*** alexpilotti has quit IRC | 00:08 | |
openstackgerrit | Merged openstack/ironic: Update devstack provision net config for multihost https://review.openstack.org/368644 | 00:09 |
---|---|---|
openstackgerrit | Merged openstack/ironic: Update ironic node names for multinode case https://review.openstack.org/368645 | 00:09 |
*** chas_ has joined #openstack-ironic | 00:15 | |
*** chas_ has quit IRC | 00:20 | |
*** pmannidi has joined #openstack-ironic | 00:25 | |
*** hoangcx has joined #openstack-ironic | 00:59 | |
*** phuongnh has joined #openstack-ironic | 01:00 | |
*** alexpilotti has joined #openstack-ironic | 01:05 | |
*** hw_wutianwei has joined #openstack-ironic | 01:15 | |
*** alexpilotti has quit IRC | 01:16 | |
*** chas_ has joined #openstack-ironic | 01:16 | |
*** chas_ has quit IRC | 01:20 | |
*** zhangjl has joined #openstack-ironic | 01:27 | |
*** godp1301 has joined #openstack-ironic | 01:28 | |
*** godp1301 has quit IRC | 01:33 | |
*** gcb has joined #openstack-ironic | 01:36 | |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: Generic power interface for soft reboot and soft power off https://review.openstack.org/216730 | 01:50 |
*** sabeen has joined #openstack-ironic | 01:53 | |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: Ipmitool power driver for soft reboot and soft power off https://review.openstack.org/216738 | 02:09 |
*** alexpilotti has joined #openstack-ironic | 02:12 | |
*** chas_ has joined #openstack-ironic | 02:17 | |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: iRMC power driver for soft reboot and soft power off https://review.openstack.org/216743 | 02:18 |
openstackgerrit | Tuan Luong-Anh proposed openstack/ironic: Timing metrics: iRMC drivers https://review.openstack.org/371284 | 02:18 |
*** chas_ has quit IRC | 02:21 | |
*** alexpilotti has quit IRC | 02:23 | |
openstackgerrit | Tuan Luong-Anh proposed openstack/ironic: Timing metrics: iRMC drivers https://review.openstack.org/371284 | 02:23 |
*** g3ek has quit IRC | 02:29 | |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: Update the existing APIs due to adding get_supported_power_states https://review.openstack.org/382194 | 02:29 |
*** g3ek has joined #openstack-ironic | 02:36 | |
*** hoonetorg has quit IRC | 02:40 | |
*** dims has quit IRC | 02:43 | |
*** awiddersheim has quit IRC | 02:45 | |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: Generic management I/F for Inject NMI https://review.openstack.org/348191 | 02:50 |
*** hoonetorg has joined #openstack-ironic | 02:52 | |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: Ipmitool management driver for Inject NMI https://review.openstack.org/348721 | 02:58 |
*** godp1301 has joined #openstack-ironic | 02:59 | |
*** awiddersheim has joined #openstack-ironic | 03:02 | |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: iRMC management driver for Inject NMI https://review.openstack.org/348732 | 03:06 |
*** godp1301 has quit IRC | 03:16 | |
*** wajdi has joined #openstack-ironic | 03:18 | |
*** rbrud has joined #openstack-ironic | 03:19 | |
*** alexpilotti has joined #openstack-ironic | 03:20 | |
*** wajdi_ has joined #openstack-ironic | 03:23 | |
*** wajdi has quit IRC | 03:26 | |
*** shuquan has joined #openstack-ironic | 03:30 | |
*** rbrud has quit IRC | 03:31 | |
*** alexpilotti has quit IRC | 03:31 | |
*** shuquan has quit IRC | 03:49 | |
*** shuquan has joined #openstack-ironic | 04:04 | |
*** shuquan has quit IRC | 04:08 | |
*** links has joined #openstack-ironic | 04:15 | |
*** yolanda has quit IRC | 04:18 | |
*** chas_ has joined #openstack-ironic | 04:18 | |
*** chas_ has quit IRC | 04:23 | |
*** moshele has joined #openstack-ironic | 04:25 | |
*** alexpilotti has joined #openstack-ironic | 04:28 | |
*** moshele has quit IRC | 04:31 | |
*** shuquan has joined #openstack-ironic | 04:31 | |
openstackgerrit | vinay kumar muddu proposed openstack/ironic-inspector: Checks if ovs bridge brbm exists https://review.openstack.org/396583 | 04:32 |
*** alexpilotti has quit IRC | 04:34 | |
*** livelace has joined #openstack-ironic | 04:37 | |
*** shuquan has quit IRC | 04:39 | |
*** shuquan has joined #openstack-ironic | 04:41 | |
*** livelace has quit IRC | 04:49 | |
*** shuquan has quit IRC | 04:50 | |
*** amotoki has joined #openstack-ironic | 05:31 | |
*** alexpilotti has joined #openstack-ironic | 05:31 | |
*** alexpilotti has quit IRC | 05:39 | |
*** shuquan has joined #openstack-ironic | 05:49 | |
*** rbudden has joined #openstack-ironic | 06:05 | |
*** jaosorior has joined #openstack-ironic | 06:09 | |
*** nicodemos has quit IRC | 06:26 | |
*** wajdi_ has quit IRC | 06:28 | |
*** shuquan_ has joined #openstack-ironic | 06:38 | |
*** shuquan has quit IRC | 06:39 | |
*** jaosorior has quit IRC | 06:39 | |
*** jaosorior has joined #openstack-ironic | 06:39 | |
*** moshele has joined #openstack-ironic | 06:40 | |
*** sabeen has quit IRC | 06:41 | |
*** aparnav has joined #openstack-ironic | 06:41 | |
*** AlexeyAbashkin has joined #openstack-ironic | 06:51 | |
*** mjura has joined #openstack-ironic | 06:54 | |
*** AlexeyAbashkin has quit IRC | 06:57 | |
*** fxpester has joined #openstack-ironic | 07:02 | |
*** shuquan_ has quit IRC | 07:15 | |
*** g3ek has quit IRC | 07:15 | |
*** shuquan has joined #openstack-ironic | 07:16 | |
*** g3ek has joined #openstack-ironic | 07:21 | |
*** jaosorior has quit IRC | 07:27 | |
*** e0ne has joined #openstack-ironic | 07:30 | |
*** fxpester|2 has joined #openstack-ironic | 07:33 | |
*** yolanda has joined #openstack-ironic | 07:33 | |
*** fxpester has quit IRC | 07:35 | |
*** alexpilotti has joined #openstack-ironic | 07:36 | |
*** parlos has joined #openstack-ironic | 07:37 | |
*** alexpilotti has quit IRC | 07:42 | |
*** roeyc has joined #openstack-ironic | 07:50 | |
*** jaosorior has joined #openstack-ironic | 07:55 | |
*** AlexeyAbashkin has joined #openstack-ironic | 07:56 | |
*** pcaruana has joined #openstack-ironic | 07:57 | |
*** parlos has quit IRC | 07:59 | |
*** ccamacho has joined #openstack-ironic | 08:01 | |
*** ralonsoh has joined #openstack-ironic | 08:03 | |
*** skazi has quit IRC | 08:04 | |
*** skazi has joined #openstack-ironic | 08:04 | |
openstackgerrit | Sukhdev Kapur proposed openstack/ironic: Add support for Security Groups for baremetal servers https://review.openstack.org/361451 | 08:04 |
*** roeyc has quit IRC | 08:05 | |
*** ohamada has joined #openstack-ironic | 08:09 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: WIP: Run all tests in one job. https://review.openstack.org/396155 | 08:10 |
*** ricardoas has quit IRC | 08:13 | |
*** xavierr has quit IRC | 08:13 | |
*** cdaniel has quit IRC | 08:13 | |
*** iurygregory has quit IRC | 08:13 | |
*** e0ne has quit IRC | 08:15 | |
*** milan has joined #openstack-ironic | 08:35 | |
milan | morning Ironic! :) | 08:36 |
*** alexpilotti has joined #openstack-ironic | 08:39 | |
*** d0ugal has joined #openstack-ironic | 08:42 | |
pas-ha | morning Ironic :) morning milan | 08:44 |
*** alexpilotti has quit IRC | 08:44 | |
milan | morning pas-ha! :) | 08:44 |
*** _vsaienko has joined #openstack-ironic | 08:44 | |
*** gcb has quit IRC | 08:46 | |
pas-ha | I've got a question re 'deploy_forces_oob_reboot' property. It is stated that it "Provides compatibility with older deploy ramdisks". As I am removing lookups/heartbeats as vendor passthru, the question is can we drop this property already? With old lookups/heartbeats removed, we already require deploy ramdisk with quite a new agent... | 08:47 |
*** jpich has joined #openstack-ironic | 08:48 | |
*** roeyc has joined #openstack-ironic | 08:49 | |
*** mgoddard has joined #openstack-ironic | 08:49 | |
*** zackf has joined #openstack-ironic | 08:54 | |
*** parlos has joined #openstack-ironic | 08:55 | |
*** chas_ has joined #openstack-ironic | 08:57 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: WIP: Run all tests in one job. https://review.openstack.org/396155 | 08:59 |
*** zackf has quit IRC | 08:59 | |
*** gcb has joined #openstack-ironic | 08:59 | |
*** zzzeek has quit IRC | 09:00 | |
*** zzzeek has joined #openstack-ironic | 09:01 | |
*** betherly has joined #openstack-ironic | 09:02 | |
*** e0ne has joined #openstack-ironic | 09:03 | |
*** AlexeyAbashkin has quit IRC | 09:04 | |
*** jtomasek has joined #openstack-ironic | 09:12 | |
*** tuanluong has joined #openstack-ironic | 09:12 | |
*** e0ne has quit IRC | 09:13 | |
*** leey has joined #openstack-ironic | 09:14 | |
lucasagomes | morning all, milan, pas-ha | 09:19 |
*** jasonzh has joined #openstack-ironic | 09:19 | |
*** ChubYann has quit IRC | 09:19 | |
*** bcornec1 has quit IRC | 09:21 | |
*** e0ne has joined #openstack-ironic | 09:21 | |
pas-ha | morning lucasagomes | 09:21 |
leey | morning | 09:23 |
*** strigazi_AFK is now known as strigazi | 09:26 | |
*** matianliang has joined #openstack-ironic | 09:26 | |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: Depends-On: I3b0d2434e608c6641c74034dd42fc4a7077c5dee https://review.openstack.org/397078 | 09:28 |
*** matianliang has quit IRC | 09:30 | |
*** athomas has joined #openstack-ironic | 09:30 | |
*** jasonzh has quit IRC | 09:31 | |
*** jasonzh has joined #openstack-ironic | 09:32 | |
*** jasonzh has quit IRC | 09:32 | |
vsaienk0 | morning pas-ha, leey, dtantsur, sambetts, lucasagomes, milan, vdrok, aarefiev and all Ironic'ers! | 09:32 |
lucasagomes | hi there | 09:33 |
*** leey has quit IRC | 09:33 | |
*** jasonzh has joined #openstack-ironic | 09:33 | |
*** matianliang has joined #openstack-ironic | 09:35 | |
*** jasonzh has quit IRC | 09:35 | |
*** leey has joined #openstack-ironic | 09:35 | |
*** leey has quit IRC | 09:36 | |
*** matianliang has quit IRC | 09:39 | |
*** ccamacho is now known as ccamacho|afk | 09:39 | |
*** alexpilotti has joined #openstack-ironic | 09:40 | |
*** e0ne_ has joined #openstack-ironic | 09:40 | |
milan | morning lucasagomes, vsaienk0! :) | 09:40 |
openstackgerrit | Vasyl Saienko proposed openstack/python-ironicclient: OSC add capability to remove node/chassis_uuid https://review.openstack.org/391812 | 09:41 |
