-dev | 03:15 | |
| *** lorin1 has joined #openstack-dev | 03:17 | |
| *** dragondm has joined #openstack-dev | 03:22 | |
| *** dragondm_ has joined #openstack-dev | 03:22 | |
| *** dolphm has joined #openstack-dev | 03:22 | |
| *** dragondm has quit IRC | 03:44 | |
| *** dragondm_ has quit IRC | 03:44 | |
| *** dolphm has quit IRC | 03:45 | |
| *** lloydde has quit IRC | 03:45 | |
| *** dragondm has joined #openstack-dev | 03:51 | |
| *** dragondm_ has joined #openstack-dev | 03:51 | |
| *** lorin1 has quit IRC | 03:51 | |
| *** heckj has quit IRC | 03:56 | |
| *** thickskin has joined #openstack-dev | 03:56 | |
| *** rsampaio has joined #openstack-dev | 04:13 | |
| *** Rajaram has joined #openstack-dev | 04:21 | |
| *** scottjg has joined #openstack-dev | 04:24 | |
| *** rsampaio has quit IRC | 04:31 | |
| *** rkukura has quit IRC | 04:33 | |
| *** rsampaio has joined #openstack-dev | 04:36 | |
| *** lloydde has joined #openstack-dev | 04:37 | |
| *** jakedahn has joined #openstack-dev | 04:40 | |
| *** rsampaio has quit IRC | 04:45 | |
| *** vladimir3p has quit IRC | 04:50 | |
| *** dwalleck has joined #openstack-dev | 04:54 | |
| *** adjohn has joined #openstack-dev | 04:57 | |
| *** spiffxp has joined #openstack-dev | 05:04 | |
| *** Abd4llA has quit IRC | 05:08 | |
| *** lloydde has quit IRC | 05:29 | |
| *** Tribaal has quit IRC | 05:34 | |
| *** uvirtbot has quit IRC | 05:34 | |
| *** scottsanchez has quit IRC | 05:34 | |
| *** mjfork has quit IRC | 05:36 | |
| openstackgerrit | Brad Hall proposed a change to openstack/nova: Make QuantumManager no longer depend on the projects table https://review.openstack.org/2166 | 05:37 |
| openstackgerrit | Brad Hall proposed a change to openstack/nova: Fix network forwarding rule initialization in QuantumManager https://review.openstack.org/2167 | 05:37 |
| *** uvirtbot has joined #openstack-dev | 05:38 | |
| *** lloydde has joined #openstack-dev | 05:39 | |
| *** Tribaal has joined #openstack-dev | 05:39 | |
| *** lloydde has quit IRC | 05:43 | |
| *** gavri has quit IRC | 05:47 | |
| *** gavri has joined #openstack-dev | 05:49 | |
| *** hugokuo has joined #openstack-dev | 05:50 | |
| *** scottsanchez has joined #openstack-dev | 05:52 | |
| *** markmc has joined #openstack-dev | 06:16 | |
| *** dwalleck has quit IRC | 06:19 | |
| *** dragondm has quit IRC | 06:41 | |
| *** dragondm_ has quit IRC | 06:41 | |
| *** zaitcev has quit IRC | 06:42 | |
| *** Mandell has quit IRC | 06:51 | |
| *** ben_duyujie has left #openstack-dev | 07:06 | |
| *** novas0x2a|laptop has quit IRC | 07:07 | |
| *** novas0x2a|laptop has joined #openstack-dev | 07:10 | |
| *** lloydde has joined #openstack-dev | 07:18 | |
| *** vladimir3p has joined #openstack-dev | 07:25 | |
| *** spiffxp has quit IRC | 08:00 | |
| *** lloydde has quit IRC | 08:01 | |
| *** jeffjapan has quit IRC | 08:03 | |
| *** winston-d has joined #openstack-dev | 08:05 | |
| *** scottjg has quit IRC | 08:14 | |
| *** LinuxJedi has quit IRC | 08:17 | |
| *** apevec has joined #openstack-dev | 09:05 | |
| *** LinuxJedi has joined #openstack-dev | 09:27 | |
| *** jakedahn has quit IRC | 09:31 | |
| *** vladimir3p has quit IRC | 09:31 | |
| *** pixelbeat has joined #openstack-dev | 09:32 | |
| *** adjohn has quit IRC | 09:47 | |
| *** adjohn has joined #openstack-dev | 09:50 | |
| *** darraghb has joined #openstack-dev | 09:54 | |
| *** Sander^work has quit IRC | 10:42 | |
| *** Sander^work has joined #openstack-dev | 10:43 | |
| *** lts has joined #openstack-dev | 11:44 | |
| *** rods has joined #openstack-dev | 11:46 | |
| *** Abd4llA has joined #openstack-dev | 12:29 | |
| *** bsza has joined #openstack-dev | 12:54 | |
| *** crobinso has joined #openstack-dev | 13:05 | |
| *** Rajaram has quit IRC | 13:09 | |
| openstackgerrit | Thierry Carrez proposed a change to openstack/nova: Revert to using IP addresses as EC2 dnsName https://review.openstack.org/2168 | 13:09 |
| *** Rajaram has joined #openstack-dev | 13:09 | |
| ttx | smoser: please review: ^ | 13:09 |
| *** dolphm has joined #openstack-dev | 13:15 | |
| smoser | ttx, fyi https://code.launchpad.net/~smoser/euca2ools/describe-ipv4/+merge/84881 | 13:16 |
| ttx | smoser: hm. So should that be fuxed in euca-describe-instances instead ? | 13:17 |
| ttx | or fixed* | 13:17 |
| smoser | well, vishy pointed out that the response does include ip addresses. | 13:17 |
| smoser | and ec2 responds with hostnames in those fields. | 13:18 |
| smoser | so its hard to argue otherwise. | 13:18 |
| smoser | but without a tools change like the euca2ools, it will definitely mess with some installations (like the "Canonistack") | 13:18 |
| * markmc grumbles about commit messages which just say "Fixes bug #XXXXXXX" | 13:21 | |
| ttx | smoser: could you comment to that effect on the merge proposal ? | 13:23 |
| smoser | sure. | 13:23 |
| ttx | smoser: thx | 13:24 |
| *** zns has joined #openstack-dev | 13:25 | |
| openstackgerrit | Stuart McLaren proposed a change to openstack/glance: Fix for bug 901609, when using v2 auth should use /v2.0/tokens path. https://review.openstack.org/2169 | 13:25 |
| uvirtbot | Launchpad bug 901609 in glance "With keystone v2 auth does not use v2 path (/v2.0/tokens)" [Undecided,New] https://launchpad.net/bugs/901609 | 13:25 |
| *** Rajaram_ has joined #openstack-dev | 13:29 | |
| smoser | ttx, sorry to nak your patch | 13:30 |
| smoser | :) | 13:30 |
| ttx | well, you complained about it, I didn't :P | 13:30 |
| *** rbasak has quit IRC | 13:30 | |
| smoser | just because you hadn't used devstack since it went in | 13:30 |
| *** markvoelker has joined #openstack-dev | 13:31 | |
| *** Rajaram has quit IRC | 13:32 | |
| *** Rajaram_ is now known as Rajaram | 13:32 | |
| *** dolphm has quit IRC | 13:32 | |
| *** popux has joined #openstack-dev | 13:37 | |
| *** lorin1 has joined #openstack-dev | 13:39 | |
| *** mdomsch_ has quit IRC | 13:41 | |
| openstackgerrit | Stuart McLaren proposed a change to openstack/glance: Allow glance keystone unit tests to run with essex keystone. https://review.openstack.org/2170 | 13:42 |
| *** rbasak has joined #openstack-dev | 13:44 | |
| *** rkukura has joined #openstack-dev | 13:48 | |
| *** dolphm has joined #openstack-dev | 14:03 | |
| *** jmckenty has joined #openstack-dev | 14:04 | |
| *** kbringard has joined #openstack-dev | 14:06 | |
| *** mjfork has joined #openstack-dev | 14:08 | |
| *** sandywalsh has quit IRC | 14:14 | |
| *** Abd4llA has quit IRC | 14:17 | |
| *** troytoman-away is now known as troytoman | 14:20 | |
| *** zns has quit IRC | 14:30 | |
| *** zns has joined #openstack-dev | 14:30 | |
| *** zns has quit IRC | 14:32 | |
| *** dolphm has quit IRC | 14:33 | |
| *** dolphm has joined #openstack-dev | 14:33 | |
| *** dovetaildan has joined #openstack-dev | 14:40 | |
| *** Abd4llA has joined #openstack-dev | 14:54 | |
| *** bcwaldon has joined #openstack-dev | 14:55 | |
| *** jmckenty has quit IRC | 14:55 | |
| openstackgerrit | Anne Gentle proposed a change to openstack/openstack-manuals: Fixes trailing whitespace https://review.openstack.org/2171 | 14:59 |
| openstackjenkins | Project nova-docs build #1,739: SUCCESS in 3 min 12 sec: https://jenkins.openstack.org/job/nova-docs/1739/ | 15:04 |
| openstackjenkins | Thierry Carrez: A more secure root-wrapper alternative | 15:04 |
| *** hugo_kuo has joined #openstack-dev | 15:06 | |
| *** troytoman is now known as troytoman-away | 15:08 | |
| *** Abd4llA has quit IRC | 15:09 | |
| *** mattray has joined #openstack-dev | 15:20 | |
| *** popux has quit IRC | 15:21 | |
| *** popux has joined #openstack-dev | 15:21 | |
| *** Abd4llA has joined #openstack-dev | 15:22 | |
| *** Abd4llA is now known as Guest59484 | 15:22 | |
| *** jmckenty has joined #openstack-dev | 15:23 | |
| *** popux has quit IRC | 15:26 | |
| markmc | jeblair, hey; Kiall says he used to be able to +1 on stable/branch but not since we changed so that only stable-maint can +2 | 15:27 |
| Kiall | markmc: yea, just double checked in case I was imagining things ;) | 15:27 |
| markmc | Kiall, cool, thanks | 15:27 |
