| NobodyCam | lol closer ... http://paste.openstack.org/show/HPTMqqpPkfxBVk8Ttzad/ | 00:00 |
|---|---|---|
| jroll | weird? | 00:03 |
| jroll | seems wrong. | 00:04 |
| jroll | any logs in conductor? | 00:04 |
| NobodyCam | that was just prob me | 00:05 |
| NobodyCam | i go it to wait call back | 00:05 |
| NobodyCam | but the deploy image is crashing | 00:05 |
| NobodyCam | i need to dbl check the basics 32/64 bit | 00:06 |
| NobodyCam | enough mem | 00:06 |
| NobodyCam | blah | 00:06 |
| NobodyCam | etc | 00:06 |
| *** yuanying has joined #openstack-ironic | 00:06 | |
| NobodyCam | but it is trying | 00:06 |
| jroll | ah hmm | 00:07 |
| jroll | do you get anything on the console? | 00:08 |
| NobodyCam | ya big blow up | 00:08 |
| NobodyCam | lol | 00:08 |
| jroll | orly | 00:08 |
| jroll | can I see? | 00:08 |
| jroll | I mean... keep in mind nobody has ran this DIB image | 00:08 |
| jroll | so it could be completely f'd | 00:08 |
| spandhe | jroll: I saw the same error messages in nova-compute.log.. turns out, I was trying to use nova.network.manager instead of Neutron.. qq on that.. do I have to run neutron at all, or is there a way around it ? | 00:12 |
| jroll | mmm. not sure. I'm using neutron. | 00:13 |
| NobodyCam | jroll: I need to check what I've done before I say its broke | 00:13 |
| jroll | I thihnk you have to | 00:13 |
| spandhe | I am just thinking out loud.. nova-compute is going to call allocate_network_async.. and thats going to invoke whichever network API you configure | 00:13 |
| spandhe | jroll: ok cool..thanks! | 00:13 |
| jroll | yeah, I feel like it should work on the nova side, but not on the ironic side :( | 00:14 |
| *** achanda has quit IRC | 00:15 | |
| *** achanda has joined #openstack-ironic | 00:15 | |
| NobodyCam | jroll: I have a tiff of the console if you really want it | 00:15 |
| jroll | NobodyCam: sure why not :P | 00:16 |
| JayF | NobodyCam: never hurts to share info | 00:16 |
| jroll | we may have seen it before | 00:16 |
| spandhe | jroll: I am going to use my own dhcp server.. hence, trying to see if I can avoid using Neutron completely :) | 00:16 |
| jroll | aha, that can be done :) | 00:16 |
| NobodyCam | how you want the tiff | 00:16 |
| jroll | we do that ourselves | 00:16 |
| jroll | NobodyCam: dunno, as a tiff? | 00:16 |
| jroll | imgur.com? | 00:16 |
| *** yuanying has quit IRC | 00:17 | |
| spandhe | jroll: yeah I have read the blogs.. have gone through the new agent you guys proposed.. but that stuff is too fancy for my use case.. | 00:17 |
| *** yuanying has joined #openstack-ironic | 00:17 | |
| jroll | lol | 00:18 |
| jroll | it's not that fancy :) | 00:18 |
| NobodyCam | http://imgur.com/8nOmR5i | 00:19 |
| NobodyCam | lol first image ever shared | 00:19 |
| *** achanda has quit IRC | 00:19 | |
| spandhe | jroll: The Ironic agent is meant to update switch port configuration, right? I am going to use static DHCP allocation.. nothing will change on the network side.. | 00:20 |
| *** rushiagr is now known as rushiagr_away | 00:20 | |
| jroll | spandhe: no... that's a downstream thing we buildt, you don't need to use that | 00:20 |
| jroll | it's not upstream yet | 00:20 |
| spandhe | jroll: right.. thats what I was talking about.. its mentioned on your OnMetal blog.. interesting stuff! | 00:21 |
| jroll | NobodyCam: weird, maybe pass an init= something? | 00:21 |
| jroll | maybe init=/path/to/systemd | 00:22 |
| jroll | JayF: ^ | 00:22 |
| jroll | spandhe: thanks :) | 00:22 |
| jroll | spandhe: upstream ironic agent is still pretty simple | 00:22 |
| *** yuanying has quit IRC | 00:22 | |
| jroll | NobodyCam: though I think maybe there's a different problem making nothing executable | 00:23 |
| *** yuanying has joined #openstack-ironic | 00:28 | |
| spandhe | jroll: havent checked that yet.. let me check.. | 00:29 |
| openstackgerrit | yunhong-jiang proposed a change to openstack/ironic: Change the force_raw_image config usage https://review.openstack.org/130880 | 00:29 |
| openstackgerrit | yunhong-jiang proposed a change to openstack/ironic: Convert qcow2 image to raw format when deploy https://review.openstack.org/130881 | 00:29 |
| *** todd_dsm has joined #openstack-ironic | 00:31 | |
| *** yuanying has quit IRC | 00:35 | |
| *** yuanying has joined #openstack-ironic | 00:36 | |
| *** yuanying has quit IRC | 00:40 | |
| spandhe | jroll: did u mean DHCP agent? | 00:41 |
| jroll | spandhe: no, ironic-python-agent and the associated ironic driver | 00:41 |
| jroll | it's a deploy driver | 00:41 |
| jroll | instead of iscsi + dd we use a python application with a rest api | 00:42 |
| jroll | (in the ramdisk) | 00:42 |
| *** marcoemorais has quit IRC | 00:44 | |
| *** marcoemorais has joined #openstack-ironic | 00:44 | |
| *** marcoemorais has quit IRC | 00:46 | |
| *** marcoemorais has joined #openstack-ironic | 00:46 | |
| *** marcoemorais has quit IRC | 00:46 | |
| *** marcoemorais has joined #openstack-ironic | 00:46 | |
| *** todd_dsm has quit IRC | 00:49 | |
| spandhe | jroll: oh, ok.. I was talking about https://github.com/rackerlabs/ironic-neutron-plugin | 00:53 |
| *** todd_dsm has joined #openstack-ironic | 00:54 | |
| *** yuanying has joined #openstack-ironic | 00:55 | |
| *** hemna has quit IRC | 00:58 | |
| *** ryanpetrello has joined #openstack-ironic | 01:02 | |
| *** todd_dsm has quit IRC | 01:03 | |
| *** Marga_ has joined #openstack-ironic | 01:06 | |
| *** yuanying has quit IRC | 01:07 | |
| *** yuanying has joined #openstack-ironic | 01:08 | |
| *** yuanying has quit IRC | 01:12 | |
| *** ryanpetrello_ has joined #openstack-ironic | 01:13 | |
| *** ryanpetrello has quit IRC | 01:13 | |
| *** ryanpetrello_ is now known as ryanpetrello | 01:14 | |
| *** marcoemorais has quit IRC | 01:17 | |
| *** lucas-dinner has quit IRC | 01:19 | |
| *** ryanpetrello has quit IRC | 01:25 | |
| *** chenglch has joined #openstack-ironic | 01:26 | |
| *** ryanpetrello has joined #openstack-ironic | 01:30 | |
| *** ryanpetrello has quit IRC | 01:36 | |
| *** Marga_ has quit IRC | 01:37 | |
| *** ryanpetrello has joined #openstack-ironic | 01:39 | |
| *** chenglch has quit IRC | 01:49 | |
| *** chenglch has joined #openstack-ironic | 01:49 | |
| *** ryanpetrello has quit IRC | 01:53 | |
| *** ryanpetrello has joined #openstack-ironic | 01:57 | |
| *** todd_dsm has joined #openstack-ironic | 01:59 | |
| *** nosnos has joined #openstack-ironic | 02:06 | |
| *** foexle has quit IRC | 02:07 | |
| *** dlaube has quit IRC | 02:08 | |
| *** yuanying has joined #openstack-ironic | 02:14 | |
| *** lazy_prince has quit IRC | 02:25 | |
| Haomeng | morning:) | 02:30 |
| *** spandhe has quit IRC | 02:30 | |
| *** Poornima has joined #openstack-ironic | 02:31 | |
| *** ryanpetrello has quit IRC | 02:35 | |
| *** Marga_ has joined #openstack-ironic | 02:39 | |
| *** killer_prince has joined #openstack-ironic | 02:55 | |
| *** killer_prince is now known as lazy_prince | 02:55 | |
| *** Marga_ has quit IRC | 03:02 | |
| *** Marga_ has joined #openstack-ironic | 03:03 | |
| *** leopoldj_ has joined #openstack-ironic | 03:19 | |
| *** leopoldj has quit IRC | 03:22 | |
| *** zz_naotok is now known as naotok | 03:24 | |
| *** jrist has quit IRC | 03:26 | |
| *** todd_dsm has quit IRC | 03:37 | |
| *** todd_dsm has joined #openstack-ironic | 03:41 | |
| *** ryanpetrello has joined #openstack-ironic | 03:43 | |
| *** nosnos has quit IRC | 03:46 | |
| *** nosnos has joined #openstack-ironic | 03:46 | |
| *** todd_dsm has quit IRC | 03:48 | |
| *** nosnos has quit IRC | 03:51 | |
| *** todd_dsm has joined #openstack-ironic | 03:51 | |
| *** ujuc has joined #openstack-ironic | 03:54 | |
| *** todd_dsm has quit IRC | 03:57 | |
| *** dlaube has joined #openstack-ironic | 04:00 | |
| *** dlaube has quit IRC | 04:02 | |
| *** marcoemorais has joined #openstack-ironic | 04:16 | |
| *** spandhe has joined #openstack-ironic | 04:16 | |
| *** marcoemorais1 has joined #openstack-ironic | 04:21 | |
| *** harlowja is now known as harlowja_away | 04:21 | |
| *** marcoemorais has quit IRC | 04:23 | |
| *** nosnos has joined #openstack-ironic | 04:27 | |
| *** ramineni has joined #openstack-ironic | 04:30 | |
| openstackgerrit | jiangfei proposed a change to openstack/ironic-python-agent: modify for CONF Fixes: bug 1386486 https://review.openstack.org/131632 | 04:31 |