*** leey has joined #openstack-ironic | 09:43 | |
*** e0ne has quit IRC | 09:43 | |
*** leey has quit IRC | 09:44 | |
*** alexpilotti has quit IRC | 09:46 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: WIP: Run all tests in one job. https://review.openstack.org/396155 | 09:47 |
*** openstackgerrit has quit IRC | 09:47 | |
*** e0ne_ has quit IRC | 09:47 | |
*** openstackgerrit has joined #openstack-ironic | 09:48 | |
*** e0ne has joined #openstack-ironic | 09:50 | |
*** leey has joined #openstack-ironic | 09:52 | |
*** leey has quit IRC | 09:54 | |
*** leey has joined #openstack-ironic | 09:57 | |
*** leey has quit IRC | 09:58 | |
*** derekh has joined #openstack-ironic | 09:59 | |
*** sambetts|afk is now known as sambetts | 10:03 | |
sambetts | Morning all | 10:03 |
openstackgerrit | Milan Kováčik proposed openstack/python-ironic-inspector-client: UUID, started_at, finished_at in the status https://review.openstack.org/395609 | 10:05 |
milan | morning sambetts! :) | 10:05 |
*** roeyc has quit IRC | 10:06 | |
*** aarefiev has joined #openstack-ironic | 10:06 | |
milan | sambetts, mind checking? ^^ ;) | 10:07 |
sambetts | milan: I'll take a look in a bit :) | 10:08 |
milan | already had a +2 but wanted to correct some nits... | 10:08 |
milan | sambetts, thanks! | 10:08 |
*** parlos has quit IRC | 10:08 | |
*** e0ne has quit IRC | 10:11 | |
*** roeyc has joined #openstack-ironic | 10:12 | |
*** zhangjl has quit IRC | 10:15 | |
*** phuongnh has quit IRC | 10:19 | |
*** aarefiev has quit IRC | 10:22 | |
openstackgerrit | Tuan Luong-Anh proposed openstack/ironic: Use __slots__ for agent driver https://review.openstack.org/397103 | 10:22 |
*** roeyc has quit IRC | 10:23 | |
tuanluong | Morning ironic | 10:24 |
*** tuanluong has quit IRC | 10:26 | |
*** alexpilotti has joined #openstack-ironic | 10:27 | |
*** dtantsur|afk is now known as dtantsur | 10:30 | |
dtantsur | Morning Ironic | 10:30 |
*** priteau has joined #openstack-ironic | 10:31 | |
*** sborkows has joined #openstack-ironic | 10:35 | |
*** aarefiev has joined #openstack-ironic | 10:37 | |
*** hoangcx has quit IRC | 10:40 | |
*** shuquan has quit IRC | 10:40 | |
milan | morning dtantsur! :) | 10:40 |
*** baoli has joined #openstack-ironic | 10:46 | |
openstackgerrit | Szymon Borkowski proposed openstack/ironic-python-agent: Add a new Hardware Manager for X710 network card https://review.openstack.org/391836 | 10:47 |
*** ccamacho|afk is now known as ccamacho | 10:49 | |
*** sborkows has quit IRC | 10:50 | |
*** baoli has quit IRC | 10:50 | |
vdrok | good morning all, pas-ha lucasagomes dtantsur sambetts milan | 10:51 |
dtantsur | o/ | 10:51 |
milan | morning vdrok! :) | 10:51 |
aarefiev | morning! | 10:51 |
milan | morning aarefiev! :) | 10:51 |
*** leey has joined #openstack-ironic | 10:51 | |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: Skip some steps for multinode case https://review.openstack.org/368646 | 10:52 |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: TEST GRENADE DEV https://review.openstack.org/397126 | 10:52 |
vdrok | morning aarefiev | 10:53 |
*** leey has quit IRC | 10:53 | |
*** mgould has joined #openstack-ironic | 10:53 | |
aarefiev | morning milan, vdrok, dtantsur | 10:53 |
dtantsur | o/ | 10:53 |
vdrok | sambetts: question on https://etherpad.openstack.org/p/network-interface-vifs-configdrive L24. after looking at your code of attach/detach on ironic side, it seems to be not needed, as the address is modified by configure_tenant_netwoek | 10:54 |
vdrok | *configure_tenant_networks | 10:54 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: WIP: Run all tests in one job. https://review.openstack.org/396155 | 10:55 |
*** d0ugal has quit IRC | 10:56 | |
sambetts | vdrok: my code needs updating, on Friday we decided we need to move the neutron port update that updates the mac address into the attach_vif function, because otherwise the network_data.json that is generated by the virt driver will be out of date | 10:58 |
dtantsur | jroll, morning! two more backports please: https://review.openstack.org/#/c/396636/ and https://review.openstack.org/#/c/396640/ | 10:58 |
patchbot | patch 396636 - ironic-lib (stable/newton) - Correctly count the number of primary partitions | 10:58 |
patchbot | patch 396640 - ironic-lib (stable/newton) - followup to disk_utils.count_mbr_partitions() | 10:58 |
dtantsur | jroll, also I think I'll propose stable releases this week. we've backported quite a few things. | 10:58 |
vdrok | sambetts: ok, gotcha | 11:01 |
mgould | morning Ironic! | 11:01 |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Add database fields for all interfaces https://review.openstack.org/396681 | 11:01 |
vdrok | morning mgould | 11:01 |
dtantsur | mgould, o/ happy Monday! | 11:01 |
*** d0ugal has joined #openstack-ironic | 11:04 | |
lucasagomes | this 1LOC in project config is changing the postgres job to use standard PXE (currently no tested in our gate): https://review.openstack.org/#/c/397129/ | 11:06 |
patchbot | patch 397129 - openstack-infra/project-config - Ironic: Switch postgres job to standard PXE | 11:06 |