| *** lloydde has joined #openstack-dev | 15:30 | |
| openstackjenkins | Project nova-docs build #1,740: SUCCESS in 3 min 19 sec: https://jenkins.openstack.org/job/nova-docs/1740/ | 15:34 |
| openstackjenkins | Mark McLoughlin: Add new cfg module | 15:34 |
| *** dwalleck has joined #openstack-dev | 15:34 | |
| *** dwalleck has quit IRC | 15:37 | |
| openstackgerrit | Brian Waldon proposed a change to openstack/python-novaclient: Adding 'absolute-limits' and 'rate-limits' https://review.openstack.org/2103 | 15:39 |
| *** code_franco has joined #openstack-dev | 15:39 | |
| *** lloydde has quit IRC | 15:39 | |
| jeblair | mtaylor: ^ openstack-stable-maint isn't synced correctly (kiall should be in the group) | 15:39 |
| *** gavri has left #openstack-dev | 15:39 | |
| jeblair | markmc: mtaylor is most familiar with the sync script, but i can get up to speed and try to fix if he isn't able | 15:40 |
| Kiall | Nope - I'm not in the openstack-stable-maint group | 15:40 |
| markmc | jeblair, anyone should be able to +1 on stable/diablo | 15:40 |
| Kiall | Oh | 15:40 |
| Kiall | I just got added ;) | 15:40 |
| markmc | oh | 15:41 |
| *** rnirmal has joined #openstack-dev | 15:41 | |
| markmc | jeblair, you just added Kiall? | 15:41 |
| jeblair | oh, no, he's pending isn't he. :) | 15:41 |
| Kiall | Oops - right, Was reading the wrong list ;) Pending. | 15:41 |
| Kiall | jeblair: I can see the gerrit settings are wrong anyway, the exclusive flag prevents inheritance of the +1 for registered users | 15:42 |
| jeblair | markmc: there are 5 people pending membership https://launchpad.net/~openstack-stable-maint/+members | 15:42 |
| *** mdomsch has joined #openstack-dev | 15:42 | |
| markmc | jeblair, right and I emailed them all about what they need to do to join | 15:42 |
| markmc | jeblair, and changed the group to invite-only | 15:42 |
| *** jmckenty has quit IRC | 15:42 | |
| jeblair | okay, so at this point, they should not be members, and should not be able to +2? | 15:43 |
| Kiall | markmc: I'm betting most of them saw the same thing I did (No more +1) and made the same assumption as me! | 15:43 |
| markmc | jeblair, right, not members and should be able to +1 | 15:43 |
| markmc | Kiall, I actually think it was a mail I sent to the list which triggered it | 15:43 |
| markmc | Kiall, you were there before the others | 15:43 |
| jeblair | i'm sorry, i misread. you can't +1. | 15:43 |
| jeblair | (i just woke up. i should have just read slashdot or something) | 15:44 |
| Kiall | lol | 15:44 |
| markmc | Kiall, and if you're looking for reviews to do :) https://review.openstack.org/#q,status:open+branch:stable/diablo+project:openstack/nova+owner:markmc,n,z | 15:44 |
| Kiall | It looks like adding Registered Users as +1 to refs/heads/stable/* would fix it.. | 15:45 |
| Kiall | markmc: yea .. you just spammed my inbox ;) | 15:45 |
| *** cp16net has joined #openstack-dev | 15:46 | |
| jeblair | Kiall: can you see if you can +1, but not +2, now? | 15:47 |
| openstackjenkins | Project nova-docs build #1,741: SUCCESS in 3 min 17 sec: https://jenkins.openstack.org/job/nova-docs/1741/ | 15:49 |
| openstackjenkins | duncan: Updated the test runner module with a sys.path insert so that tests run in and | 15:49 |
| *** Rajaram has quit IRC | 15:50 | |
| *** reed has joined #openstack-dev | 15:56 | |
| *** bcwaldon has quit IRC | 16:00 | |
| *** adjohn has joined #openstack-dev | 16:00 | |
| Kiall | jeblair: yup - all fixed | 16:03 |
| jeblair | cool, sorry about that! | 16:04 |
| markmc | jeblair, cool, thanks | 16:04 |
| *** linuxjedi_ has joined #openstack-dev | 16:06 | |
| *** linuxjedi_ has quit IRC | 16:06 | |
| *** reidrac has quit IRC | 16:10 | |
| openstackgerrit | James E. Blair proposed a change to openstack/openstack-ci: Document fix to code review perms. https://review.openstack.org/2189 | 16:10 |
| *** bcwaldon has joined #openstack-dev | 16:11 | |
| jeblair | mtaylor: ^ nevermind :) | 16:12 |
| Kiall | markmc: my poor inbox ;) | 16:14 |
| markmc | Kiall, get filtering :) | 16:15 |
| *** hub_cap has joined #openstack-dev | 16:21 | |
| openstackgerrit | Thierry Carrez proposed a change to openstack/openstack-ci-puppet: Use launchpadlib instead of emails for bug updates https://review.openstack.org/2190 | 16:21 |
| *** sandywalsh has joined #openstack-dev | 16:24 | |
| *** zns has joined #openstack-dev | 16:31 | |
| openstackgerrit | Verification of a change to openstack/melange failed: Force using the melange branch of openstack.common. https://review.openstack.org/2119 | 16:38 |
| *** Mandell has joined #openstack-dev | 16:38 | |
| openstackgerrit | Thierry Carrez proposed a change to openstack/openstack-ci-puppet: Use launchpadlib instead of emails for bug updates https://review.openstack.org/2190 | 16:40 |
| *** zaitcev has joined #openstack-dev | 16:41 | |
| *** lloydde has joined #openstack-dev | 16:41 | |
| jkoelker | jeblair, mtaylor: Question on the git urls in pip-requires, re ^melange build failure. does it not run the pip-requires every time? | 16:45 |
| jeblair | jkoelker: looking | 16:45 |
| *** lloydde has quit IRC | 16:46 | |
| jkoelker | i had changed the pip-requires line to get from a specific branch so we can start with a clean slate for openstack-common and not break melange | 16:46 |
| *** adjohn has quit IRC | 16:47 | |
| *** adjohn has joined #openstack-dev | 16:47 | |
| *** AlanClark has joined #openstack-dev | 16:47 | |
| jeblair | hm. so the glance job runs "python tools/install_venv.py --upgrade", but it looks like that didn't make it into the melange job. one of the reasons i want to combine these. i'll add that to melange | 16:48 |
| *** rsampaio has joined #openstack-dev | 16:49 | |
| jeblair | jkoelker: okay, that's there and it's running now. | 16:49 |
| jkoelker | thanks | 16:49 |
| openstackgerrit | Brian Lamar proposed a change to openstack/tempest: Removed storm references from README.rst https://review.openstack.org/2191 | 16:52 |
| openstackgerrit | p-draigbrady proposed a change to openstack/nova: Bug#898257 support handling images with libguestfs https://review.openstack.org/1994 | 16:53 |
| openstackgerrit | p-draigbrady proposed a change to openstack/nova: Bug#898257 abstract out disk image access methods https://review.openstack.org/1993 | 16:53 |
| *** apevec has quit IRC | 16:59 | |
| *** novas0x2a|laptop has quit IRC | 17:05 | |
| *** adjohn has quit IRC | 17:07 | |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: Adds network model and network info cache. https://review.openstack.org/2028 | 17:07 |
| *** nati2 has joined #openstack-dev | 17:09 | |
| *** lloydde has joined #openstack-dev | 17:13 | |
| *** sleepsonthefloo has joined #openstack-dev | 17:15 | |
| openstackgerrit | Brad Hall proposed a change to openstack/nova: Make QuantumManager no longer depend on the projects table https://review.openstack.org/2166 | 17:15 |
| *** reed has quit IRC | 17:18 | |
| openstackjenkins | Project nova-docs build #1,742: SUCCESS in 3 min 43 sec: https://jenkins.openstack.org/job/nova-docs/1742/ | 17:19 |
| openstackjenkins | jenkins: Imported Translations from Launchpad | 17:19 |
| *** dolphm has quit IRC | 17:20 | |
| vishy | ttx: ping | 17:21 |
| *** maplebed has joined #openstack-dev | 17:23 | |
| * Vek is very tempted to mark his keystone bug as "confirmed" so he can set it to high priority. | 17:32 | |
| Vek | but it would probably be easier if I could just find a keystone dev. | 17:32 |
| * Vek would submit a patch, but there should be a more extensive unit test added as well, and he has his own stuff he needs to do... | 17:33 | |
| *** lloydde has quit IRC | 17:36 | |
| *** jmckenty has joined #openstack-dev | 17:36 | |
| openstackgerrit | Brian Lamar proposed a change to openstack/tempest: PEP8 fixes. https://review.openstack.org/2192 | 17:41 |
| Vek | ====================================================================== | 17:46 |
| Vek | ERROR: Failure: AttributeError ('module' object has no attribute 'skip') | 17:46 |
| Vek | ---------------------------------------------------------------------- | 17:46 |