| *** ryanpetrello has quit IRC | 04:34 | |
| *** Poornima has quit IRC | 04:34 | |
| *** pcrews has quit IRC | 04:37 | |
| *** todd_dsm has joined #openstack-ironic | 04:41 | |
| *** pcrews has joined #openstack-ironic | 04:41 | |
| *** chenglch has quit IRC | 04:46 | |
| *** achanda has joined #openstack-ironic | 04:49 | |
| *** chenglch has joined #openstack-ironic | 04:53 | |
| *** pcrews has quit IRC | 04:57 | |
| *** Poornima has joined #openstack-ironic | 05:05 | |
| *** todd_dsm has quit IRC | 05:08 | |
| *** todd_dsm has joined #openstack-ironic | 05:08 | |
| *** todd_dsm has quit IRC | 05:22 | |
| *** pensu has joined #openstack-ironic | 05:24 | |
| *** pcrews has joined #openstack-ironic | 05:26 | |
| *** rushiagr_away is now known as rushiagr | 05:29 | |
| *** pensu has quit IRC | 05:29 | |
| *** rakesh_hs has joined #openstack-ironic | 05:31 | |
| *** pcrews has quit IRC | 05:32 | |
| *** pensu has joined #openstack-ironic | 05:32 | |
| openstackgerrit | Harshada Mangesh Kakad proposed a change to openstack/ironic-specs: Seamicro Serial Console https://review.openstack.org/131382 | 05:36 |
| *** loki184 has joined #openstack-ironic | 05:44 | |
| *** achanda has quit IRC | 06:12 | |
| *** achanda has joined #openstack-ironic | 06:14 | |
| *** spandhe has quit IRC | 06:20 | |
| *** achanda has quit IRC | 06:20 | |
| *** achanda has joined #openstack-ironic | 06:25 | |
| *** lazy_prince has quit IRC | 06:38 | |
| *** achanda has quit IRC | 06:39 | |
| *** Marga_ has quit IRC | 06:46 | |
| *** killer_prince has joined #openstack-ironic | 06:46 | |
| *** chenglch|2 has joined #openstack-ironic | 06:46 | |
| *** killer_prince is now known as lazy_prince | 06:46 | |
| *** chenglch|2 has left #openstack-ironic | 06:46 | |
| *** lazy_prince has quit IRC | 06:50 | |
| *** killer_prince has joined #openstack-ironic | 06:59 | |
| *** killer_prince is now known as lazy_prince | 06:59 | |
| *** lazy_prince has quit IRC | 07:20 | |
| *** lazy_prince has joined #openstack-ironic | 07:21 | |
| *** nosnos has quit IRC | 07:25 | |
| *** nosnos has joined #openstack-ironic | 07:27 | |
| *** leopoldj_ has quit IRC | 07:29 | |
| *** dtantsur|afk is now known as dtantsur | 07:36 | |
| dtantsur | Morning Ironic | 07:37 |
| *** teju has joined #openstack-ironic | 07:46 | |
| teju | Haomeng : morning :) | 07:48 |
| *** foexle has joined #openstack-ironic | 07:51 | |
| *** lazy_prince has quit IRC | 07:58 | |
| *** RioS2 has quit IRC | 08:05 | |
| *** killer_prince has joined #openstack-ironic | 08:05 | |
| *** killer_prince is now known as lazy_prince | 08:06 | |
| *** jcoufal has joined #openstack-ironic | 08:07 | |
| *** lazy_prince has quit IRC | 08:10 | |
| *** andreykurilin_ has joined #openstack-ironic | 08:12 | |
| *** killer_prince has joined #openstack-ironic | 08:19 | |
| *** killer_prince is now known as lazy_prince | 08:19 | |
| *** kongfy has joined #openstack-ironic | 08:22 | |
| *** RioS2 has joined #openstack-ironic | 08:23 | |
| *** rushiagr is now known as rushiagr_away | 08:27 | |
| *** k4n0 has joined #openstack-ironic | 08:28 | |
| openstackgerrit | A change was merged to openstack/ironic: Update install guide to install tftp https://review.openstack.org/131547 | 08:29 |
| *** ndipanov_gone is now known as ndipanov | 08:29 | |
| *** lazy_prince has quit IRC | 08:39 | |
| *** lazy_prince has joined #openstack-ironic | 08:39 | |
| *** rushiagr_away is now known as rushiagr | 08:39 | |
| *** andreykurilin_ has quit IRC | 08:47 | |
| Haomeng | teju: morning:) | 08:51 |
| Haomeng | teju: how are you | 08:51 |
| teju | Haomeng : i'm cool..how r u & howz Ironic? | 08:52 |
| ekarlso | morning :D | 08:54 |
| Haomeng | teju: fine:) | 08:55 |
| Haomeng | ekarlso: morning:) | 08:55 |
| *** yuanying has quit IRC | 08:57 | |
| *** jistr has joined #openstack-ironic | 09:06 | |
| *** derekh has joined #openstack-ironic | 09:07 | |
| *** ujuc has quit IRC | 09:12 | |
| *** leopoldj has joined #openstack-ironic | 09:13 | |
| *** lucasagomes has joined #openstack-ironic | 09:24 | |
| *** kongfy has quit IRC | 09:24 | |
| *** igordcard has joined #openstack-ironic | 09:27 | |
| *** athomas has joined #openstack-ironic | 09:28 | |
| *** pradipta_away is now known as pradipta | 09:39 | |
| *** achanda has joined #openstack-ironic | 09:40 | |
| *** marcoemorais1 has quit IRC | 09:41 | |
| *** lazy_prince has quit IRC | 09:43 | |
| *** pelix has joined #openstack-ironic | 09:44 | |
| *** achanda has quit IRC | 09:44 | |
| *** mkerrin has quit IRC | 09:49 | |
| *** mkerrin has joined #openstack-ironic | 09:54 | |
| *** chenglch has quit IRC | 10:07 | |
| *** Marga_ has joined #openstack-ironic | 10:11 | |
| yuriyz | morning all | 10:23 |
| yuriyz | lucasagomes, old client bug again https://bugs.launchpad.net/ironic/+bug/1387101 | 10:25 |
| yuriyz | fresh Devstack install on the clean OS | 10:26 |
| *** rushiagr is now known as rushiagr_away | 10:39 | |
| lucasagomes | yuriyz, morning, will take a look | 10:45 |
| *** Marga_ has quit IRC | 10:46 | |
| *** Marga_ has joined #openstack-ironic | 10:46 | |
| lucasagomes | yuriyz, I think devananda have spotted that problem yesterday https://bugs.launchpad.net/python-ironicclient/+bug/1386470 | 10:49 |
| lucasagomes | I will take a look at it | 10:49 |
| *** killer_prince has joined #openstack-ironic | 10:54 | |
| *** killer_prince is now known as lazy_prince | 10:54 | |
| yuriyz | thanks lucasagomes | 10:55 |
| *** sambetts has joined #openstack-ironic | 10:56 | |
| *** athomas has quit IRC | 10:58 | |
| *** ifarkas has joined #openstack-ironic | 10:58 | |
| yuriyz | and this bug reproduced with OS_USERNAME=admin in my environment | 10:59 |
| yuriyz | for error messages | 11:01 |
| *** loki184 has quit IRC | 11:01 | |
| sambetts | Good Morning Ironic! | 11:03 |
| lucasagomes | sambetts, morning | 11:04 |
| *** Marga_ has quit IRC | 11:05 | |
| *** ramineni has quit IRC | 11:05 | |
| *** athomas has joined #openstack-ironic | 11:06 | |
| sambetts | morning lucasagomes | 11:07 |
| *** boris-42 has quit IRC | 11:11 | |
| *** rameshg87 has joined #openstack-ironic | 11:17 | |
| *** teju has quit IRC | 11:19 | |
| *** kongfy has joined #openstack-ironic | 11:24 | |
| *** rushiagr_away is now known as rushiagr | 11:25 | |
| *** pradipta is now known as pradipta_away | 11:28 | |
| *** Poornima has quit IRC | 11:37 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add a mechanism to route vendor methods https://review.openstack.org/129261 | 11:42 |
| lucasagomes | sambetts, ^ using ur patch as base :) | 11:42 |
| lucasagomes | will write the bp now | 11:42 |
| lucasagomes | (wanted to know how it would look like first) | 11:42 |
| sambetts | Nice! | 11:42 |
| *** lazy_prince is now known as killer_prince | 11:44 | |
| sambetts | I'm just updating my patch with the comments you left | 11:45 |
| lucasagomes | sambetts, yeah I applied the functools.wraps on my patch, I will rebase on top of urs again once u fix it there | 11:46 |
| lucasagomes | no rush :) | 11:46 |
| *** ryanpetrello has joined #openstack-ironic | 11:53 | |
| *** Poornima has joined #openstack-ironic | 11:54 | |
| *** pensu has quit IRC | 11:57 | |
| *** naotok is now known as zz_naotok | 12:07 | |
| *** dprince has joined #openstack-ironic | 12:10 | |
| dtantsur | What do you folks think about having periodic tasks for drivers? E.g. any driver could declare it's periodic task... | 12:12 |
| *** nosnos has quit IRC | 12:15 | |
| *** rameshg87 has left #openstack-ironic | 12:17 | |
| *** killer_prince has quit IRC | 12:20 | |
| lucasagomes | dtantsur, what is the use case? But to be honest in general my first reaction is urgh | 12:21 |
| lucasagomes | not because it's a bad idea | 12:21 |
| lucasagomes | but because periodic tasks in openstack sucks so badly | 12:21 |
| lucasagomes | if they at least could run in parallel | 12:21 |
| lucasagomes | but they don't, they run one after the other in a single greenthread | 12:22 |
| dtantsur | lucasagomes, it's for DRAC RAID spec (see comments): https://review.openstack.org/#/c/107981/9/specs/juno/drac-raid-mgmt.rst | 12:22 |
| dtantsur | lucasagomes, yeah, implementation sucks, but the very idea looks sane to me | 12:23 |