*** priteau has quit IRC | 11:07 | |
*** athomas has quit IRC | 11:08 | |
*** e0ne has joined #openstack-ironic | 11:09 | |
dtantsur | lucasagomes, +1 and good morning | 11:09 |
*** priteau has joined #openstack-ironic | 11:09 | |
lucasagomes | dtantsur, heh morning | 11:09 |
lucasagomes | ty | 11:09 |
openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector-specs: Endpoint to list introspection statuses https://review.openstack.org/375045 | 11:10 |
dtantsur | sambetts, do you plan on re-reviewing https://review.openstack.org/374381? I'd like to avoid it hanging there forever, so we can land the main thing and iterate over code later | 11:10 |
patchbot | patch 374381 - ironic-inspector-specs - Add LLDP processing hook and new CLI commands | 11:10 |
mgould | morning vdrok sambetts dtantsur lucasagomes aarefiev | 11:10 |
milan | morning mgould! :) | 11:11 |
aarefiev | morning mgould | 11:11 |
mgould | morning milan | 11:13 |
sambetts | dtantsur: I'll take a run through in a minute, I'm pretty happy with it from my last read through only one issue and that was fixed in the last patchset | 11:13 |
*** athomas has joined #openstack-ironic | 11:13 | |
dtantsur | thanks! | 11:14 |
openstackgerrit | Merged openstack/ironic-inspector-specs: Add LLDP processing hook and new CLI commands https://review.openstack.org/374381 | 11:27 |
*** AlexeyAbashkin has joined #openstack-ironic | 11:27 | |
milan | \o/ ^ :D | 11:28 |
mgould | \o/ | 11:28 |
*** chas_ has quit IRC | 11:29 | |
bfournie | morning ironic, sambetts: thanks for the review!! | 11:31 |
sambetts | np :) | 11:32 |
milan | aarefiev, w/r the not_state I feel the same: let's have a =nin: operator | 11:37 |
* milan has to propose a patch to that wg thing again I'm afraid | 11:38 | |
*** roeyc has joined #openstack-ironic | 11:38 | |
aarefiev | milan: I'd go as is for now, but sure you can propose | 11:38 |
milan | dtantsur, sambetts ^ guys, wdyt about blocking the spec again? https://review.openstack.org/#/c/375045 | 11:39 |
patchbot | patch 375045 - ironic-inspector-specs - Endpoint to list introspection statuses | 11:39 |
dtantsur | ugh | 11:40 |
aarefiev | imo we will lost a lot of time adding all missing functionality to wg first | 11:40 |
dtantsur | yep | 11:41 |
dtantsur | milan, should we maybe ask them how exactly they'd like to see it, and then just implement it? | 11:41 |
aarefiev | +1 | 11:42 |
milan | aarefiev, dtantsur I think there would be a bike shedding w/r how to name it: not_in, nin, what ever, but yeah, let's ping them first | 11:43 |
* milan joins #openstack-sdks | 11:44 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add portgroup to api-ref https://review.openstack.org/393834 | 11:44 |
dtantsur | aarefiev, fun, isn't it: https://bugzilla.redhat.com/show_bug.cgi?id=1394744 ? | 11:45 |
openstack | bugzilla.redhat.com bug 1394744 in openstack-tempest "TestSecurityGroupsBasicOps.test_port_security_disable_security_group test gets halted by giving request timed out." [Unspecified,New] - Assigned to dmellado | 11:45 |
aarefiev | dtantsur: ? I need context on | 11:47 |
*** fxpester has joined #openstack-ironic | 11:47 | |
dtantsur | aarefiev, a neutron tempest test fails if inspector tempest plugin is merely present | 11:48 |
dtantsur | this is not call for actions, I'm just showing you how strange our world is | 11:48 |
* dtantsur hates tempest already | 11:48 | |
*** fxpester|2 has quit IRC | 11:49 | |
dmellado | dtantsur: lol | 11:50 |
dtantsur | :) | 11:50 |
dtantsur | dmellado, really, I dunno, our plugin source code seems as dumb as it's possible | 11:51 |
dmellado | dtantsur: I do agree, so I'm trying to check if some timeout's the guilty | 11:51 |
dmellado | in any case there's a proposal for ocata (/me crosses fingers) | 11:51 |
dmellado | to have an experimental gate which would install all the plugins available | 11:51 |
dmellado | as there are some weird interactions... | 11:51 |
dtantsur | hah, interesting | 11:51 |
dtantsur | if we go further with "tempest-as-a-product" idea, we'll need help from QA folks sorting out such "interactions" | 11:52 |
*** rbartal has quit IRC | 11:52 | |
dmellado | also, what happens is that my bandwith is limited... | 11:52 |
* dmellado handling quite the work on manila now regarding to that... | 11:53 | |
* milan feels alone in #openstack-sdks :-/ | 11:53 | |
openstackgerrit | Galyna Zholtkevych proposed openstack/ironic: Retry connecting to vm after power-on https://review.openstack.org/394901 | 11:54 |
aarefiev | dtantsur: sometimes I have feelings that such tests fail due the reason how we configure network | 11:54 |
dtantsur | aarefiev, well, this is not devstack this time | 11:56 |
*** g3ek has quit IRC | 11:56 | |
aarefiev | dtantsur: I see, I agree it's really weird tempest plugin doesn't change any configuration | 12:00 |