| Vek | Looks like we let a Python 2.7-ism slip into the test suite. | 17:47 |
| *** jeremyb has quit IRC | 17:48 | |
| *** sandywalsh has quit IRC | 17:49 | |
| *** zul has quit IRC | 17:51 | |
| lorin1 | Vek: It's been logged as a bug. | 17:51 |
| lorin1 | Vek: But apparently there are other 2.6 problems as well... | 17:51 |
| *** zul has joined #openstack-dev | 17:51 | |
| lorin1 | Vek: https://bugs.launchpad.net/nova/+bug/901730 | 17:52 |
| uvirtbot | Launchpad bug 901730 in nova "Unit test errors on Python 2.6" [Undecided,New] | 17:52 |
| *** jdurgin has joined #openstack-dev | 17:53 | |
| *** lts has quit IRC | 17:53 | |
| vishy | smoser: ping | 17:53 |
| smoser | here | 17:54 |
| *** andrewbogott has joined #openstack-dev | 17:54 | |
| *** jeremyb has joined #openstack-dev | 17:55 | |
| vishy | hey | 17:55 |
| vishy | smoser: how is your ipv6 knowledge? | 17:55 |
| *** jeremyb has quit IRC | 17:56 | |
| *** jeremyb has joined #openstack-dev | 17:56 | |
| smoser | hm... right up there with my knowledge of intergalactic travel. | 17:56 |
| smoser | but i know that Daviey at least has some. | 17:57 |
| smoser | he's probably busy right now though. | 17:57 |
| Vek | lorin1: so I see. | 17:57 |
| vishy | smoser: darn | 17:57 |
| vishy | smoser: we're discussing an alternative to dhcp | 17:58 |
| Vek | ConfigParser.SafeConfigParser() does not take allow_no_value as an argument, and Python 2.6 does not have assertRaisesRegexp. | 17:58 |
| Vek | (nor does it have assertIsNone) | 17:58 |
| vishy | smoser: we prefer a net service for config over an attached drive | 17:58 |
| vishy | smoser: but apparently the ipv4 link local doesn't quite work | 17:59 |
| vishy | we are wondering if using ipv6 might solve the problem | 17:59 |
| smoser | well, i've heard soren and (i believe Daviey ) saying things like that. | 18:00 |
| openstackgerrit | Brian Waldon proposed a change to openstack/glance: Add the missing column header to list-cached https://review.openstack.org/2193 | 18:00 |
| *** sandywalsh has joined #openstack-dev | 18:01 | |
| *** dragondm has joined #openstack-dev | 18:03 | |
| *** markvoelker has quit IRC | 18:03 | |
| *** dwalleck has joined #openstack-dev | 18:04 | |
| *** ramon has joined #openstack-dev | 18:10 | |
| zns | Vek: keystone dev here. Which bug? | 18:11 |
| *** sandywalsh has quit IRC | 18:13 | |
| *** darraghb has quit IRC | 18:15 | |
| *** jog0 has joined #openstack-dev | 18:17 | |
| *** novas0x2a|laptop has joined #openstack-dev | 18:22 | |
| *** jaypipes has quit IRC | 18:23 | |
| *** reed has joined #openstack-dev | 18:26 | |
| *** jaypipes has joined #openstack-dev | 18:38 | |
| *** novas0x2a|laptop has quit IRC | 18:40 | |
| zul | Daviey is susppose to be pretty familar with ipv5 | 18:41 |
| zul | grr....ipv6 even | 18:41 |
| *** novas0x2a|laptop has joined #openstack-dev | 18:41 | |
| openstackgerrit | Verification of a change to openstack/glance failed: Allow glance keystone unit tests to run with essex keystone. https://review.openstack.org/2170 | 18:45 |
| openstackgerrit | Brian Waldon proposed a change to openstack/glance: Lock keystone to specific commit in pip-requires https://review.openstack.org/2194 | 18:51 |
| vishy | ah, soren, good call | 18:51 |
| vishy | soren: ping | 18:51 |
| *** zigo has joined #openstack-dev | 18:53 | |
| openstackgerrit | Verification of a change to openstack/glance failed: Allow glance keystone unit tests to run with essex keystone. https://review.openstack.org/2170 | 18:54 |
| vishy | mtaylor: ping | 18:54 |
| *** dwalleck has quit IRC | 18:54 | |
| mtaylor | vishy: pong | 18:54 |
| openstackgerrit | Brian Waldon proposed a change to openstack/glance: Lock keystone to specific commit in pip-requires https://review.openstack.org/2194 | 18:56 |
| *** hugo_kuo has quit IRC | 18:56 | |
| vishy | mtaylor: did you see my comment here: https://review.openstack.org/#change,2165 | 18:57 |
| andrewbogott | ameade (or others): I'm working on adding dns naming to floating ips. Do you think I should do that by adding to the existing floating_ip extension, or by creating a new one? | 18:58 |
| mtaylor | vishy: I didn't - but good point | 18:58 |
| bcwaldon | andrewbogott: what specifically are you interested in adding to the API? | 18:58 |
| mtaylor | vishy: essentially, euca2ools apparently require boto v2 at least ... I figured just tossing a patch up for review would be the easiest way to start a conversation about that :) | 18:59 |
| andrewbogott | bcwaldon: https://blueprints.launchpad.net/nova/+spec/public-and-private-dns | 18:59 |
| bcwaldon | andrewbogott: sounds like a new extensions to me | 19:00 |
| bcwaldon | extension* | 19:00 |
| vishy | mtaylor: also i am really confused about the test failure in venv | 19:00 |
| andrewbogott | At the moment I'm just trying to add a way to query the network manager for a list of available zones. That doesn't fit very well into the existing floating_ip interface. | 19:00 |
| andrewbogott | bcwaldon: OK, that's what I guessed, but didn't want to add needless clutter. | 19:00 |
| vishy | mtaylor: I would like to make that change for sure | 19:01 |
| vishy | mtaylor: it should work fine | 19:01 |
| vishy | markmc: ping | 19:01 |
| bcwaldon | andrewbogott: extensions are a way for us to expose new featuresets, and this seems like just that | 19:01 |
| mtaylor | vishy: seems dprince's problem is a deficiency in FakeHttplibConnection | 19:01 |
| andrewbogott | fair enough | 19:02 |
| mtaylor | vishy: I'll take it back to the drawing board in a little bit | 19:02 |
| vishy | mtaylor: but why would running a venv expose that? | 19:02 |
| vishy | mtaylor: strange | 19:02 |
| mtaylor | vishy: god only knows | 19:02 |
| vishy | mtaylor: oh looks like a simple addition | 19:03 |
| *** markvoelker has joined #openstack-dev | 19:03 | |
| dovetaildan | mtaylor/vish: I'm using boto 2.0 with python-adminclient just fine BTW. | 19:05 |
| vishy | with nova-adminclient? | 19:06 |
| dovetaildan | mtaylor/vish: Essentially I'm doing that in order to get the nova smoke tests running on Oneiric. | 19:06 |
| vishy | so we can probably relax the dep there and push up a new version? | 19:06 |
| dovetaildan | vishy: so long as the test pass I'm cool with your concern there. | 19:06 |
| vishy | dovetaildan == dprince? | 19:07 |
| dovetaildan | vishy: yes. | 19:07 |
| vishy | who is maintainer of nova-adminclient? | 19:07 |
| dovetaildan | vishy: sorry. I had to use another box. Couldn't get on freenode at work. | 19:07 |
| vishy | or should we just remove the dep in nova? | 19:07 |
| vishy | (on nova-adminclient) | 19:07 |
| dovetaildan | vishy: If we remove the few smoke tests which use it I'm fine if it goes away altogether. | 19:08 |
| dovetaildan | vishy: Still using it to create a user, download a tarball, etc w/ the smoketests however. | 19:09 |
| vishy | ok I will make a patch that gets rid of all of that | 19:09 |
| dovetaildan | vishy: once we have that once the unit tests pass on mtaylors branch I'm good w/ bumping boto. | 19:10 |
| andrewbogott | bcwaldon: Next question... I can parrot an existing ExtensionDescriptor, but I don't much understand what my new code means. Can you suggest any other guidance beyond http://nova.openstack.org/devref/addmethod.openstackapi.html? | 19:11 |
| andrewbogott | (Otherwise I can just write a naive implementation and submit it to gerrit. *shrug* ) | 19:12 |
| bcwaldon | andrewbogott: ok, one sec | 19:13 |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: Relax dependency on boto 1.9b and nova-adminclient https://review.openstack.org/2195 | 19:19 |
| vishy | mtaylor, dovetaildan: ^^ | 19:20 |
| *** vladimir3p has joined #openstack-dev | 19:20 | |
| bcwaldon | andrewbogott: ok, so I'm thinking you're going to want to design the REST interface you want to expose then write a ResourceExtension | 19:20 |
| bcwaldon | andrewbogott: you then need to write an ExtensionDescriptor that documents that resource | 19:21 |