| lucasagomes | oh I see | 12:23 |
| lucasagomes | yeah, I see it being useful | 12:23 |
| *** shikui__ has joined #openstack-ironic | 12:28 | |
| dtantsur | lucasagomes, btw why aren't periodic tasks fixed? It should not be too hard to move them to different greenthreads... | 12:34 |
| openstackgerrit | Sam Betts proposed a change to openstack/ironic: Add logging to driver vendor_passthru functions https://review.openstack.org/129298 | 12:35 |
| lucasagomes | dtantsur, no idea, I bet that for now it may be "okish" to be the way it is | 12:36 |
| lucasagomes | also it's still part of the oslo incubation, it doesn't have it's own library | 12:37 |
| lucasagomes | maybe someone is working on that and improving it at the same time (just like the oslo.config stuff) | 12:37 |
| lucasagomes | worth investigation, idk exactly, these are just assumptions | 12:37 |
| dtantsur | lucasagomes, one more topic to discuss, if we have some time | 12:37 |
| lucasagomes | yeah, dhellmann might know it | 12:37 |
| openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic-specs: DRAC vendor passthru for RAID management https://review.openstack.org/107981 | 12:39 |
| lucasagomes | yuriyz, :/ I did a fresh install but can't reproduce the error | 12:43 |
| lucasagomes | yuriyz, http://paste.openstack.org/show/126464/ | 12:43 |
| yuriyz | lucasagomes, try request with missing parameter, ironic node-create for example w/o driver | 12:45 |
| lucasagomes | ack | 12:46 |
| lucasagomes | stack@virtual-machine:~/devstack$ ironic node-create -i test=test | 12:46 |
| lucasagomes | Authentication failure: Could not find user: dsasddsa (Disable debug mode to suppress these details.) (HTTP 401) | 12:46 |
| lucasagomes | yuriyz, I will investigate more, but seems grand | 12:46 |
| yuriyz | with valid user | 12:46 |
| lucasagomes | yuriyz, ohhhh there you go | 12:47 |
| lucasagomes | stack@virtual-machine:~/devstack$ ironic node-create -i test=test | 12:47 |
| lucasagomes | 'unicode' object has no attribute 'get' | 12:47 |
| lucasagomes | yuriyz, cheers | 12:47 |
| *** Krast has joined #openstack-ironic | 12:48 | |
| *** kongfy has quit IRC | 12:49 | |
| sambetts | lucasagomes: What do you think about dtantsur's comments on my patch https://review.openstack.org/#/c/129298/9/ironic/drivers/base.py? I like the idea of changing the name to @passthru instead of @register_passthru, and I think we could make the method param optional, if not provided take the func.__name__ minus the leading '_', ideas? | 12:49 |
| lucasagomes | sambetts, I like the idea of having a shorter name @passthru and @driver_passthru seems good to me | 12:50 |
| lucasagomes | sambetts, right, about making it optional hmm, I kinda like to be explicit because that's the method name people are adding to the HTTP request | 12:51 |
| lucasagomes | v1/nodes/<uuid>vendor_passthru/<method name> | 12:51 |
| sambetts | yeah thats what I had thought | 12:51 |
| lucasagomes | sambetts, I wouldn't be against taking it from __name__ but, I think explicit may be clearer | 12:52 |
| sambetts | dtantsur, ideas? | 12:52 |
| dtantsur | it's overexplicit to me | 12:52 |
| dtantsur | @passthru('method') def method() | 12:52 |
| lucasagomes | dtantsur, right, so use __name__ but not remove the prefixed '_' right? | 12:53 |
| lucasagomes | dtantsur, some functions are not like that as well... see pxe driver | 12:53 |
| dtantsur | I don't like leading _ anyway, these methods are not private | 12:53 |
| dtantsur | lucasagomes, if name does change, we can use this optional argument | 12:54 |
| lucasagomes | the vendor method is called "pass_deploy_info", and the method itself is called "_continue_deploy" | 12:54 |
| sambetts | there is an example in the ilo driver where its @passthru('pass_deploy_info') def _set_boot_device | 12:54 |
| dtantsur | IMO them we should rename the method, thus making code more clear | 12:54 |
| lucasagomes | dtantsur, right, yeah I'm not against it | 12:54 |
| dtantsur | pass_deploy_info vs _set_boot_device looks somewhat weird to me :) | 12:55 |
| lucasagomes | dtantsur, yeah, it's because the ilo is inherting from pxe | 12:55 |
| lucasagomes | so all it does it set the boot device, and then call the parents _continue_deploy | 12:55 |
| dtantsur | lucasagomes, mmm, this inheritance thing... I have concerns about it | 12:56 |
| dtantsur | at least we'll have double error handling | 12:56 |
| lucasagomes | dtantsur, yea me too, but if we talk about it we are diverting the conversation focus | 12:56 |
| dtantsur | right | 12:56 |
| dtantsur | so I think we should try to use method name as often as it's possible and leave 'method' as optional argument | 12:57 |
| lucasagomes | dtantsur, sambetts so to be honest I see benefits on both sides, and could argue about both | 12:57 |
| lucasagomes | dtantsur, idea of making it option, and still provide a way to reuse it if needed sounds good to me | 12:57 |
| sambetts | thats why I'm thinking optional | 12:58 |
| lucasagomes | to rename it* | 12:58 |
| dtantsur | lgtm too | 12:58 |
| lucasagomes | right, I will add it to the spec | 12:58 |
| sambetts | what did we decide about the leading underscore? | 12:58 |
| lucasagomes | I don't like the idea of removing it from the name if it exists | 12:59 |
| dtantsur | I don't like that it exists :) | 12:59 |
| lucasagomes | imo that's where things get too obscured | 12:59 |
| dtantsur | these are not private methods, which the name implies | 12:59 |
| dtantsur | they are used from outside | 12:59 |
| lucasagomes | I'm more into renaming those functions and remove the leading _ than removing it as part of the decorator | 12:59 |
| dtantsur | so @passthru() def pass_deploy_info() right? | 13:00 |
| lucasagomes | dtantsur, well you still can rename it and keep the _continue deploy | 13:00 |
| sambetts | yeah | 13:00 |
| lucasagomes | in case it's going to affect too many tests | 13:00 |
| lucasagomes | but yeah... | 13:00 |
| dtantsur | lunch time, brb | 13:01 |
| lucasagomes | enjoy, I will eat something quickly too | 13:01 |
| *** lucasagomes is now known as lucas-hungry | 13:01 | |
| jroll | dtantsur, lucas-hungry, sambetts, please make it explicit | 13:03 |
| jroll | don't couple the method name to the API | 13:03 |
| jroll | a simple refactoring could end up backwards-incompatible | 13:03 |
| jroll | we went through this discussion in ipa already :| | 13:04 |
| jroll | also.. dtantsur ++ for drivers defining periodic tasks | 13:04 |
| jroll | also also... good morning all :) | 13:04 |
| *** kongfy has joined #openstack-ironic | 13:11 | |
| *** Haomeng has quit IRC | 13:13 | |
| dtantsur | jroll, we already have RPC using method names | 13:13 |
| dtantsur | and g'morning of course :) | 13:14 |
| *** Haomeng has joined #openstack-ironic | 13:15 | |
| *** mjturek has joined #openstack-ironic | 13:18 | |
| jroll | dtantsur: not for passthru | 13:19 |
| dtantsur | jroll, right, but there "a simple refactoring could end up backwards-incompatible" | 13:20 |
| jroll | sigh. | 13:21 |
| jroll | however, just because one thing sucks, doesn't mean we should make another thing suck | 13:22 |
| jroll | and this is a user-facing api | 13:22 |
| jroll | RPC is only driver-facing | 13:22 |
| dtantsur | well, maybe you're right and it's not such a big duplication (though it also does not look a huge problem to me) | 13:23 |
| jroll | dtantsur: another example... | 13:25 |
| jroll | ilo has a set_boot_device() call | 13:26 |
| jroll | that is used from passthru, and not from passthru | 13:26 |
| jroll | so the one called from passthru gets the good name if we couple to names | 13:26 |
| jroll | whereas the one that actually does stuff needs a different name | 13:27 |
| dtantsur | jroll, it anyway needs, because we'll be baking exception handling inside the @decorator. and caller may expect it to raise | 13:28 |
| jroll | needs what? | 13:28 |
| dtantsur | needs a different name | 13:29 |
| jroll | right | 13:29 |
| jroll | https://review.openstack.org/#/c/129298/9/ironic/drivers/modules/ilo/deploy.py | 13:29 |
| jroll | I guess dropping the _ would be fine | 13:29 |
| dtantsur | jroll, is this call used anywhere else? I would suppose ilo_common.set_boot_device is used | 13:31 |
| jroll | yeah, you're right | 13:31 |