*** jkilpatr has joined #openstack-ironic | 12:00 | |
*** roeyc has quit IRC | 12:02 | |
*** g3ek has joined #openstack-ironic | 12:03 | |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: WIP: TEST GRENADE DEV https://review.openstack.org/397126 | 12:03 |
*** rbartal has joined #openstack-ironic | 12:05 | |
openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector-specs: Endpoint to list introspection statuses https://review.openstack.org/375045 | 12:07 |
mgould | milan: so you're holding off on a generic not-in operator for now? | 12:15 |
mgould | good idea | 12:15 |
openstackgerrit | Szymon Borkowski proposed openstack/ironic-python-agent: Add a new Hardware Manager for X710 network card https://review.openstack.org/391836 | 12:15 |
milan | aarefiev, dtantsur I feel like the resolution w/r not in will be a new operator (there's already a neq operator), just the name might be a question; I've posted to the [api] list as channel gives no answers :-/ | 12:15 |
milan | mgould, I'd prefer that and aarefiev would as well I think | 12:15 |
milan | mgould, but I think it will delay all again :-/ | 12:16 |
milan | as I think the name for the operator might need agreeing upon | 12:16 |
mgould | yeah | 12:17 |
mgould | land what you've got, propose not-in as a separate change | 12:17 |
mgould | STTCPW | 12:17 |
mgould | small changes | 12:18 |
mgould | agile | 12:18 |
mgould | all those good things | 12:18 |
milan | mgould, my spec has not_set and set fields instead | 12:18 |
milan | not sure what's smaller change here | 12:18 |
*** priteau has quit IRC | 12:19 | |
milan | to deprecate not_set or =nin: if someone dislikes industrial rock bands ;) | 12:19 |
*** priteau has joined #openstack-ironic | 12:19 | |
* mgould is surprised the API WG haven't ruled on this already, TBH | 12:19 | |
milan | mgould, same here | 12:20 |
mgould | it's not like Boolean logic is a recent invention | 12:20 |
milan | lol | 12:20 |
mgould | "guys, 1854 just called, they say you've left out an operator" | 12:20 |
*** dprince has joined #openstack-ironic | 12:21 | |
* milan going to suggest a parentheses and negation to filtering api wg spec ;) | 12:21 | |
mgould | heh | 12:21 |
milan | * one day | 12:22 |
mgould | hard to fit into HTTP params, I guess | 12:22 |
milan | ** April the 1st | 12:22 |
*** gcb has quit IRC | 12:22 | |
milan | or February the 30th | 12:22 |
* milan not sure yet which one | 12:22 | |
mgould | it's enough to NOT individual tests, of course | 12:22 |
milan | ?<field>=[not:]<op>:value[,value...] | 12:23 |
mgould | probably not great UX, though | 12:23 |
mgould | yeah | 12:23 |
mgould | "first convert your query into conjunctive normal form..." | 12:24 |
*** priteau has quit IRC | 12:24 | |
*** priteau has joined #openstack-ironic | 12:25 | |
milan | mgould, that might actually be a good idea | 12:25 |
lucasagomes | dtantsur, https://review.openstack.org/#/c/397160/ (switching postgres to ipmitool) | 12:26 |
patchbot | patch 397160 - openstack-infra/project-config - Ironic: Switch postgres job to ipmitool | 12:26 |
openstackgerrit | Karthik S proposed openstack/ironic-inspector-specs: Spec for retrieving NUMA node information https://review.openstack.org/396147 | 12:27 |
milan | mgould, but a negated operator :not:in is just not_in imho | 12:28 |
*** wajdi has joined #openstack-ironic | 12:29 | |
milan | as there's no parentheses for us | 12:29 |
mgould | true | 12:29 |
mgould | do we want to negate all operators? | 12:29 |
*** wajdi_ has joined #openstack-ironic | 12:29 | |
mgould | lt, ge etc? | 12:30 |
mgould | not:lt is just ge, etc | 12:30 |
*** lucasagomes is now known as lucas-hungry | 12:30 | |
milan | mgould, that would be redundant I'm afraid | 12:30 |
mgould | so yeah, I'm not convinced this is a great idea | 12:30 |
mgould | it's consistent and formally adequate to express all possible queries, but that's about all it has going for it | 12:30 |
dtantsur | lucas-hungry, cool! I'm working on https://review.openstack.org/#/c/397164/ in the meantime | 12:31 |
patchbot | patch 397164 - openstack-infra/project-config - Make the whole matrix of Ironic jobs voting on New... | 12:31 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Prepare devstack to run all tests in one job. https://review.openstack.org/396155 | 12:32 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Test ironic concurrent tests https://review.openstack.org/397168 | 12:32 |
aarefiev | milan: I see a lot of frameworks using `nin` operator, so should be fine | 12:32 |
milan | aarefiev, cool :) | 12:33 |
*** wajdi has quit IRC | 12:33 | |
milan | aarefiev, let's put it into the list spec right away then | 12:33 |
aarefiev | "nin" sounds good for me, what others think | 12:34 |
*** AlexeyAbashkin has quit IRC | 12:35 | |
dtantsur | could someone please double-check me on project-config change https://review.openstack.org/#/c/397164/ ? | 12:35 |
patchbot | patch 397164 - openstack-infra/project-config - Make the whole matrix of Ironic jobs voting on New... | 12:35 |
milan | mgould, the query string field repetition is considered to be the field conjunction, so if we have set operators =in: and =nin: and if we use multiple queries as disjunction we have sort of a "racy" disjunctive normal form (as the items may shift between consecutive requests to the api) | 12:36 |