| dovetaildan | mtaylor/jeblair: How would I go about getting a non-core account for bellows? Essentially I want to automatically leave merge prop feedback w/ test results. I'd rather not use my own account for this... | 19:22 |
| andrewbogott | bcwaldon: Yeah, that looks right. I think it's mostly serialization that's going to challenge me, but that may just be a matter of trial and error. | 19:25 |
| *** ccorrigan has quit IRC | 19:26 | |
| bcwaldon | andrewbogott: ok, I can help when you get that point if you need me to | 19:26 |
| andrewbogott | thanks! | 19:26 |
| *** dolphm has joined #openstack-dev | 19:31 | |
| *** rbasak has quit IRC | 19:36 | |
| *** mattray has quit IRC | 19:43 | |
| *** dolphm has quit IRC | 19:49 | |
| *** jmckenty has quit IRC | 19:50 | |
| openstackgerrit | Jay Pipes proposed a change to openstack/tempest: Bunch of cleanups in images tests and clients https://review.openstack.org/2196 | 19:57 |
| *** jmckenty has joined #openstack-dev | 20:01 | |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: Adds network model and network info cache. https://review.openstack.org/2028 | 20:15 |
| *** GheRivero_ has joined #openstack-dev | 20:17 | |
| *** mdomsch has quit IRC | 20:24 | |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: Adds network model and network info cache. https://review.openstack.org/2028 | 20:24 |
| lorin1 | annegentle: Are you around? | 20:24 |
| annegentle | lorin1: hey, yup! | 20:26 |
| lorin1 | annegentle: Can I +2 something in docs that I proposed if it's been +1'd by somebody else in doc-core? | 20:27 |
| *** lloydde has joined #openstack-dev | 20:28 | |
| annegentle | lorin1: yes, you can, because you're doc-core (you can manually add up your +1 with the other +1 and make your vote a +2) | 20:28 |
| lorin1 | annegentle: thanks | 20:29 |
| soren | vishy: What's up? | 20:34 |
| vishy | soren: we have some crazy ideas regarding metadata service | 20:34 |
| vishy | i was wondering if you are familiar with ipv6 link local addresses | 20:34 |
| soren | I'm still trying to wrap my head around ipv5. HAven't even started on ipv6. | 20:35 |
| soren | I hear ipv7 will be awesome, though. | 20:36 |
| kbringard | lol | 20:37 |
| *** zigo has quit IRC | 20:37 | |
| zul | ttx: ping when you are around | 20:41 |
| vishy | soren: ok | 20:42 |
| vishy | soren: how about ipv4 link-local addresses? | 20:43 |
| soren | vishy: What's the problem with the.. | 20:43 |
| soren | vishy: Right that. | 20:43 |
| vishy | i.e. 169.254.169.254 | 20:43 |
| soren | vishy: Why do we need ipv6? | 20:43 |
| vishy | well it doesn't seem that guests actually give themselves a link local address | 20:44 |
| vishy | * all guests | 20:44 |
| *** anotherjesse has joined #openstack-dev | 20:44 | |
| vishy | but the main concern is that it isn't unique | 20:44 |
| anotherjesse | mtaylor: https://review.openstack.org/#change,2195 isn't going through .. | 20:44 |
| soren | vishy: What isn't unique? 169.254.169.254? | 20:45 |
| vishy | soren: so the way nova-metadata knows which info to pass is by the source ip | 20:45 |
| vishy | soren: so if the guest is assigning itself an ip | 20:45 |
| vishy | how do we know which info to send? | 20:46 |
| soren | Well... IT doesn't. It gets an IP over dhcp. | 20:46 |
| soren | ?!? | 20:46 |
| soren | Right? | 20:46 |
| vishy | soren: we're trying to figure out how to do it w/o dhcp | 20:46 |
| soren | What am I missing? | 20:46 |
| soren | Why? | 20:46 |
| vishy | there are some people that don't like the requirement of having nova-network have a bridge on the guest network | 20:47 |
| vishy | which it needs to dhcp properly | 20:47 |
| jeblair | anotherjesse: https://jenkins.openstack.org/job/gate-nova-unittests/202/ | 20:47 |
| vishy | so we're trying to figure out if there is a way to setup the instance without dhcp | 20:47 |
| vishy | but still be able to get metadata | 20:48 |
| anotherjesse | jeblair: odd - doesn't jenkin's usually comment that it is starting? | 20:48 |
| vishy | so cloud-init (for example) could get metadata and set up interfaces based on it | 20:48 |
| jeblair | anotherjesse: ah, i turned that off because it was generating a lot of spam | 20:48 |
| soren | vishy: What's the problem with nova-network having a bridge on the network? | 20:48 |
| vishy | soren: security concerns | 20:49 |
| vishy | bridging management networks and guest networks | 20:49 |
| soren | vishy: They'd be more comfrotable with an address assignment thingamajig we invent ourselves? | 20:49 |
| *** mattray has joined #openstack-dev | 20:49 | |
| vishy | well we aren't inventing it | 20:50 |
| soren | DHCP (and all its idiosyncracies) are well understood. I don't believe not using DHCP to do dynanic host configuration is a valid solution. | 20:50 |
| vishy | soren: cloud servers already doesn't use dhcp | 20:50 |
| vishy | so it is an existing solution | 20:51 |
| vishy | whether or not it is valid. | 20:51 |
| openstackgerrit | James E. Blair proposed a change to openstack/openstack-ci: Use a new image name each time we snapshot. https://review.openstack.org/2197 | 20:51 |
| kbringard | doesn't it use some crazy Open vSwitch setup with a bunch of custom stuff on the backend? | 20:51 |
| soren | vishy: Yes, but it's dreadful. | 20:51 |
| soren | vishy: It only works for cloud servers, because they don't let anyone else provide images. | 20:51 |
| soren | I'd really much rather have a discussion about this where the solution ahsn't been decided beforehand. | 20:52 |
| vishy | soren: ok so your answer is don't do that. | 20:52 |
| soren | Yes. | 20:52 |
| vishy | soren: not super helpful :) | 20:53 |
| soren | It's also not superhelpful to say "we don't want to use dhcp, because of security concerns". | 20:53 |
| soren | For all I know those security concerns could be complete BS. | 20:53 |
| soren | Or not. | 20:53 |
| soren | I don't know. | 20:54 |
| soren | BEcause I haven't heard them. | 20:54 |
| soren | My starting point for this discussion is this: STick with DHCP. | 20:54 |
| soren | Yours is clearly different. | 20:54 |
| vishy | so you think investigating an alternative means of configuring an image is not worth the time. | 20:54 |
| soren | Until the problem at hand is understood, none of us can say which is the right solutiuon. | 20:54 |
| soren | I think we'd need a *very* good reason to stop using dhcp. | 20:55 |
| soren | "security concerns" isn't a reason. It's a headline. | 20:55 |
| kbringard | not that my opinion matters, but I would 200% agree with that | 20:56 |
| soren | Sure it does. | 20:57 |
| jog0 | ls | 20:57 |
| soren | . .. | 20:57 |
| kbringard | hehe, I just meant I didn't mean in interject like I was some kind of authority | 20:58 |
| soren | vishy: I would love to have this discussion on the mailing list. | 21:00 |
| soren | vishy: My current stance is that I quite firmly believe that not using DHCP will simply replace whatever problems we have with another set of problems. | 21:01 |
| *** jmckenty has quit IRC | 21:02 | |
| *** jmckenty has joined #openstack-dev | 21:03 | |
| openstackjenkins | Project nova-docs build #1,743: SUCCESS in 3 min 25 sec: https://jenkins.openstack.org/job/nova-docs/1743/ | 21:04 |
| openstackjenkins | Vishvananda Ishaya: Relax dependency on boto 1.9b and nova-adminclient | 21:04 |
| markmc | vishy, what's up? | 21:13 |
| openstackgerrit | Lorin Hochstein proposed a change to openstack/openstack-manuals: Fixed how to add how to endpoints. https://review.openstack.org/2198 | 21:13 |
| openstackgerrit | Lorin Hochstein proposed a change to openstack/openstack-manuals: Specify HOST_IP in endpoint template https://review.openstack.org/2021 | 21:13 |
| *** f-yang has joined #openstack-dev | 21:17 | |
| openstackgerrit | Scott Moser proposed a change to openstack-dev/devstack: use cgroup-lite package https://review.openstack.org/2199 | 21:23 |
| openstackgerrit | Scott Moser proposed a change to openstack-dev/devstack: install libvirt as a dependency of nova https://review.openstack.org/2200 | 21:23 |
| openstackgerrit | Scott Moser proposed a change to openstack-dev/devstack: support non "uec style" IMAGE_URLS https://review.openstack.org/2201 | 21:23 |