| jroll | I just remembered because we were talking about how to decorate that | 13:31 |
| openstackgerrit | Sam Betts proposed a change to openstack/ironic: Add logging to driver vendor_passthru functions https://review.openstack.org/129298 | 13:31 |
| sambetts | I've only implemented the shorten decorator name until we decide what we want to do with the method param | 13:32 |
| jroll | I'd really prefer that being explicit | 13:36 |
| jroll | but that's just me | 13:36 |
| *** igordcard has quit IRC | 13:37 | |
| sambetts | I think I prefer the explictness because the user facing API may wish describe things differently to the code/function name, if its optional it isn't tied so staticly to the function name it just reduces the duplication in instances when they would be the same | 13:39 |
| jroll | I agree | 13:41 |
| jroll | optional might be ok | 13:41 |
| sambetts | I'll put together a patch which has it as an optional param so we can see what it looks like and if its no good then we can easily revert it | 13:42 |
| *** Krast has quit IRC | 13:43 | |
| jroll | cool, ty | 13:44 |
| *** rushiagr is now known as rushiagr_away | 13:47 | |
| *** stendulker has joined #openstack-ironic | 13:53 | |
| *** lucas-hungry is now known as lucasagomes | 13:56 | |
| *** romcheg has joined #openstack-ironic | 13:56 | |
| lucasagomes | jroll, morning | 13:56 |
| jroll | heya lucas | 13:57 |
| stendulker | Hello | 13:57 |
| stendulker | I get following error during deploy | 13:58 |
| stendulker | ERROR ironic.drivers.modules.iscsi_deploy [-] Deploy failed for instance None. Error: Unexpected error while running command. Command: sudo ironic-rootwrap /etc/ironic/rootwrap.conf dd if=/var/lib/ironic/images/f785923c-dfea-4d01-af2b-c26af86111ca/disk of=/dev/disk/by-path/ip-10.10.1.217:3260-iscsi-iqn-f785923c-dfea-4d01 -af2b-c26af86111ca-lun-1-part1 bs=1M oflag=direct Exit code: 1 Stdout: '' Stderr: "/bin/dd: error w | 13:58 |
| jroll | stendulker: you cut off at Stderr: "/bin/dd: error w | 13:58 |
| jroll | is there more? | 13:59 |
| jroll | stderr seems relevant here :P | 13:59 |
| stendulker | jroll : No further error | 13:59 |
| jroll | really? | 13:59 |
| jroll | seems wrong. | 14:00 |
| stendulker | Is there anything to be modified in /etc/ironic/rootwrap.conf. I have a default settings in the same | 14:00 |
| jroll | it looks like dd ran, so rootwrap probably worked | 14:00 |
| jroll | but I can't see why dd failed | 14:00 |
| jroll | so I really can't help | 14:00 |
| jroll | there must be more | 14:00 |
| stendulker | 2014-10-29 22:51:36.823 DEBUG ironic.common.utils [-] Command stderr is: "" from (pid=8835) execute /opt/stack/ironic/ironic/common/utils.py:86 2014-10-29 22:51:36.825 ERROR ironic.drivers.modules.iscsi_deploy [-] Deploy failed for instance None. Error: Unexpected error while running command. Command: sudo ironic-rootwrap /etc/ironic/rootwrap.conf dd if=/var/lib/ironic/images/f785923c-dfea-4d01-af2b-c26af86111ca/disk of | 14:01 |
| stendulker | Nothing more.. | 14:01 |
| stendulker | 2014-10-29 22:51:36.823 DEBUG ironic.common.utils [-] Command stderr is: "" from (pid=8835) execute /opt/stack/ironic/ironic/common/utils.py:86 2014-10-29 22:51:36.825 ERROR ironic.drivers.modules.iscsi_deploy [-] Deploy failed for instance None. Error: Unexpected error while running command. Command: sudo ironic-rootwrap /etc/ironic/rootwrap.conf dd if=/var/lib/ironic/images/f785923c-dfea-4d01-af2b-c26af86111ca/disk of | 14:01 |
| stendulker | I think my messages are getting truncated | 14:02 |
| jroll | right... maybe by your terminal or whatever application you're using to read the logs? | 14:03 |
| jroll | they shouldn't be truncated on disk | 14:03 |
| stendulker | 2014-10-29 22:51:36.823 DEBUG ironic.common.utils [-] Command stderr is: "" from (pid=8835) execute /opt/stack/ironic/ironic/common/utils.py:86 | 14:04 |
| stendulker | 2014-10-29 22:51:36.825 ERROR ironic.drivers.modules.iscsi_deploy [-] Deploy failed for instance None. Error: Unexpected error while running command. | 14:04 |
| stendulker | Command: sudo ironic-rootwrap /etc/ironic/rootwrap.conf dd if=/var/lib/ironic/images/f785923c-dfea-4d01-af2b-c26af86111ca/disk of=/dev/disk/by-path/ | 14:04 |
| stendulker | ip-10.10.1.217:3260-iscsi-iqn-f785923c-dfea-4d01-af2b-c26af86111ca-lun-1-part1 bs=1M oflag=direct Exit code: 1 | 14:04 |
| stendulker | Stdout: '' | 14:04 |
| stendulker | Stderr: "/bin/dd: error writing '/dev/disk/by-path/ip-10.10.1.217:3260-iscsi-iqn-f785923c-dfea-4d01-af2b-c26af86111ca-lun-1-part1': | 14:05 |
| stendulker | Input/output error\n1128+0 records in\n1127+0 records out\n1181745152 bytes (1.2 GB) copied, 264.387 s, 4.5 MB/s\n" | 14:05 |
| stendulker | 2014-10-29 22:51:37.068 WARNING ironic.conductor.utils [-] Not going to change_node_power_state because current state = requested state = 'power off'. | 14:05 |
| jroll | perfect | 14:05 |
| stendulker | These are the logs | 14:05 |
| jroll | let's use paste.openstack.org next time, but that's great | 14:05 |
| jroll | so I see Input/output error\n1128+0 records in\n1127+0 records out | 14:06 |
| stendulker | sure. was not aware if paste.openstack.org | 14:06 |
| jroll | so looks like 1) error writing to the disk, or 2) error with iscsi | 14:06 |
| jroll | or both | 14:06 |
| stendulker | Anything I can try for further investigation | 14:08 |
| jroll | I am not an expert on testing disks, and have never used iSCSI | 14:08 |
| *** todd_dsm has joined #openstack-ironic | 14:08 | |
| jroll | but perhaps someone else can help :) | 14:09 |
| stendulker | jroll : ok. thanks for looking at it. | 14:09 |
| jroll | np :) | 14:09 |
| openstackgerrit | Sam Betts proposed a change to openstack/ironic: Add logging to driver vendor_passthru functions https://review.openstack.org/129298 | 14:14 |
| sambetts | lucasagomes, dtantsur, jroll: ^ optional method param | 14:15 |
| *** Marga_ has joined #openstack-ironic | 14:15 | |
| jroll | whee | 14:15 |
| * dtantsur is looking | 14:15 | |
| *** foexle has quit IRC | 14:15 | |
| NobodyCam | good morning Ironic | 14:16 |
| jroll | morning NobodyCam | 14:17 |
| jroll | hmm | 14:17 |
| jroll | sambetts: the idea looks fine... but this is breaking APIs | 14:17 |
| jroll | e.g. https://review.openstack.org/#/c/129298/11/ironic/drivers/modules/ipmitool.py | 14:17 |
| jroll | oh wait I'm an idiot | 14:18 |
| jroll | yeah ignore me | 14:18 |
| jroll | :) | 14:18 |
| dtantsur | jroll, sambetts I'm more worried about using new decorator above @task_manager.require_exclusive_lock | 14:18 |
| jroll | yeah, should probably be below | 14:18 |
| jroll | wrap the whole thing in a lock | 14:19 |
| *** Haomeng|2 has joined #openstack-ironic | 14:19 | |
| dtantsur | jroll, 1. yes, 2. task_manager should also be fixed to use @functools.wraps | 14:19 |
| sambetts | really? I would have thought you want to keep the lock as close to the code that needs the locked code as possible? | 14:19 |
| dtantsur | sambetts, I think fixing the lock decorator should be enough | 14:20 |
| jroll | actually | 14:20 |
| jroll | it might be fine | 14:20 |
| jroll | the new decorator doesn't do anything that needs a lock | 14:20 |
| NobodyCam | morning jroll, dtantsur | 14:20 |
| *** Haomeng has quit IRC | 14:20 | |
| dtantsur | NobodyCam, morning! | 14:21 |
| *** zz_jgrimm is now known as jgrimm | 14:22 | |
| sambetts | jroll: that is what I thought, logging and exceptions are all thread/process safe, so I decided to make my decorator wrap the already locked code | 14:22 |
| sambetts | thats interesting about it losing the __name__ though | 14:23 |
| jroll | does it? | 14:25 |
| *** rloo has joined #openstack-ironic | 14:26 | |
| dtantsur | jroll, without @functools.wraps - it should (to "wrapper") | 14:26 |
| jroll | mmm | 14:26 |
| jroll | that's right | 14:26 |
| sambetts | Yeah, I'm just adding the .wraps to the wrapper function | 14:27 |
| openstackgerrit | Sam Betts proposed a change to openstack/ironic: Add logging to driver vendor_passthru functions https://review.openstack.org/129298 | 14:29 |
| sambetts | hows that ^ | 14:29 |
| * dtantsur is looking again | 14:30 | |
| dtantsur | sambetts, lgtm | 14:33 |
| sambetts | woop :D | 14:33 |