mgould | milan: yes, that's what I meant | 12:37 |
mgould | sorry, I was mixing up DNF and CNF | 12:37 |
milan | aarefiev, I'll post an update to our spec let's see what others will vote there | 12:40 |
mgould | who's working on ironic-tempest-plugin at the moment? | 12:41 |
* milan relocates O:-) | 12:44 | |
aarefiev | mgould: community? :) | 12:45 |
*** hw_wutianwei has quit IRC | 12:45 | |
*** milan has quit IRC | 12:46 | |
*** aparnav has quit IRC | 12:47 | |
mgould | aarefiev: sure, but is anyone specific working on it? | 12:49 |
mgould | alternatively, is anyone working on splitting it out into its own branchless repo as the Tempest folks request? | 12:51 |
aarefiev | mgould: hard to say what do you mean by "on ironic-tempest-plugin", a lot of people adding new tests | 12:51 |
mgould | <- not me, please not me... | 12:51 |
mgould | aarefiev: OK, thanks | 12:51 |
aarefiev | not sure | 12:51 |
mgould | ^^^ dmellado | 12:51 |
* dtantsur is still sure that we don't have anyone specifically dedicated to tempest plugins | 12:51 | |
aarefiev | mgould is volunteer :) | 12:52 |
mgould | fffffffuuuuuuuuuuuu.... | 12:53 |
jroll | dtantsur: I'm working on getting it out of tempest | 12:55 |
jroll | I could also work on separate repo | 12:55 |
jroll | it isn't hard, only lots of time-consuming changes | 12:55 |
jroll | mgould: ^^ | 12:55 |
jroll | also morning. | 12:55 |
mgould | phew | 12:55 |
mgould | morning jroll | 12:55 |
vdrok | morning jroll ! | 12:55 |
dtantsur | morning jroll :) I think dmellado is interested on more of full-time-ish accountability for our two tempest plugins | 12:56 |
*** bfournie has quit IRC | 12:57 | |
*** makowals has quit IRC | 12:57 | |
mgould | dtantsur: the second one being the grenade plugin? | 12:57 |
dtantsur | mgould, no, two tempest plugins: ironic and inspector | 12:58 |
dtantsur | mgould, and then two grenade plugins, similarly | 12:58 |
jroll | dtantsur: as in, 'who cares about it enough to deal with bugs and fix things'? | 12:58 |
mgould | dtantsur: gotcha | 12:58 |
dtantsur | jroll, 'who cares about its life outside of upsteam CI enough to deal with tricky bugs' | 12:59 |
jroll | dtantsur: I didn't see dmellado asking, only you, sorry :) | 12:59 |
jroll | ah | 12:59 |
dtantsur | yeah, the conversation started downstream, sorry | 12:59 |
*** makowals has joined #openstack-ironic | 12:59 | |
jroll | as always, accountability for a thing defaults to ptl, so I guess that's me :P | 13:00 |
mgould | dtantsur: I think the basic answer is "we only care about tempest as a CI tool" | 13:00 |
dtantsur | we've found someone to blame \o/ | 13:00 |
moshele | JayF: can you review my ironic-agent changes - https://review.openstack.org/#/c/334816/ | 13:00 |
patchbot | patch 334816 - ironic-python-agent - Add a new Hardware Manager for Mellanox NICs | 13:00 |
dtantsur | mgould, this is the point of future traction between project teams and the QA team, I guess :) | 13:00 |
jroll | dtantsur: I would like to make our tempest plugin not terrible (it's so terrible), if I can find a balance between hacking on that and everything else :) | 13:01 |
dtantsur | mgould, they want to see it as product, but they want us to be responsible for that (sorry dmellado if it sounds like a rude and shallow summary) | 13:01 |
jroll | and by make it not terrible, I mean be able to run it against more than devstack | 13:01 |
dtantsur | jroll, our plugin is terrible mostly because tempest is not so great either IMO | 13:01 |
jroll | (hm, maybe if we can make it run against real clouds, we can just fix our devstack plugin to match) | 13:01 |
jroll | dtantsur: well, we're beyond that now, don't have much choice | 13:02 |
dtantsur | the problem here, "real clouds" may be pretty diverse, some closer to devstack, some (undercloud in tripleo) substantially further | 13:02 |
jroll | well, they all have the same APIs | 13:02 |
jroll | that's the first thing we need to do, ensure all APIs work | 13:03 |
TheJulia | Good morning | 13:03 |
mgould | morning TheJulia | 13:03 |
dtantsur | but leaving aside undercloud, tempest's manner to load and execute all entrypoints by default does give us some headache | 13:03 |
dtantsur | morning TheJulia | 13:03 |
jroll | (but, it's weird in ironic because our resources are *real things*, so we can't really just issue node-create etc at will | 13:03 |
*** xavierr has joined #openstack-ironic | 13:03 | |
dtantsur | right | 13:04 |
dtantsur | anyway, I don't think we're going to move far with this work, if we say, "ok, the ptl is responsible for everything" | 13:05 |
mgould | is "must hit every API endpoint" a tempest requirement? | 13:05 |
dtantsur | an ideal outcome would be to run ironic tempest plugin against something that is NOT devstack | 13:05 |
dtantsur | mgould, dunno. I'd say it's a job for functional tests. | 13:06 |
jroll | mgould: tempest's primary goal is to test APIs | 13:06 |
jroll | or, test that things work via APIs | 13:06 |
mgould | yeah | 13:06 |
jroll | like a black box | 13:06 |