| openstackgerrit | Scott Moser proposed a change to openstack-dev/devstack: stackrc: replace ttylinux-uec images with cirros https://review.openstack.org/2202 | 21:23 |
| smoser | anotherjesse, see above. sorry for the spam of 4 reviews, but i'd like you to take alook at those. | 21:25 |
| openstackgerrit | Jake Dahn proposed a change to openstack/nova: blueprint osapi-console-log bug 876809 https://review.openstack.org/2203 | 21:25 |
| uvirtbot | Launchpad bug 876809 in nova "Console log should allow you to tail from the api" [Wishlist,Confirmed] https://launchpad.net/bugs/876809 | 21:25 |
| smoser | i'd comment that there should be some documentation on how to get gerrit and devstack set up, but i'm affraid someone would say "then write it" | 21:25 |
| openstackgerrit | anotherjesse proposed a change to openstack-dev/devstack: support versions of novaclient that uses api_key https://review.openstack.org/2204 | 21:25 |
| jeblair | smoser: do you have feedback about http://wiki.openstack.org/GerritWorkflow | 21:26 |
| jaypipes | smoser: you're a busy little devstacker today. :) | 21:26 |
| anotherjesse | smoser: I agree - we want to improve the docs … | 21:26 |
| jeblair | or do mean specifically setting up a repo inside of a devstack host to submit code from there? | 21:27 |
| anotherjesse | smoser: as someone who has felt the pain of missing docs if you could share the questions that the docs should address it would be helpful | 21:27 |
| reed | what's the best way to see who committed a fix to a bug? is this information visible on launchpad? | 21:29 |
| smoser | jeblair, that doc is actually fine. | 21:29 |
| smoser | it was just figuring out that i had to set | 21:29 |
| smoser | [remote "gerrit"] | 21:29 |
| smoser | url = ssh://smoser@review.openstack.org:29418/openstack-dev/devstack.git | 21:29 |
| smoser | fetch = +refs/heads/*:refs/remotes/gerrit/* | 21:29 |
| jeblair | smoser: did you use git-review ("pip install git-review")? | 21:29 |
| smoser | i did. and it said "i dont know what that is" | 21:30 |
| vishy | markmc: hi | 21:30 |
| smoser | (in other, probably nicer) words | 21:30 |
| *** ramon has quit IRC | 21:30 | |
| pvo | soren: I think the concern for "security issues' with what vishy was talking about is the bridging of L2 networks. Control plane then has to be bridged to customer plane | 21:30 |
| smoser | anotherjesse, the primary issue was just i didn't know where to checkout and push to. for some reason that was clear to me initially when I set up openstack git checkouts. | 21:30 |
| jeblair | smoser: that's really strange. when you use git-review, it should automatically set up that remote for you. i don't know why pip installing it wouldn't work. could another pypi server or http proxy or some-such be in the way? | 21:31 |
| smoser | from history: http://paste.ubuntu.com/764271/ | 21:32 |
| *** lorin1 has quit IRC | 21:32 | |
| vishy | markmc: I have questions about kvm and fedora for you | 21:32 |
| smoser | s/history/scrollback/ jeblair | 21:32 |
| markmc | vishy, awesome :) | 21:32 |
| soren | pvo: Won't these things have to be wired up anyway for the vm's to be able to reach the metadata serveR? | 21:33 |
| jeblair | smoser: that looks like the old rfc.sh script. if you "pip install git-review", you'll get a much nicer program that understands what url to use for the remote. | 21:34 |
| openstackgerrit | Jake Dahn proposed a change to openstack/nova: blueprint osapi-console-log bug 876809 https://review.openstack.org/1071 | 21:34 |
| uvirtbot | Launchpad bug 876809 in nova "Console log should allow you to tail from the api" [Wishlist,Confirmed] https://launchpad.net/bugs/876809 | 21:34 |
| jeblair | we're going to retire the rfc.sh scripts soon | 21:34 |
| smoser | i thought i did pip install.. | 21:34 |
| smoser | maybe i forgot after i apt-get installed pip | 21:35 |
| smoser | sorry, jeblair | 21:35 |
| smoser | it seems like i did | 21:35 |
| jeblair | no problem; if you pip install, that _should_ override the alias you previously set up to run rfc.sh | 21:35 |
| jeblair | it's possible that it's not for some reason. you might remove that alias just to clear things up | 21:35 |
| jeblair | perhaps after doing that, you'll see that git isn't finding the git-review program, or something. | 21:35 |
| pvo | soren: yes, but we were brainstorming ways to get data over to a service that wasn't tied back to the control plane | 21:36 |
| anotherjesse | yikes - devstack now gets python-novaclient from nova's tools/pip-requires - which gives us 2.6.5 which doesn't work with keystone | 21:36 |
| anotherjesse | for last day | 21:36 |
| markmc | vishy, keeping me in suspense? :) | 21:37 |
| vishy | markmc: i was waiting for you to come back | 21:37 |
| * soren calls it a day | 21:37 | |
| openstackgerrit | Jake Dahn proposed a change to openstack/nova: blueprint osapi-console-log bug 876809 https://review.openstack.org/1071 | 21:37 |
| uvirtbot | Launchpad bug 876809 in nova "Console log should allow you to tail from the api" [Wishlist,Confirmed] https://launchpad.net/bugs/876809 | 21:37 |
| openstackgerrit | James E. Blair proposed a change to openstack/openstack-ci: Ignore exceptions when deleting images. https://review.openstack.org/2205 | 21:37 |
| markmc | vishy, I said "vishy: awesome :)" | 21:37 |
| vishy | markmc: question 1) is there some analog to cloud-init in fedora? | 21:38 |
| *** crobinso has quit IRC | 21:38 | |
| markmc | vishy, someone's working on packaging cloud-init for fedora | 21:38 |
| * markmc is a big fan; wish it was there yesterday | 21:38 | |
| markmc | smoser, :) | 21:38 |
| vishy | and is there some magic image somewhere that will work as a guest on all hypervisors? | 21:38 |
| vishy | a la uec images | 21:38 |
| smoser | it is packaged i believe | 21:38 |
| kbringard | I made an old version work with cent 5.6 a few months back | 21:39 |
| kbringard | well enough to get meta-data and keys :-) | 21:39 |
| kbringard | err, user-data | 21:39 |
| markmc | vishy, oh wait, it's there - http://pkgs.fedoraproject.org/gitweb/?p=cloud-init.git | 21:39 |
| smoser | the uec images should work in all hypervisors (well, at least 2 of them.. should work on xen-pvm, xen-hvm, and kvm) | 21:39 |
| markmc | vishy, no, we don't have any such magic images | 21:40 |
| vishy | markmc: seems like that would be valuable | 21:41 |
| markmc | kbringard, amazon did a bunch of work on cloud-init for fedora/rhel and then, predictably, didn't contribute it upstream | 21:41 |
| markmc | kbringard, even went so far as to obfuscate the change | 21:41 |
| kbringard | that's because they're jerks :-p | 21:41 |
| vishy | since we have both xen and kvm openstack clouds | 21:41 |
| vishy | markmc: second question | 21:41 |
| kbringard | I think the build I did I stole an amazon src rpm I found | 21:41 |
| markmc | vishy, well, I'd like us to publish openstack/kvm images for a start :) | 21:42 |
| kbringard | unpacked it, made a few changes and repackaged it | 21:42 |
| vishy | markmc: is there any way for libvirt to gracefully degrade | 21:42 |
| vishy | markmc: for network and block drivers | 21:42 |
| vishy | markmc: as in, try to use virtio, but fallback if there is a failure | 21:42 |
| markmc | vishy, oh, right - you mean provide a guest some sort of dual e1000/virtio device | 21:42 |
| vishy | markmc: right | 21:42 |
| vishy | and same with block | 21:43 |
| markmc | vishy, I know RHEV provides a dual rtl8139/virtio_net, but I don't think I ever figured out how that works | 21:43 |
| vishy | markmc: wow that is cool | 21:43 |
| markmc | cdub, how about you? ^^^ | 21:43 |
| markmc | vishy, I don't think they do it for block, though | 21:43 |
| * cdub scrolss up | 21:43 | |
| cdub | scrolls too | 21:43 |
| markmc | cdub, dual rtl8139/virtio_net in RHEV | 21:44 |
| * markmc reads the code | 21:44 | |
| markmc | vishy, http://git.fedorahosted.org/git/?p=vdsm.git :) | 21:44 |
| mtaylor | markmc: do you have magical plans to make the run_tests.sh stuff agree across all of the projects? | 21:46 |