| jroll | sambetts: hmm, looking at https://review.openstack.org/#/c/129298/12/ironic/tests/drivers/test_agent.py | 14:37 |
| jroll | would doing mock.patch.object(..., 'heartbeat') avoid the need to set __name__ etc | 14:37 |
| jroll | just curious, change lgtm either way | 14:38 |
| jroll | maybe we can clean those tests up later | 14:38 |
| jroll | +2 | 14:39 |
| jroll | brb | 14:39 |
| *** Marga_ has quit IRC | 14:49 | |
| *** todd_dsm has quit IRC | 14:55 | |
| *** jcoufal has quit IRC | 14:58 | |
| *** jcoufal has joined #openstack-ironic | 15:01 | |
| *** todd_dsm has joined #openstack-ironic | 15:03 | |
| *** rushiagr_away is now known as rushiagr | 15:08 | |
| lucasagomes | dtantsur, I'm seem a odd error with the decorator thing, you may know | 15:08 |
| lucasagomes | dtantsur, when I call the function reference from the map I generated in the decorator, I'm getting a "_private_method() takes exactly 3 arguments (2 given) (HTTP 500)" | 15:09 |
| lucasagomes | basically the self parameter is not being automatically passed to that function | 15:09 |
| lucasagomes | any ideas? | 15:09 |
| jroll | mmm | 15:09 |
| jroll | you need to bind self | 15:09 |
| jroll | in the decorator iirc | 15:09 |
| lucasagomes | jroll, right hmm | 15:10 |
| lucasagomes | jroll, https://review.openstack.org/#/c/129261/11/ironic/drivers/base.py | 15:10 |
| lucasagomes | L372 | 15:10 |
| lucasagomes | that how I was saving the reference | 15:10 |
| jroll | right | 15:11 |
| jroll | so I think passthru_handler can take self and pass it on to func() | 15:11 |
| jroll | but I guess *args would do that | 15:11 |
| lucasagomes | so I should take self, *args, **kwargs? | 15:11 |
| lucasagomes | me too | 15:12 |
| lucasagomes | apparently it doesn't | 15:12 |
| lucasagomes | lemme see | 15:12 |
| jroll | maybe | 15:12 |
| *** pcrews has joined #openstack-ironic | 15:12 | |
| *** shikui__ has quit IRC | 15:14 | |
| lucasagomes | same :/ | 15:17 |
| lucasagomes | damn | 15:17 |
| jroll | huh. | 15:19 |
| *** kongfy has quit IRC | 15:21 | |
| *** pensu has joined #openstack-ironic | 15:21 | |
| jroll | oh lucasagomes | 15:22 |
| jroll | you might be able to like... bind self in the decorator | 15:22 |
| jroll | https://github.com/openstack/ironic-python-agent/blob/master/ironic_python_agent/extensions/base.py#L211 | 15:22 |
| jroll | so passthru_handler should take self | 15:22 |
| jroll | func = functools.partial(func, self) | 15:23 |
| lucasagomes | oh | 15:23 |
| lucasagomes | lemme try that | 15:23 |
| *** Marga_ has joined #openstack-ironic | 15:23 | |
| jroll | since you aren't calling it with object.method() | 15:23 |
| jroll | you're just calling method() | 15:23 |
| jroll | I think that should work | 15:23 |
| openstackgerrit | Ilya Pekelny proposed a change to openstack/ironic: Test migrations with Alembic, using Oslo.db https://review.openstack.org/111984 | 15:23 |
| *** hemna__ is now known as hemna | 15:30 | |
| lucasagomes | jroll, nop :( | 15:31 |
| dtantsur | lucasagomes, hmmm.. it should work w/o self, as self is already bound... | 15:34 |
| dtantsur | where does this error happen? | 15:34 |
| dtantsur | ah no, self is not bound | 15:35 |
| lucasagomes | dtantsur, in the worker thread | 15:35 |
| lucasagomes | dtantsur, I could reproduce here making a test code | 15:35 |
| lucasagomes | lemme give it to ya | 15:35 |
| lucasagomes | dtantsur, jroll http://paste.openstack.org/show/126536/ | 15:35 |
| lucasagomes | yeah I invoke ha( ... ) same error happens | 15:36 |
| dtantsur | lucasagomes, lines 20 and 22 bound self twice | 15:36 |
| dtantsur | actually line 20 is not needed | 15:36 |
| sambetts | lucasagomes: I think you could partial it out in the VendorPassThru init when you set the dict, you just make sure that every element has partial(func, self) instead of func?? | 15:37 |
| lucasagomes | dtantsur, yeah that was a experiment, but removing it fails same way | 15:37 |
| dtantsur | lucasagomes, I can hardly belieave it's _the same_ way, it should be at least a bit different | 15:38 |
| lucasagomes | sambetts, I don't have access to self at that point :/ | 15:38 |
| sambetts | lucasagomes: you have self in the init function ... | 15:38 |
| sambetts | def __init__(self) <- | 15:39 |
| lucasagomes | oh I see what u mean | 15:39 |
| sambetts | inject it at that point | 15:39 |
| dtantsur | lucasagomes, I see several things wrong there, lemme try to fix | 15:39 |
| lucasagomes | yeah this is an experimentation code | 15:40 |
| lucasagomes | please try :) | 15:40 |
| dtantsur | lucasagomes, http://paste.openstack.org/show/126539/ | 15:44 |
| lucasagomes | ha(b, 'aaa', abc='mole') | 15:44 |
| lucasagomes | so I need access to the class instance | 15:45 |
| dtantsur | lucasagomes, of course, you need self to call a method :) | 15:45 |
| dtantsur | lucasagomes, or you can store partial in metadata | 15:46 |
| dtantsur | hmm no you can't | 15:46 |
| dtantsur | so yes, you need self to pass | 15:46 |
| lucasagomes | dtantsur, I thought I could get a ref of the class method and self would be passed implicitly | 15:46 |
| lucasagomes | oh dear | 15:47 |
| lucasagomes | :/ | 15:47 |
| dtantsur | lucasagomes, problems is: you don't get a ref to an instance (not class) method | 15:47 |
| dtantsur | lucasagomes, decorator work before binding | 15:47 |
| lucasagomes | yeah I see that now | 15:47 |
| lucasagomes | hmm that actually sucks | 15:47 |
| *** ifarkas has quit IRC | 15:48 | |
| lucasagomes | wondering if I should look at __class__ | 15:48 |
| *** todd_dsm has quit IRC | 15:48 | |
| dtantsur | lucasagomes, you have an instance at the call place, what's the problem? | 15:49 |
| dtantsur | https://review.openstack.org/#/c/129261/11/ironic/conductor/manager.py line 410 you have vendor_iface | 15:49 |
| dtantsur | vendor_iface is self for the method | 15:49 |
| *** todd_dsm has joined #openstack-ironic | 15:49 | |
| lucasagomes | dtantsur, yeah that's fine, but can be trick when it comes to Mixin | 15:49 |
| dtantsur | right | 15:50 |
| lucasagomes | I could do check if that's mixin interface and get the real class instance from the map | 15:51 |
| lucasagomes | but that looks pretty horrible | 15:51 |
| dtantsur | oh well | 15:51 |
| dtantsur | that's why I don't like overusing decorators actually... | 15:52 |
| *** todd_dsm has quit IRC | 15:52 | |
| lucasagomes | yeah | 15:53 |
| lucasagomes | dtantsur, sounds a bit odd, but another way to do it would be to use a metaclass to create the routes | 15:54 |
| dtantsur | lucasagomes, I 99% prefer class decorator to a metaclass, but yeah, it's possible | 15:54 |
| rloo | or no decorator (sorry, haven't been following but I see the 'help me' signs) | 15:55 |
| *** todd_dsm has joined #openstack-ironic | 15:55 | |
| lucasagomes | vendor classes are inheriting from a base class | 15:55 |
| lucasagomes | so a metaclass makes more sense imo | 15:55 |
| lucasagomes | rloo, yeah or no decorator and back to manual mapping | 15:55 |
| lucasagomes | which is fine too | 15:55 |
| dtantsur | I guess jroll would call metaclass approach too implicit and I tend to agree :) | 15:56 |
| rloo | lucasagomes: I'm fine with either. If going with decorators might make code/debugging harder to understand, I would say make it simple. | 15:56 |
| lucasagomes | dtantsur, well it's already implicit | 15:56 |
| lucasagomes | I don't see the diference | 15:56 |
| lucasagomes | rloo, yeah the manual explicity vendor_routes and driver_routes | 15:57 |
| dtantsur | lucasagomes, you see decorator on every function. with metaclass you see nothing (or you want metaclass + decorator?) | 15:57 |
| lucasagomes | is way simpler to debug | 15:57 |
| lucasagomes | dtantsur, functions will have a decorator | 15:57 |
| lucasagomes | lemme try to make a sample of what I'm thinking here | 15:58 |
| *** jinxly has quit IRC | 15:58 | |
| openstackgerrit | Jim Mankovich proposed a change to openstack/ironic-specs: Support per Driver Sensor Meters https://review.openstack.org/130074 | 15:59 |
| openstackgerrit | Ilya Pekelny proposed a change to openstack/ironic: Test migrations with Alembic, using Oslo.db https://review.openstack.org/111984 | 16:03 |
| *** k4n0 has quit IRC | 16:03 | |
| *** Poornima has quit IRC | 16:06 | |
| lucasagomes | dtantsur, http://paste.openstack.org/show/126558/ | 16:06 |