jroll | I think "can add a node to the pool" is a useful test for any ironic install :) | 13:06 |
mgould | I'm wondering if we can test a minimal subset of our API in the tempest plugin, and then run deeper tests via in-tree functional tests | 13:06 |
dtantsur | mgould, this is the plan iirc | 13:07 |
jroll | yeah, I think that's reasonable | 13:07 |
mgould | dtantsur: yep | 13:07 |
jroll | dtantsur: to be clear, what I was saying is that if nobody else will step up then it falls on me | 13:07 |
dtantsur | dunno if you scale well enough :) | 13:09 |
* dtantsur thinks of HA jroll | 13:09 | |
jroll | heh, agree :) | 13:09 |
jroll | so we need someone to step up :P | 13:09 |
*** ccamacho is now known as ccamacho|lunch | 13:11 | |
karthiks | dtantsur, regarding the spec for NUMA node information (https://review.openstack.org/#/c/396147) , we would like to bring it up in today's meeting. Can we add it under RFE review section ? | 13:11 |
patchbot | patch 396147 - ironic-inspector-specs - Spec for retrieving NUMA node information | 13:11 |
dtantsur | karthiks, what do you want to discuss? | 13:12 |
karthiks | dtantsur, Just to bring it to the team's notice. | 13:13 |
dtantsur | karthiks, then just mention it in the open discussion. | 13:13 |
karthiks | Thanks dtantsur | 13:13 |
dtantsur | np | 13:14 |
vsaienk0 | dtantsur, jroll could you please have look on https://review.openstack.org/#/c/396155/ The approach allows to run tests wit different driver/boot_mode/boot_option/image concurrently http://logs.openstack.org/68/397168/1/check/gate-tempest-dsvm-ironic-ipa-partition-agent_ipmitool-tinyipa-ubuntu-xenial-nv/5b0ab2b/console.html#_2016-11-14_13_01_19_800135 | 13:14 |
patchbot | patch 396155 - ironic - Prepare devstack to run all tests in one job. | 13:14 |
*** jtomasek has quit IRC | 13:14 | |
*** makowals has quit IRC | 13:16 | |
dtantsur | interesting! | 13:16 |
dtantsur | vsaienk0, do you think we can create flavors on fly instead of listing them? | 13:16 |
dtantsur | this relates to the previous discussion: we should not rely so much on devstack | 13:16 |
sambetts | ++ | 13:17 |
vsaienk0 | dtantsur: I think yes, we may create flavor on the fly | 13:17 |
jroll | this is cool, thought I'm not a huge fan of all the tempest configs there | 13:17 |
dtantsur | ditto for drivers, boot modes, etc: I think we should just get a bunch of nodes from devstack/any stack and then configure them | 13:18 |
dtantsur | vsaienk0, ^^^ | 13:18 |
sambetts | yeah it would be good for our tempest plugin to adapt to the environment its being run against instead of being configured before hand | 13:19 |
vsaienk0 | dtantsur: for boot mode we need to configure VM correctly, not only ironic node | 13:19 |
sambetts | vsaienk0: devstack should handle that though | 13:19 |
dtantsur | vsaienk0, right, ok, let's leave boot mode there. but boot option and driver can be configured on a node, right? | 13:19 |
sambetts | vsaienk0: right? | 13:19 |
sambetts | IMO tempest shouldn't even nessesarly be aware of the nodes configuration, it should just be aware that it did nova boot and it results in a deployed node | 13:21 |
sambetts | if we configure 5 nodes of different types in Ironic, then we just do nova boot 5 times and it'll test them all | 13:21 |
*** chas_ has joined #openstack-ironic | 13:22 | |
*** shuquan has joined #openstack-ironic | 13:24 | |
vsaienk0 | dtantsur: so if we are going to prepare the node before test we may even do not create a flavor for this test, and use availability zone to place instance on needed node http://paste.openstack.org/show/550163/ | 13:24 |
vsaienk0 | sambetts: we still need to be sured that instance from specific tests was placed to correct node | 13:25 |
vsaienk0 | sambetts: for example when using different images, we can't get partitioned with localboot working on CI with cirros, because it doesn't have grub | 13:26 |
TheJulia | sambetts: but, we can't assume, nor should we assume that the only ironic client will be nova with tempest. We can very easily write tests to connect and work directly with the ironic api for instance nodes. We just have to write the test such that it recognizes if the test is not viable based on what node configuration/properties are present. | 13:26 |
*** bfournie has joined #openstack-ironic | 13:27 | |
jroll | silly question: do we care if nova is involved for boot mode, boot option, etc? | 13:28 |
jroll | with nova we only care that we can boot a thing, and probably that multitenant works, yeah? | 13:28 |
sambetts | yeah | 13:29 |
TheJulia | I think that we really shouldn't care | 13:29 |
jroll | ok | 13:29 |
jroll | feels like we should do most testing without nova | 13:29 |
sambetts | ++ | 13:29 |
jroll | which makes it easy to make sure the right node is chosen | 13:29 |
TheJulia | +1000 | 13:29 |
vdrok | morning TheJulia | 13:29 |
*** makowals has joined #openstack-ironic | 13:30 | |
*** hoangcx has joined #openstack-ironic | 13:30 | |
TheJulia | jroll: exactly, and also if a 3rd party CI is such that a feature is not available, then it is super easy for us to identify that and gracefully handle, as opposed to making an unhappy CI :) | 13:30 |