| markmc | mtaylor, nope, but wasn't run_tests.py moved into nova/testing during the week? | 21:46 |
| markmc | mtaylor, something like that might make sense in common | 21:46 |
| mtaylor | markmc: yeah. in nova. | 21:46 |
| mtaylor | markmc: I would love to see all of the stuff that was in run_tests.py go somewhere in a generic way, and then have run_test.sh be something really small and consistent that just consumes it ... | 21:47 |
| mtaylor | of course, I'd really love to see all of it go away and just become patches/options/plugins for existing test runner tools- but I'm an extremist | 21:47 |
| markmc | mtaylor, I'd love nova's run_test.sh not to take an age compared to just running nosetests | 21:47 |
| mtaylor | ++ | 21:48 |
| * markmc really should look into that | 21:48 | |
| *** anotherjesse has quit IRC | 21:48 | |
| * mtaylor would love nose to just work, and to have a plugin if necessary to do our magical virtualenv stuff - so that you could just install nove-virtualenv or something and it would just be happy | 21:48 | |
| * mtaylor sits back and waits for markmc to produce magic | 21:49 | |
| vishy | mtaylor: patch nose to provide pretty colored output? | 21:49 |
| markmc | mtaylor, I'm busying making cfg work on python 2.6 | 21:49 |
| * markmc makes grumbling noises | 21:49 | |
| mtaylor | oh yeah. | 21:49 |
| mtaylor | have fun with that | 21:49 |
| mtaylor | vishy: or do a nose plugin to provide pretty colored output - yeah | 21:50 |
| mtaylor | vishy: I mean, it would mean less copied code between projects, and people do like colors :) | 21:50 |
| markmc | vishy, you know, I think the rtl8139/virtio thing is just two NICs with the same MAC connected to the same bridge | 21:50 |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: Adds network model and network info cache. https://review.openstack.org/2028 | 21:51 |
| cdub | markmc: yeah, there's no device that acts as both | 21:51 |
| markmc | vishy, http://git.fedorahosted.org/git/?p=vdsm.git;a=blob;f=vdsm/libvirtvm.py;h=857bd63c3a8ce4083eb2b99ef180d75030f81e14;hb=HEAD#l668 | 21:51 |
| mtaylor | but I think the main features of our test running are the seamless use of virtualenv, the colored output, and the options for running pep8 - all of those could be handled as nose plugins or setuptools extensions. in theory. while I'm imagining work for people | 21:51 |
| cdub | markmc: perhaps some hotplug + mac magic | 21:51 |
| markmc | vishy, where nicModel can be "rtl8139,pv" | 21:51 |
| markmc | cdub, just two NICs I think | 21:51 |
| cdub | markmc: boring ;) | 21:52 |
| markmc | cdub, yeah, so much for magic | 21:52 |
| *** anotherjesse has joined #openstack-dev | 21:52 | |
| markmc | cdub, there must be magic in the windows virtio drivers to disable the rtl8139 | 21:52 |
| soren | vishy: There are several plugins for nose that makes it produce pretty, coloured output. | 21:54 |
| soren | vishy: yanc, for one, is quite nice. | 21:54 |
| vishy | markmc: craziness | 21:54 |
| vishy | markmc: that actually works? | 21:54 |
| markmc | vishy, I'm guessing only for windows guests - i.e. the magic hack is probably in the windows virtio drivers | 21:55 |
| vishy | markmc: ah | 21:55 |
| cdub | markmc: could be, i've not looked at those drivers, or in the guest agent | 21:55 |
| vishy | markmc: sad face | 21:55 |
| mtaylor | soren: we should make porting our extra stuff into nose and or nose plugins a todo list items somewhere | 21:56 |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: Relax novaclient and remove redis dependency https://review.openstack.org/2206 | 21:56 |
| mtaylor | soren: I'd love for the testing answer in all of the projects to be "python setup.py tests" | 21:56 |
| *** zaitcev has quit IRC | 21:56 | |
| mtaylor | vishy: re: change above - should we go ahead and change the python-novaclient entry to a -e git url? | 21:57 |
| vishy | mtaylor: sure we could use the one from devstack! | 21:58 |
| * vishy goes to fix | 21:58 | |
| mtaylor | vishy: is devstack pointing to github url or gerrit url? (would prefer gerrit url) | 21:58 |
| soren | mtaylor: Or "python setup.py nosetests", really. | 21:59 |
| mtaylor | soren: well - I'd love it to be whatever dh is going to try to run | 21:59 |
| vishy | mtaylor: -e git+https://github.com/openstack/python-novaclient.git#egg=python_novaclient-dev | 22:00 |
| vishy | what is the equivalent gerrit? | 22:00 |
| mtaylor | vishy: one sec... | 22:00 |
| mtaylor | jeblair: ^^^ | 22:01 |
| jeblair | hi | 22:01 |
| vishy | mtaylor: it setup.py develops from whichever branch you specify. Github by default i think? | 22:01 |
| jeblair | https://review.openstack.org/p/openstack/python-novaclient.git | 22:02 |
| mtaylor | vishy: well, if we use ^^ instead of the github mirror, it'll be more robust (github goes down _way_ more than gerrit does) | 22:02 |
| openstackgerrit | Jake Dahn proposed a change to openstack/nova: blueprint osapi-console-log bug 876809 https://review.openstack.org/1071 | 22:02 |
| uvirtbot | Launchpad bug 876809 in nova "Console log should allow you to tail from the api" [Wishlist,Confirmed] https://launchpad.net/bugs/876809 | 22:02 |
| mtaylor | vishy: but both urls should fetch the same thing | 22:02 |
| jeblair | where are we talking about? | 22:02 |
| mtaylor | jeblair: in the nova pip-requires file | 22:02 |
| mtaylor | jeblair: vish was putting a patch in to remove the strict version require on python-novaclient in the pip-requires, so I suggested just going ahead and putting in the git-url since that was the plan for the client libs | 22:03 |
| soren | mtaylor: Ah, good point. | 22:04 |
| *** anotherjesse has quit IRC | 22:04 | |
| mtaylor | vishy: so, to be complete here: "-e git+https://review.openstack.org/p/openstack/python-novaclient.git#egg=python_novaclient-dev" | 22:04 |
| vishy | yeah | 22:04 |
| vishy | updated | 22:05 |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: Relax novaclient and remove redis dependency https://review.openstack.org/2206 | 22:05 |
| mtaylor | rock | 22:05 |
| jeblair | mtaylor: when we do integration testing, we'll be providing our own python-novaclient repo | 22:05 |
| mtaylor | yes | 22:05 |
| jeblair | so that url won't be used by jenkins | 22:05 |
| mtaylor | but for unittesting in the venvs it will be | 22:05 |
| jeblair | okay | 22:05 |
| vishy | I guess when we branch for essex, we should update that to current version | 22:05 |
| vishy | as in @stable/diablo | 22:05 |
| vishy | * @stable/essex | 22:05 |
| vishy | perhaps we should also do it for the milestones? | 22:06 |
| vishy | @essex-2 ? | 22:06 |
| larissa | vishy: Error: "essex-2" is not a valid command. | 22:06 |
| jeblair | larissa: who are you? | 22:06 |
| larissa | jeblair: Error: "who" is not a valid command. | 22:06 |
| vishy | mtaylor, jeblair: ^^ | 22:06 |
| mtaylor | vishy: I'm not as worried about it for the milestones - but I could be wrong there | 22:07 |
| mtaylor | vishy: we _can_ reference tags or revisions in that git url for pip | 22:07 |
| mtaylor | vishy: so if we do need to pin the milestone branches to a version, it's totally doable pretty easily | 22:07 |
| bcwaldon | mtaylor: can you add a job for gating tempest on pep8 in jenkins? | 22:07 |
| mtaylor | bcwaldon: hell yes | 22:07 |
| *** hhoover has joined #openstack-dev | 22:07 | |
| jeblair | mtaylor: w00t: https://jenkins.openstack.org/job/dev-gate-integration-tests-devstack-vm/155/ | 22:08 |
| mtaylor | bcwaldon: is it clean? | 22:08 |
| mtaylor | vishy: ^^^ check the devstack job there | 22:08 |
| bcwaldon | mtaylor: blamar has a branch that went in today | 22:08 |
| bcwaldon | mtaylor: shoudl be 0 | 22:08 |
| jeblair | mtaylor: i'll send out email about that now that it looks like we can run it again. | 22:10 |
| vishy | mtaylor: sexy! | 22:11 |
| mtaylor | jeblair: would you check me on this: https://jenkins.openstack.org/job/gate-tempest-pep8/ | 22:11 |
| bcwaldon | mtaylor: well damn | 22:11 |
| bcwaldon | mtaylor: I think I lied | 22:11 |
| mtaylor | bcwaldon: awesome | 22:12 |
| bcwaldon | mtaylor: he fixed pep8 in tempest/ | 22:12 |