| lucasagomes | jroll, ^ | 16:06 |
| lucasagomes | rloo, ^ u too :) | 16:07 |
| dtantsur | hmm looks nice | 16:07 |
| jroll | metaclasses make me cry a little | 16:08 |
| *** todd_dsm has quit IRC | 16:09 | |
| *** dlaube has joined #openstack-ironic | 16:09 | |
| lucasagomes | well | 16:09 |
| lucasagomes | I also dislike metaclasses, but I don't really know a better way to solve it | 16:09 |
| lucasagomes | I may, but I'm tired at this point too :P | 16:10 |
| jroll | yeah, agree | 16:10 |
| jroll | I mean, whatever works | 16:10 |
| lucasagomes | jroll, we could also not use a decorator as rloo pointed out, and have that explicit routes as the spec says | 16:10 |
| jroll | oh, right | 16:11 |
| rloo | to be honest, I'd rather get rid of the decorators then. | 16:11 |
| rloo | the code seems complex to do something that would be simpler with the explicit maps, no? | 16:11 |
| lucasagomes | rloo, yup | 16:11 |
| lucasagomes | I mean it's not super complex, it just uses some more advanced python features | 16:12 |
| * jroll bbiab | 16:12 | |
| rloo | lucasagomes: like the metaclass stuff. Seems like if we can avoid it, why not avoid it. BUT I'm not a python expert. Which is more 'pythonic'? | 16:13 |
| rloo | with the explicit routes, the driver writers can easily see which functions are being used. With the decorator, they'd have to scan the file. | 16:14 |
| NobodyCam | brb | 16:14 |
| lucasagomes | rloo, right... ok I prefer the simple case as well | 16:17 |
| *** jmanko has joined #openstack-ironic | 16:17 | |
| lucasagomes | sambetts, hey, I think I won't use the decorator then :( | 16:17 |
| rloo | is Shrews around? | 16:17 |
| rloo | (trying to remember who asked for decorators) | 16:18 |
| lucasagomes | I think it was Shrews | 16:18 |
| rloo | To be honest, I think that the use of decorators for what sambetts did, is reasonable. It is just that it doesn't seem to 'extend' easily for what lucasagomes wants wrt the blueprint. | 16:19 |
| *** marcoemorais has joined #openstack-ironic | 16:19 | |
| lucasagomes | makes sense for the log and exceptions, but if we try to expand it, then it becomes messy | 16:19 |
| lucasagomes | yeha | 16:19 |
| Shrews | here | 16:19 |
| *** jmank has quit IRC | 16:20 | |
| Shrews | just got off the a long conf call. gotta catch up on scrollback | 16:20 |
| rloo | hi Shrews. ^^ lucasagomes has been experimenting with the decorators that sambetts added, and I think it would be simpler to not use decorators. wanted your feedback. | 16:20 |
| lucasagomes | (feels sorry for taking the time of ur all to look at it, but I don't want to have to write a spec again and again and again) | 16:21 |
| Shrews | rloo: lucasagomes: sambetts: i mentioned the use of decorators to make sure we considered using them as what sambetts wanted to do seemed to fit the pattern for that. if it turns out they are not right, then that's fine by me. | 16:22 |
| rloo | lucasagomes: no worries. I'd rather the discussion happen now rather than later. | 16:22 |
| * Shrews trusts lucasagomes's opinion | 16:22 | |
| lucasagomes | Shrews, they r fine for the logging and converting exceptions... problem is extending it to create the mapping for the vendor functions dynamically | 16:23 |
| rloo | Shrews: yeah, that's what I think. Would you take a quick look at lucasagomes' prototype and see what you think? http://paste.openstack.org/show/126558/ | 16:23 |
| rloo | Shrews: it's taking those decorated functions and creating a dictionary with entries for them. | 16:24 |
| *** todd_dsm has joined #openstack-ironic | 16:24 | |
| Shrews | looking | 16:24 |
| *** rameshg87 has joined #openstack-ironic | 16:28 | |
| *** BLZbubba has quit IRC | 16:29 | |
| *** BLZbubba has joined #openstack-ironic | 16:29 | |
| *** dtantsur is now known as dtantsur|afk | 16:29 | |
| openstackgerrit | Jim Mankovich proposed a change to openstack/ironic-specs: Remove Sensor RecordID from IPMI Sensor ID https://review.openstack.org/130070 | 16:30 |
| Shrews | lucasagomes: rloo: that seems acceptable to me (though metaclasses are handled differently in python3). | 16:32 |
| *** Marga__ has joined #openstack-ironic | 16:32 | |
| rloo | thx Shrews. | 16:33 |
| lucasagomes | Shrews, right so metaclass is it? yeah for py3 we use six | 16:33 |
| lucasagomes | this is just a prototype | 16:33 |
| Shrews | sure | 16:34 |
| *** Marga_ has quit IRC | 16:34 | |
| openstackgerrit | Jim Mankovich proposed a change to openstack/ironic-specs: Support per Driver Sensor Meters https://review.openstack.org/130074 | 16:35 |
| Shrews | lucasagomes: what is 'lalau' ??? anything meaningful? | 16:36 |
| lucasagomes | Shrews, no, just sounds funny in portuguese | 16:37 |
| Shrews | :) | 16:37 |
| *** Marga__ has quit IRC | 16:37 | |
| *** victor_lowther__ has quit IRC | 16:37 | |
| lucasagomes | naming things are hard :P specially for prototypes | 16:37 |
| * Shrews wants to hear "dhsauhdsuhdsaudshudsah" in portuguese | 16:37 | |
| *** rameshg87 has left #openstack-ironic | 16:38 | |
| lucasagomes | hah I was a bit mad things weren't working :P | 16:38 |
| *** spandhe has joined #openstack-ironic | 16:42 | |
| *** rloo has quit IRC | 16:43 | |
| *** marck has quit IRC | 16:43 | |
| *** marcoemorais has quit IRC | 16:43 | |
| *** spandhe_ has joined #openstack-ironic | 16:46 | |
| *** spandhe has quit IRC | 16:47 | |
| *** spandhe_ is now known as spandhe | 16:47 | |
| *** marcoemorais has joined #openstack-ironic | 16:47 | |
| *** jcoufal has quit IRC | 16:55 | |
| *** rakesh_hs has quit IRC | 17:04 | |
| *** Marga_ has joined #openstack-ironic | 17:05 | |
| *** Marga_ has quit IRC | 17:14 | |
| *** harlowja_away is now known as harlowja | 17:16 | |
| *** Poornima has joined #openstack-ironic | 17:21 | |
| *** harlowja_ has joined #openstack-ironic | 17:21 | |
| *** harlowja has quit IRC | 17:21 | |
| *** stendulker has quit IRC | 17:22 | |
| *** achanda has joined #openstack-ironic | 17:23 | |
| *** derekh has quit IRC | 17:25 | |
| *** jcoufal has joined #openstack-ironic | 17:27 | |
| *** leopoldj has quit IRC | 17:29 | |
| *** spandhe has quit IRC | 17:32 | |
| *** romcheg has quit IRC | 17:32 | |
| *** victor_lowther__ has joined #openstack-ironic | 17:33 | |
| *** achanda has quit IRC | 17:35 | |
| *** jistr has quit IRC | 17:37 | |
| *** marcoemorais1 has joined #openstack-ironic | 17:40 | |
| *** marcoemorais has quit IRC | 17:40 | |
| *** marcoemorais1 has quit IRC | 17:42 | |
| *** marcoemorais has joined #openstack-ironic | 17:43 | |
| *** ryanpetrello has quit IRC | 17:43 | |
| *** achanda has joined #openstack-ironic | 17:48 | |
| *** achanda has quit IRC | 17:51 | |
| *** harlowja_ has quit IRC | 17:55 | |
| *** harlowja has joined #openstack-ironic | 17:56 | |
| *** rloo has joined #openstack-ironic | 17:59 | |
| *** marcoemorais has quit IRC | 18:01 | |
| *** marcoemorais has joined #openstack-ironic | 18:01 | |
| *** marcoemorais has quit IRC | 18:02 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add a mechanism to route vendor methods https://review.openstack.org/129261 | 18:02 |
| *** todd_dsm has quit IRC | 18:02 | |
| *** marcoemorais has joined #openstack-ironic | 18:02 | |
| lucasagomes | rloo, Shrews ^ no metaclass | 18:04 |
| lucasagomes | clearner imo | 18:04 |
| rloo | looking... | 18:04 |
| * sambetts looking | 18:05 | |
| lucasagomes | I just put on the __init__ of the VendorPassthru what I was doing in the MetaClass | 18:05 |
| *** pcrews has quit IRC | 18:07 | |
| sambetts | I think thats nicer, no metaclass, however doesn't that rely on method being the function name??? | 18:08 |
| lucasagomes | sambetts, nop why? | 18:08 |
| lucasagomes | I don't get the func reference in the decorator anymore, since the decorator will wrap it and return another method | 18:09 |
| lucasagomes | what I do is on the Vendor class instantiation, I look for methods that contains the vendor metadata | 18:09 |
| NobodyCam | any ipa folks about? | 18:09 |
| lucasagomes | sambetts, see base.py | 18:09 |
| jroll | hi NobodyCam | 18:09 |
| NobodyCam | hey hey jroll | 18:09 |
| lucasagomes | sambetts, L428 | 18:09 |
| NobodyCam | would you happen to have a sec to look at: http://paste.openstack.org/show/FymqYB0fz1jkIqKRr75h/ | 18:10 |
| jroll | ofc | 18:10 |