| mtaylor | well - there's a gating job now - so fix those problems | 22:12 |
| bcwaldon | mtaylor: that works | 22:12 |
| *** anotherjesse has joined #openstack-dev | 22:12 | |
| bcwaldon | mtaylor: we'll fix it right now | 22:12 |
| bcwaldon | mtaylor: thanks! | 22:12 |
| mtaylor | bcwaldon: oh - I'm only running pep8 on tempest... | 22:12 |
| bcwaldon | mtaylor: oh! then we're good | 22:12 |
| mtaylor | bcwaldon: should I add other dirs? | 22:13 |
| bcwaldon | mtaylor: I'd say no, but I may get back to you on that | 22:13 |
| jeblair | No such project 'template-tempest-copy-venv'. Did you mean 'template-melange-copy-venv'? | 22:13 |
| mtaylor | bcwaldon: k | 22:13 |
| bcwaldon | mtaylor: and what pep8 version are you running | 22:13 |
| bcwaldon | mtaylor: so we can reproduce failures locally | 22:13 |
| jeblair | i don't think tempest has a venv, so i will delete that build step | 22:13 |
| mtaylor | jeblair: yes. thanks | 22:13 |
| mtaylor | bcwaldon: 0.6.1 | 22:13 |
| bcwaldon | mtaylor: kk, I'll try to get a runner locally in tempest | 22:13 |
| *** zns has quit IRC | 22:13 | |
| jeblair | https://jenkins.openstack.org/view/Tempest/ | 22:14 |
| jeblair | added a view | 22:14 |
| *** anotherjesse has quit IRC | 22:14 | |
| *** zaitcev has joined #openstack-dev | 22:15 | |
| jeblair | https://jenkins.openstack.org/view/Tempest/job/gate-tempest-pep8/1/ | 22:16 |
| jeblair | bcwaldon, mtaylor: ^ i retriggered the most recent patch, looks good. | 22:16 |
| *** jmckenty has quit IRC | 22:16 | |
| bcwaldon | jeblair: thanks! | 22:16 |
| mtaylor | sexy | 22:16 |
| *** zns has joined #openstack-dev | 22:17 | |
| openstackgerrit | Jay Pipes proposed a change to openstack/tempest: Bunch of cleanups in images tests and clients https://review.openstack.org/2196 | 22:19 |
| soren | mtaylor: From what I can tell, dh won't even call "python setup.py test". | 22:22 |
| *** rnirmal has quit IRC | 22:22 | |
| mtaylor | soren: really? wow. that's kind of weird | 22:22 |
| mtaylor | soren: since it calls make check by default for autotools projects | 22:23 |
| soren | mtaylor: If there's a makefile, it'll check if there's a check or test target and run that. | 22:23 |
| vishy | jeblair: is that devstack vm running in cloudservers? | 22:23 |
| mtaylor | I guess not enough python projects implement setup.py test for it to make sense | 22:23 |
| soren | mtaylor: AFAICT, it has no such magic for python. | 22:23 |
| mtaylor | soren: ass | 22:23 |
| mtaylor | vishy: yes | 22:23 |
| vishy | mtaylor: are you using pvgrubbed oneiric? | 22:23 |
| mtaylor | vishy: we have been - _although_ they just released oneiric images, so it might be possible to base our image on top of that - but for now pvgrubbed oneiric | 22:24 |
| openstackgerrit | Jay Pipes proposed a change to openstack/glance: Fixes LP Bug#901534 - Lost properties in upload https://review.openstack.org/2207 | 22:24 |
| vishy | wow, finally! | 22:24 |
| mtaylor | right? | 22:24 |
| *** cp16net has quit IRC | 22:25 | |
| soren | Gawd, I'm looking forward to a cloud provider that isn't Amazon that lets you upload your own images. | 22:25 |
| mtaylor | soren: ++ | 22:26 |
| jeblair | i believe all new images are supposed to have pvgrub turned on; our oneiric shouldn't be much different than the "official" one. | 22:27 |
| vishy | jeblair: interesting | 22:38 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/glance: Add new cfg module https://review.openstack.org/2141 | 22:38 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/glance: Convert glance to use the new cfg module https://review.openstack.org/2142 | 22:38 |
| vishy | markmc: does that mean you fixed it for python 2.6? | 22:39 |
| markmc | vishy, it does | 22:39 |
| vishy | markmc: w00t | 22:39 |
| markmc | vishy, mostly unittest stuff, but also ConfigParser in 2.6 doesn't support allow_no_value | 22:40 |
| markmc | vishy, which I think is about allowing 'foo = None' in the config files | 22:40 |
| * markmc shrugs | 22:40 | |
| markmc | vishy, hmm, so it got merged into nova without working on 2.6 ? | 22:41 |
| markmc | vishy, what is this heresy? | 22:41 |
| vishy | markmc: yes nobody noticed | 22:41 |
| markmc | vishy, we don't gate on 2.6? | 22:41 |
| * mtaylor needs to get up 2.6 builders | 22:41 | |
| markmc | ah, okay | 22:41 |
| vishy | markmc: hence my comment earlier that we should have a gating job | 22:41 |
| mtaylor | yes. | 22:42 |
| * mtaylor just put it on his todo list | 22:42 | |
| *** anotherjesse has joined #openstack-dev | 22:44 | |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/nova: Make cfg work on python 2.6 https://review.openstack.org/2208 | 22:44 |
| *** anotherjesse has quit IRC | 22:44 | |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/nova: Port nova.flags to cfg https://review.openstack.org/2147 | 22:44 |
| * markmc kicks off run_tests.sh to be sure to be sure and goes to bed | 22:45 | |
| vishy | markmc: you reproposed the original patch which was already merged? | 22:47 |
| markmc | vishy, no, a new patch | 22:48 |
| vishy | the old one is there too ^ | 22:48 |
| markmc | vishy, and a new version of a patch that hadn't been merged yet | 22:48 |
| vishy | is there a difference? | 22:48 |
| markmc | vishy, it was rebased | 22:48 |
| vishy | hmm | 22:48 |
| vishy | will that work? | 22:48 |
| markmc | why would it not? | 22:48 |
| vishy | hmm wait i guess it isn't merged yet | 22:49 |
| markmc | right, the patch to add cfg was merged | 22:49 |
| vishy | ah it was adding the cfg that was merged | 22:49 |
| vishy | my bad | 22:49 |
| markmc | the patch to *use* it wasn't | 22:49 |
| vishy | continue going to bed | 22:49 |
| vishy | :) | 22:49 |
| markmc | thank you for your indulgence :) | 22:49 |
| * markmc tears himself away from twitter and really goes | 22:49 | |
| vishy | i guess we better merge before we get 2.6 gating in | 22:49 |
| vishy | :) | 22:50 |
| openstackgerrit | Alex Meade proposed a change to openstack/nova: Add preparation for asynchronous instance faults https://review.openstack.org/2209 | 22:50 |
| *** lloydde has quit IRC | 22:50 | |
| *** markvoelker has quit IRC | 22:54 | |
| *** dwalleck has joined #openstack-dev | 22:58 | |
| *** lloydde has joined #openstack-dev | 22:58 | |
| *** dwalleck has quit IRC | 23:01 | |
| openstackjenkins | Project nova-docs build #1,744: FAILURE in 1 min 20 sec: https://jenkins.openstack.org/job/nova-docs/1744/ | 23:02 |
| openstackjenkins | * Rick Harris: Add ability to see deleted and active records. | 23:02 |
| openstackjenkins | * Vishvananda Ishaya: Relax novaclient and remove redis dependency | 23:02 |
| *** hhoover has quit IRC | 23:02 | |
| openstackgerrit | Verification of a change to openstack/nova failed: Make QuantumManager no longer depend on the projects table https://review.openstack.org/2166 | 23:03 |
| *** mattray has quit IRC | 23:04 | |
| *** anotherjesse has joined #openstack-dev | 23:04 | |
| *** bcwaldon has quit IRC | 23:06 | |
| vishy | mtaylor: fail | 23:06 |
| vishy | error in nova setup command: 'install_requires' must be a string or list of strings containing valid project/version requirement specifiers | 23:06 |
| mtaylor | vishy: that failure doesn't look good | 23:06 |
| vishy | install_requires is using pip-requires to build | 23:06 |
| mtaylor | oh. well, tits | 23:07 |
| *** rsampaio has quit IRC | 23:07 | |
| mtaylor | lemme poke for a minute | 23:07 |
| vishy | yeah stupid | 23:07 |
| vishy | so we can't use the -e in pip-requires unless we change it | 23:07 |
| *** hub_cap has quit IRC | 23:07 | |
| vishy | or setup.py develop bgoes boom | 23:07 |
| *** kbringard has quit IRC | 23:08 | |
| openstackgerrit | Verification of a change to openstack/nova failed: Make cfg work on python 2.6 https://review.openstack.org/2208 | 23:08 |
| vishy | mtaylor: i guess we have to switch it back and just update the version of novaclient? | 23:09 |
| mtaylor | vishy: so... | 23:09 |
| mtaylor | vishy: I think we can patch the function in setup.py | 23:09 |
| vishy | mtaylor: and it broke your jenkins too | 23:09 |