| NobodyCam | I'm sure I have not set something | 18:10 |
| sambetts | yeah, I was thinking you were using the name field for getattr but thats not it | 18:10 |
| jroll | huh | 18:10 |
| jroll | NobodyCam: that's trying to download your deploy image | 18:10 |
| jroll | NobodyCam: probably wrong deploy k/r ID in driver info? | 18:10 |
| rloo | lucasagomes: I skimmed it. That looks much better! | 18:12 |
| lucasagomes | rloo, :) | 18:12 |
| lucasagomes | yeah, no black metaclass magic | 18:12 |
| NobodyCam | jroll: hum ... yep | 18:12 |
| sambetts | lucasagomes: that looks nice, why did you pick to make metadata a set instead of a dict so you could access it by key?? | 18:13 |
| rloo | lucasagomes: assuming it works, but I does, right? I'll confirm when I really review it ;) | 18:13 |
| lucasagomes | rloo, yeah I'm testing it now and gate will test it too | 18:13 |
| lucasagomes | rloo, I made a basic vendor method and a driver method | 18:13 |
| lucasagomes | it seems fine | 18:14 |
| jroll | lucasagomes: so uh, on thing | 18:14 |
| jroll | one | 18:14 |
| jroll | you're looping over all of the methods anyway | 18:14 |
| lucasagomes | sambetts, a set instead of dict? A tuple? | 18:14 |
| jroll | oh wait | 18:14 |
| jroll | ignore me | 18:14 |
| jroll | lucasagomes: with vendor_meta[1], what other keys are there? | 18:14 |
| lucasagomes | sambetts, I will extend it later to be (method, {'supported': ['GET', 'POST'], 'async': True}) | 18:14 |
| rloo | need to get the spec approved. maybe we can do some sort of spec-review thing at the summit/pod. | 18:14 |
| lucasagomes | jroll, ^ | 18:14 |
| jroll | aha :D | 18:15 |
| lucasagomes | it's like a plumbing to add the rest of the metadata for that method | 18:15 |
| *** romcheg has joined #openstack-ironic | 18:15 | |
| jroll | yeah | 18:15 |
| lucasagomes | jroll, makes sense? | 18:15 |
| jroll | yep | 18:15 |
| jroll | :) | 18:15 |
| lucasagomes | rloo, yup, yeah I will work on the spec now that I've a clear view of how it will wokr | 18:16 |
| lucasagomes | work* | 18:16 |
| sambetts | lucasagomes: I thought that which is why I wondered why you have a tuple wrapping it, you could just have 'method': whatever in the dict? | 18:16 |
| lucasagomes | sambetts, yup having a dict there works as well | 18:17 |
| lucasagomes | I did a tuple cause it's easier to access the elements by index | 18:17 |
| sambetts | I thought that it would be more readable to access it by key thats all, either works :-) | 18:18 |
| lucasagomes | sambetts, it will be only 1 key, because it's a tuple or dict per method | 18:19 |
| *** dprince has quit IRC | 18:20 | |
| sambetts | OOOOh I see why you've done it now! Ignore me :D | 18:22 |
| *** marcoemorais1 has joined #openstack-ironic | 18:23 | |
| sambetts | It makes sense once I realised the mapping was like this: vendor_meta[0]: vendor_meta[1] | 18:23 |
| lucasagomes | yup | 18:23 |
| *** marcoemorais has quit IRC | 18:24 | |
| *** Marga_ has joined #openstack-ironic | 18:24 | |
| sambetts | :D lgtm then | 18:24 |
| lucasagomes | \o/ | 18:25 |
| NobodyCam | lol jroll so close: http://i.imgur.com/26dpd4H.png | 18:25 |
| jroll | :( | 18:25 |
| jroll | NobodyCam: how much ram do you have? | 18:25 |
| *** Poornima has quit IRC | 18:28 | |
| *** todd_dsm has joined #openstack-ironic | 18:28 | |
| *** Marga_ has quit IRC | 18:31 | |
| *** marcoemorais1 has quit IRC | 18:32 | |
| *** marcoemorais has joined #openstack-ironic | 18:32 | |
| *** marcoemorais has quit IRC | 18:32 | |
| lucasagomes | aight I will take a break, good night everyone | 18:33 |
| *** marcoemorais has joined #openstack-ironic | 18:33 | |
| lucasagomes | ta much for all the comments and suggestions! | 18:33 |
| NobodyCam | night lucasagomes | 18:33 |
| NobodyCam | jroll: not much <currentMemory unit='KiB'>2097152</currentMemory> | 18:33 |
| sambetts | good night lucasagomes | 18:33 |
| jroll | NobodyCam: try bumping that up | 18:34 |
| jroll | wait | 18:34 |
| *** marcoemorais has quit IRC | 18:34 | |
| *** lucasagomes is now known as lucas-dinner | 18:34 | |
| jroll | that's 2 gigs, should be fine | 18:34 |
| *** marcoemorais has joined #openstack-ironic | 18:34 | |
| *** dprince has joined #openstack-ironic | 18:37 | |
| *** athomas has quit IRC | 18:40 | |
| NobodyCam | brb | 18:41 |
| *** marcoemorais has quit IRC | 18:45 | |
| *** marcoemorais has joined #openstack-ironic | 18:45 | |
| *** marcoemorais has quit IRC | 18:49 | |
| *** marcoemorais has joined #openstack-ironic | 18:49 | |
| *** achanda has joined #openstack-ironic | 18:52 | |
| *** marcoemorais has quit IRC | 18:54 | |
| *** marcoemorais has joined #openstack-ironic | 18:55 | |
| *** achanda has quit IRC | 18:57 | |
| sambetts | I'm off for the night, see you tomorrow ironic o/ | 18:58 |
| *** marcoemorais has quit IRC | 18:59 | |
| NobodyCam | night sambetts | 19:00 |
| sambetts | night NobodyCam | 19:00 |
| *** sambetts has quit IRC | 19:00 | |
| *** pcrews has joined #openstack-ironic | 19:07 | |
| *** andreykurilin_ has joined #openstack-ironic | 19:08 | |
| *** rushiagr is now known as rushiagr_away | 19:10 | |
| NobodyCam | jroll: is there a knowen good ipa deploy image I can pull down for testing? | 19:12 |
| jroll | NobodyCam: https://github.com/openstack-dev/devstack/blob/master/lib/ironic#L78-79 | 19:13 |
| jroll | Shrews: around? | 19:13 |
| jroll | or anyone that knows about devstack(-gate) and firewalls that happen or don't happen to the "baremetal" vms | 19:13 |
| NobodyCam | TY | 19:14 |
| jroll | np | 19:15 |
| *** romcheg1 has joined #openstack-ironic | 19:15 | |
| *** romcheg has quit IRC | 19:16 | |
| *** marcoemorais has joined #openstack-ironic | 19:22 | |
| jroll | https://github.com/openstack-dev/devstack/blob/master/lib/ironic#L526 | 19:22 |
| jroll | muahahaha | 19:22 |
| jroll | :( | 19:22 |
| jroll | this is why agent jobs are failing | 19:23 |
| * jroll will fix shortly | 19:23 | |
| Shrews | jroll: i'm not familiar with the devstack iptable magic | 19:28 |
| *** jcoufal has quit IRC | 19:30 | |
| jroll | Shrews: no worries, I found it | 19:42 |
| *** pcrews has quit IRC | 19:53 | |
| *** marcoemorais has quit IRC | 19:54 | |
| *** Marga_ has joined #openstack-ironic | 19:56 | |
| *** pelix has quit IRC | 19:57 | |
| *** pensu has quit IRC | 19:58 | |
| *** achanda has joined #openstack-ironic | 19:59 | |
| *** Marga_ has quit IRC | 20:05 | |
| *** Marga_ has joined #openstack-ironic | 20:06 | |
| *** andreykurilin_ has quit IRC | 20:08 | |
| *** igordcard has joined #openstack-ironic | 20:10 | |
| *** marcoemorais has joined #openstack-ironic | 20:18 | |
| *** dprince has quit IRC | 20:19 | |
| *** pcrews has joined #openstack-ironic | 20:23 | |
| *** adam_g_gone is now known as adam_g | 20:30 | |
| adam_g | o/ | 20:30 |
| NobodyCam | morning adam_g | 20:30 |
| *** Shrews has quit IRC | 20:44 | |
| *** Shrews has joined #openstack-ironic | 20:46 | |
| *** Shrews has quit IRC | 20:46 | |
| *** achanda has quit IRC | 20:50 | |
| *** marcoemorais1 has joined #openstack-ironic | 21:02 | |
| *** marcoemorais has quit IRC | 21:02 | |
| *** marcoemorais1 has quit IRC | 21:03 | |
| *** marcoemorais has joined #openstack-ironic | 21:04 | |
| *** todd_dsm has quit IRC | 21:07 | |
| *** Shrews has joined #openstack-ironic | 21:08 | |
| jroll | adam_g: \o | 21:09 |
| jroll | NobodyCam: get anywhere? | 21:09 |
| *** Marga_ has quit IRC | 21:09 | |
| NobodyCam | let ya know in a few | 21:12 |
| *** todd_dsm has joined #openstack-ironic | 21:14 | |
| *** Marga_ has joined #openstack-ironic | 21:16 | |
| jroll | JayF, adam_g, https://review.openstack.org/#/c/131876/ | 21:16 |
| jroll | NobodyCam: ^ that was my problem in devstack | 21:16 |
| adam_g | jroll, ah!! | 21:17 |
| jroll | :) | 21:17 |
| adam_g | jroll, one small nit | 21:18 |
| jroll | sure | 21:18 |
| jroll | touché | 21:18 |
| jroll | fixed, ty | 21:19 |
| adam_g | i miss anything exciting the last week or has it been pre-summit crickets? | 21:20 |
| *** anderbubble has joined #openstack-ironic | 21:20 | |
| jroll | yeah, not a lot going on | 21:21 |
| *** achanda has joined #openstack-ironic | 21:24 | |
| *** todd_dsm has quit IRC | 21:26 | |
| *** todd_dsm has joined #openstack-ironic | 21:29 | |
| NobodyCam | jroll: agent boots attempts to download image and errors | 21:35 |
| NobodyCam | fyi: the sample agent you pointed my to | 21:35 |
| NobodyCam | I see Could not download image with id 35804f18-fe1f-4fa2-8223-3e5421375626. in the log | 21:36 |
| NobodyCam | that id is in glance | 35804f18-fe1f-4fa2-8223-3e5421375626 | testimage | qcow2 | bare | 280100864 | active | | 21:36 |
| *** mjturek has quit IRC | 21:38 | |
| *** eghobo has joined #openstack-ironic | 21:39 | |
| *** eghobo has quit IRC | 21:39 | |
| anderbubble | I admit that this is really just a general openstack question; but I’m going through the instructions at http://docs.openstack.org/developer/ironic/deploy/install-guide.html to set up ironic, and I’m wondering why I have to configure explicit addresses and urls for individual services to configure ironic. Why can’t I just configure rabbitmq and have it find the services via the mq? | 21:43 |
| jroll | NobodyCam: conductor logs may have better errors? | 21:45 |
| jroll | anderbubble: not everything talks over rabbitmq... any config in particular you're thinking of? | 21:45 |
| jroll | anderbubble: e.g. nova uses a rest api to talk to ironic | 21:45 |
| anderbubble | in general, I think it would be nice if all of the services advertised on the mq; so then I could, for example, configure ironic to know where rabbitmq was, and then just have it broadcast a request for “where is nova?” “where is neutron?” “where is glance?” etc | 21:47 |
| anderbubble | without having to hardcode the different ip addresses in the config file | 21:47 |
| anderbubble | full disclosure: this is my first openstack experience; it’s just a thought | 21:47 |
| NobodyCam | jroll: yep... looks like swift proxy configs : http://paste.openstack.org/show/pCsynsXtZY73tlHuhfFr/ | 21:47 |
| NobodyCam | brb | 21:48 |
| jroll | NobodyCam: any logs from before this | 21:48 |
| jroll | anderbubble: aha, so | 21:49 |
| jroll | keystone has a concept of a service catalog | 21:49 |
| jroll | services register with keystone and other services can query this | 21:49 |
| jroll | it tends to work.. you shouldn't need to configure the neutron or glance url | 21:49 |
| anderbubble | ah | 21:49 |
| jroll | nor the ironic endpoint | 21:50 |
| anderbubble | so it’s just in the instructions because… bit rot? ;) | 21:50 |
| jroll | because it's optional | 21:50 |
| NobodyCam | http://paste.openstack.org/show/8arBmdhe9UBhvVrCC8xk | 21:50 |
| jroll | I prefer to be explicit, so I configure them | 21:50 |
| *** openstackgerrit has quit IRC | 21:50 | |
| jroll | I also work at rackspace, where our service catalog is hilariously broken | 21:50 |
| anderbubble | ok; cool | 21:50 |
| jroll | so I configure them | 21:50 |
| jroll | :) | 21:50 |
| anderbubble | except that, then, if they change, you have to change it in *two* places :p | 21:50 |
| anderbubble | ;) | 21:50 |
| anderbubble | anyway: good to know. I have been enlightened. | 21:51 |
| jroll | right :P | 21:51 |
| NobodyCam | brb | 21:51 |
| jroll | I would love to use etcd or something for service discovery | 21:51 |
| jroll | NobodyCam: huh, I wonder if the tempurl expired... kind of doubt it though | 21:52 |
| anderbubble | jroll, to be fair, the instructions say things like “This configuration file must be modified on the Compute Service’s controller nodes and compute nodes." | 21:52 |
| anderbubble | with emphasis | 21:52 |
| jroll | right | 21:53 |
| jroll | I still don't know what controller/compute nodes are | 21:53 |
| jroll | at any rate | 21:53 |
| jroll | the service catalog should work for all of those urls afaik | 21:53 |
| jroll | (if it's configured correctly) | 21:53 |
| jroll | which are the "keystone service-create" and "keystone endpoint-create" calls | 21:53 |
| anderbubble | I see that now; thanks | 21:54 |
| anderbubble | is the same true for the database connection? | 21:54 |
| jroll | no, db needs to be explicitly configured | 21:56 |
| *** eghobo has joined #openstack-ironic | 21:57 | |
| *** achanda_ has joined #openstack-ironic | 22:04 | |
| *** marcoemorais has quit IRC | 22:04 | |
| *** marcoemorais has joined #openstack-ironic | 22:05 | |
| *** achanda has quit IRC | 22:08 | |
| *** trown is now known as trown|outttypeww | 22:10 | |
| *** igordcard has quit IRC | 22:15 | |
| *** marcoemorais has quit IRC | 22:15 | |
| *** todd_dsm has quit IRC | 22:17 | |
| *** andreykurilin_ has joined #openstack-ironic | 22:22 | |
| *** anderbubble has quit IRC | 22:22 | |
| *** Marga_ has quit IRC | 22:22 | |
| *** ociuhandu has quit IRC | 22:25 | |
| *** todd_dsm has joined #openstack-ironic | 22:27 | |
| *** ociuhandu has joined #openstack-ironic | 22:28 | |
| *** marcoemorais has joined #openstack-ironic | 22:29 | |
| *** ChuckC has quit IRC | 22:33 | |
| *** achanda_ has quit IRC | 22:34 | |
| dlaube | hey guys, for those of you running openstack services configured to use rabbitmq, are you using a single vhost or separate vhosts per service? | 22:41 |
| JayF | NobodyCam: did you see jroll's devstack patch? | 22:41 |
| JayF | NobodyCam: I wonder if you're hitting firewall issues like the job is | 22:41 |
| jroll | mayyyyyybe | 22:42 |
| NobodyCam | hummm | 22:42 |
| jroll | dlaube: we use one vhost | 22:42 |
| *** ChuckC has joined #openstack-ironic | 22:45 | |
| dlaube | same here, just making sure | 22:45 |
| dlaube | thanks jroll | 22:45 |
| jroll | np :) | 22:45 |
| *** romcheg1 has quit IRC | 22:46 | |
| *** todd_dsm has quit IRC | 22:46 | |
| *** spandhe has joined #openstack-ironic | 22:48 | |
| *** hemna is now known as hemna__ | 22:57 | |
| *** achanda has joined #openstack-ironic | 22:58 | |
| *** ociuhandu_ has joined #openstack-ironic | 22:59 | |
| *** ociuhandu has quit IRC | 23:00 | |
| *** ociuhandu_ is now known as ociuhandu | 23:00 | |
| *** eghobo has quit IRC | 23:02 | |
| *** yuanying has joined #openstack-ironic | 23:07 | |
| BLZbubba | where can I find a juno-compatible python-ironic? | 23:16 |
| BLZbubba | testing 1 2 3 | 23:16 |
| *** BLZbubba has quit IRC | 23:16 | |
| *** BLZbubba has joined #openstack-ironic | 23:16 | |
| rloo | BLZbubba: for ubuntu? | 23:16 |
| BLZbubba | hello there, what does it take to get a juno compatible version of python-ironic? | 23:17 |
| BLZbubba | yes, 14.04 with the cloud-archive juno build | 23:17 |
| BLZbubba | it has a newer version of sql-alchemy that trusty's python-ironic hates | 23:17 |
| BLZbubba | sucks not to be a real openstack project anymore | 23:18 |
| *** andreykurilin_ has quit IRC | 23:18 | |
| rloo | BLZbubba: oh, yeah, I thought I saw someone else mention something about sql-alchemy. I'm not sure what they did. (Or was that you that mentioned it) | 23:19 |
| BLZbubba | it was me, i haven't been able to look at it for a couple of days | 23:20 |
| BLZbubba | i'm going to have to build my diskimage code on the juno baremetal version, i guess | 23:21 |
| BLZbubba | why did it get demoted? | 23:21 |
| rloo | BLZbubba: I don't know. | 23:22 |
| rloo | BLZbubba: don't know if this helps or not: https://github.com/openstack/requirements/blob/master/global-requirements.txt#L132 | 23:27 |
| rloo | g'night all. | 23:30 |
| *** rloo has quit IRC | 23:30 | |
| jroll | 23:18:16 BLZbubba | sucks not to be a real openstack project anymore <- eh? | 23:31 |
| *** shikui__ has joined #openstack-ironic | 23:34 | |
| *** lucas-dinner has quit IRC | 23:37 | |
| *** annegentle has quit IRC | 23:42 | |
| *** annegentle has joined #openstack-ironic | 23:42 | |
| *** yjiang5 is now known as yjiang5_away | 23:43 | |
| *** Shrews has quit IRC | 23:46 | |
| *** annegentle has quit IRC | 23:46 | |
| *** annegentle has joined #openstack-ironic | 23:47 | |
| *** Shrews has joined #openstack-ironic | 23:47 | |
| *** annegentle has quit IRC | 23:48 | |
| *** annegentle has joined #openstack-ironic | 23:48 | |
| *** annegentle has quit IRC | 23:49 | |
| *** annegentle has joined #openstack-ironic | 23:49 | |
| *** Marga_ has joined #openstack-ironic | 23:50 | |
| *** Haomeng has joined #openstack-ironic | 23:53 | |
| *** Haomeng|2 has quit IRC | 23:54 | |
| spandhe | hey jroll ! I managed to make some progress.. have setup Neutron for now because it seems nova needs it.. so its going well.. | 23:56 |
| *** Marga_ has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