| vishy | boom! | 23:09 |
| mtaylor | vishy: yeah. I'll bet it did | 23:09 |
| openstackgerrit | Jake Dahn proposed a change to openstack/nova: blueprint osapi-console-log bug 876809 https://review.openstack.org/1071 | 23:09 |
| uvirtbot | Launchpad bug 876809 in nova "Console log should allow you to tail from the api" [Wishlist,Confirmed] https://launchpad.net/bugs/876809 | 23:09 |
| vishy | ImportError: No module named novaclient | 23:09 |
| vishy | mtaylor how do you propose we patch the function? | 23:10 |
| mtaylor | vishy: sigh. ok - just make it a non-git line for now, and I'll work on a more robust solution | 23:10 |
| jeblair | vishy: what's up with "sudo apt-get install -y python-m2crypto" in install_venv | 23:10 |
| vishy | jeblair: the pip version splodes on oneiric | 23:10 |
| jeblair | i kind of think it's uncool for something like that to be doing sudo stuff | 23:11 |
| vishy | mtaylor | 23:11 |
| vishy | okies | 23:11 |
| jeblair | (esp, since it's already installed) | 23:11 |
| mtaylor | vishy: develop vs. venv is tricky | 23:11 |
| mtaylor | vishy: I'd just been thinking of venv so far ... | 23:11 |
| * mtaylor punches self in head | 23:12 | |
| jeblair | the pip-requires file gets used by setup? | 23:12 |
| jeblair | i thought those were separate, and you were planning on taking advantage of that? | 23:12 |
| jeblair | def load_required_packages(): with file('tools/pip-requires', 'r') as f: | 23:13 |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: revert using git for novaclient https://review.openstack.org/2210 | 23:13 |
| jeblair | ah. i see. :) | 23:13 |
| mtaylor | I was. | 23:13 |
| openstackgerrit | Gabriel Hurley proposed a change to openstack/horizon: PEP8 fixes. https://review.openstack.org/2211 | 23:13 |
| vishy | jeblair: it is now | 23:13 |
| mtaylor | BUT - there is still a question mark of what to do for the python setup.py develop workflow | 23:13 |
| vishy | not so sure I like it because it caused multiple issues | 23:13 |
| vishy | mtaylor: how do we push 2.5.8 to pip? | 23:13 |
| vishy | assuming we wanted to | 23:13 |
| vishy | mtaylor: should we ninja approve that patch to fix jenkins? | 23:14 |
| mtaylor | vishy: yes | 23:14 |
| vishy | mtaylor: done | 23:15 |
| openstackgerrit | James E. Blair proposed a change to openstack/nova: Rename .nova-venv to .venv. https://review.openstack.org/2212 | 23:15 |
| jeblair | whoops. authors. crap. | 23:16 |
| *** code_franco has quit IRC | 23:17 | |
| openstackgerrit | Verification of a change to openstack/nova failed: revert using git for novaclient https://review.openstack.org/2210 | 23:17 |
| *** dragondm_ has joined #openstack-dev | 23:18 | |
| openstackgerrit | James E. Blair proposed a change to openstack/nova: Rename .nova-venv to .venv. https://review.openstack.org/2212 | 23:18 |
| *** GheRivero_ has quit IRC | 23:19 | |
| *** jakedahn has joined #openstack-dev | 23:19 | |
| *** bsza has quit IRC | 23:20 | |
| vishy | mtaylor: we might need secret powers | 23:21 |
| vishy | https://jenkins.openstack.org/job/gate-nova-unittests/207/ | 23:22 |
| jeblair | mtaylor: should we add a virtualenv upgrade to the template copy job for nova? | 23:22 |
| jeblair | ie, is the problem here that it's running with an out of date venv? | 23:22 |
| jeblair | i believe that's in there for glance, and i just added it to another job this morning, but it's not on all of them (something that will be easier to fix once the venv renames are done) | 23:23 |
| jeblair | "python tools/install_venv.py --upgrade" | 23:23 |
| vishy | jeblair: can you fix that failure above? | 23:23 |
| jeblair | oh, nova doesn't have that option :/ | 23:23 |
| vishy | the patch to fix the pip-requires is failing because the builder doesn't have the right pip requires :| | 23:24 |
| jeblair | vishy: is there a way to upgrade the venv? | 23:24 |
| vishy | jeblair: activate it | 23:24 |
| vishy | and pip install python-novaclient | 23:25 |
| vishy | :) | 23:25 |
| vishy | or pip install -E <env> python-novaclient | 23:25 |
| jeblair | i mean automatically; glance has "python tools/install_venv.py --upgrade" | 23:27 |
| vishy | yeah i don't think we have one of those | 23:28 |
| vishy | :( | 23:28 |
| dovetaildan | vishy: Can you look at this: https://bugs.launchpad.net/nova/+bug/901925 | 23:29 |
| uvirtbot | Launchpad bug 901925 in nova "python2.6 setup.py clean -a returned exit code 1" [High,New] | 23:29 |
| vishy | yeah | 23:29 |
| vishy | dovetaildan: it is the thing we are trying to fix right now | 23:30 |
| openstackgerrit | Verification of a change to openstack/nova failed: revert using git for novaclient https://review.openstack.org/2210 | 23:30 |
| dovetaildan | vishy: sorry man. I just jumped back online. Thanks. I'll check back later tonight. | 23:30 |
| vishy | cool | 23:30 |
| vishy | mtaylor, jeblair: ideas on how we can unbreak this? | 23:31 |
| jeblair | so ideally, we make sure the venv is up to date right before we use it. but there doesn't appear to be a way to do that. | 23:31 |
| jeblair | without that, i think the venv caching scheme is broken | 23:31 |
| jeblair | this is a problem that will present again, even if we ninja fix it now | 23:32 |
| vishy | jeblair: sure, can we ninja fix to get it unbroken now | 23:32 |
| jeblair | but i think we maybe should just ninja fix it for now and figure out the right thing later | 23:32 |
| jeblair | :) | 23:32 |
| vishy | then we can figure out how to update the venv properly | 23:32 |
| vishy | exactly | 23:32 |
| vishy | +1 | 23:32 |
| mtaylor | sorry - got pulled in to a call | 23:34 |
| jeblair | vishy, mtaylor: i am going to merge the change in gerrit, overriding jenkins. sound good? | 23:34 |
| mtaylor | jeblair: SO - perhaps the re-build venv job should attempt to do the gating job functions on tip of trunk using the venv before declaring it successful? | 23:34 |
| mtaylor | jeblair: yes | 23:34 |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: Adds network model and network info cache. https://review.openstack.org/2028 | 23:34 |
| jeblair | then the venv will be rebuilt, and things should be good | 23:34 |
| vishy | mtaylor: coolness | 23:35 |
| vishy | and jeblair | 23:35 |
| vishy | :) | 23:35 |
| openstackgerrit | Gabriel Hurley proposed a change to openstack/horizon: PEP8 fixes. https://review.openstack.org/2211 | 23:36 |
| jeblair | vishy: merged, venv rebuild is running: https://jenkins.openstack.org/job/nova-venv/127/ | 23:37 |
| openstackgerrit | Verification of a change to openstack/nova failed: revert using git for novaclient https://review.openstack.org/2210 | 23:38 |
| jeblair | (you can ignore that) | 23:39 |
| vishy | hehe | 23:41 |
| vishy | ok it is in and rebuilt now? | 23:41 |
| vishy | there were a couple other patches that need to be rekicked | 23:42 |
| *** zns1 has joined #openstack-dev | 23:44 | |
| andrewbogott | markmc: I was talking to you earlier about a dns extension, right? (My backscroll buffer is tiny and my memory faint.) | 23:46 |
| *** zns has quit IRC | 23:46 | |
| openstackgerrit | Jake Dahn proposed a change to openstack/nova: blueprint osapi-console-log bug 876809 https://review.openstack.org/1071 | 23:47 |
| uvirtbot | Launchpad bug 876809 in nova "Console log should allow you to tail from the api" [Wishlist,Confirmed] https://launchpad.net/bugs/876809 | 23:47 |
| openstackjenkins | Yippie, build fixed! | 23:49 |
| openstackjenkins | Project nova-docs build #1,745: FIXED in 3 min 17 sec: https://jenkins.openstack.org/job/nova-docs/1745/ | 23:49 |
| openstackjenkins | Vishvananda Ishaya: revert using git for novaclient | 23:49 |
| openstackgerrit | Paul Voccio proposed a change to openstack/nova: removing TODO as we support Windows+XenServer and have no plans to support quiesce or VSS at the moment https://review.openstack.org/2213 | 23:51 |
| openstackgerrit | Kevin L. Mitchell proposed a change to openstack/nova: Add templates for selected resource extensions. https://review.openstack.org/2214 | 23:56 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!