jroll | maybe we should write up where we want to be, before we start diving into changes | 13:30 |
TheJulia | +++++++ | 13:30 |
jroll | TheJulia: true that | 13:30 |
vdrok | had a patch on adding a basic methods and a test without nova - https://review.openstack.org/327238, will try to update it this week | 13:30 |
patchbot | patch 327238 - ironic - Add node provisioning methods to the tempest plugin | 13:31 |
jroll | vdrok: that's missing my point - let's agree on what we want the end result to be, first, before we start throwing code around | 13:31 |
TheJulia | vdrok and I both have revs up for doing direct testing if memory serves | 13:31 |
* jroll could write a short novel this week | 13:31 | |
vdrok | jroll: sure | 13:32 |
TheJulia | jroll: would a higher bandwidth discussion be more suitable for gathering/context setting for direct tempest testing? | 13:32 |
jroll | TheJulia: maybe? | 13:32 |
TheJulia | I kind of feel like context setting is necessary in that, which makes me wonder if IRC is the right medium for that discussion | 13:33 |
TheJulia | I dunno | 13:33 |
jroll | TheJulia: thinking I could write up a thing first, and then go from there | 13:33 |
jroll | yeah | 13:33 |
TheJulia | jroll: If your up to the typing, sure :) | 13:33 |
jroll | TheJulia: I haven't typed much all weekend, fingers are ready to go :) | 13:34 |
TheJulia | heh :) | 13:34 |
dmellado | hey dtantsur jroll mgould | 13:34 |
dmellado | just read it | 13:34 |
dmellado | xD | 13:34 |
jroll | \o dmellado | 13:34 |
*** dims has joined #openstack-ironic | 13:35 | |
dmellado | jroll: I just wanted to know if there was someobody handling it | 13:36 |
dmellado | s/someobody/somebody | 13:36 |
vsaienk0 | jroll: so your suggestion is to introduce an Ironic only tests to test different combinations of drivers/boot_modes/boot_options/images whatever ironic related and launch them concurrently with current baremetal_basic ops? | 13:37 |
jroll | dmellado: what, specifically, is 'it'? the bug you mentioned? the productization of the tempest plugin? making our tempest plugin not terrible? | 13:37 |
dmellado | jroll: basically, the latter | 13:37 |
jroll | vsaienk0: something like that, I think | 13:37 |
dmellado | I've found out that there were some issues while running ironic-inspector plugin | 13:37 |
dmellado | but I'd need to check also the upstream gates to see if this reproduces | 13:38 |
jroll | dmellado: team effort, I guess, there's people working on some individual efforts, but we need a bigger picture view (which we just decided I will start working on and sounds like other folks are interested in helping) | 13:39 |
vsaienk0 | jroll: will work in that direction :) | 13:39 |
dmellado | jroll: ack, sounds great | 13:40 |
jroll | vsaienk0: the same thing I said to vdrok: that's missing my point - let's agree on what we want the end result to be, first, before we start throwing code around | 13:40 |
dmellado | let me know if I can somehow help | 13:40 |
jroll | vsaienk0: I don't want to waste your time if we don't have the same clear picture in our heads | 13:40 |
jroll | dmellado: will do, thanks :) | 13:40 |
*** baoli has joined #openstack-ironic | 13:41 | |
*** baoli has quit IRC | 13:42 | |
*** lucas-hungry is now known as lucasagomes | 13:43 | |
lucasagomes | dtantsur, awesome! | 13:43 |
*** Michael-ZTE has joined #openstack-ironic | 13:44 | |
vsaienk0 | jroll: concerned about one thing, that we will re-implement a lot of nova stuff for stand alone tests, might be just use availability zones and place instance directly on node we want by "--availability-zone ::f2a411e0-814f-4fe8-9078-9a47cc45c7c5" | 13:44 |
*** krtaylor has quit IRC | 13:45 | |
*** Michael-ZTE has quit IRC | 13:46 | |
*** ccamacho|lunch is now known as ccamacho | 13:47 | |
*** xavierr has quit IRC | 13:47 | |
*** jcoufal has joined #openstack-ironic | 13:48 | |
TheJulia | vsaienk0: I don't think we would re-implement that much, because we know when it will work and not work. But I think I'd like to see what jroll writes before we dig into it much further. | 13:49 |
*** AlexeyAbashkin has joined #openstack-ironic | 13:49 | |
vsaienk0 | TheJulia: well at least we need to fill instance_info, manage neutron ports by our own. | 13:50 |
jroll | well, tempest shouldn't be assuming availability zones exist, nor creating them | 13:50 |
openstackgerrit | Merged openstack/ironic-ui: Fixed bug in representation of ironic state machine https://review.openstack.org/394997 | 13:51 |
jroll | that's my concern, our tempest plugin shouldn't be requiring or making things on the clouds it's testing | 13:51 |
vsaienk0 | jroll, no need to create availability zones "--availability-zone ::<ironic-node-uuid>" | 13:51 |
openstackgerrit | Merged openstack/ironic-ui: Add a check for node last_error equal to null https://review.openstack.org/394985 | 13:51 |
TheJulia | vsaienk0: yes, also what roll just said. These tests should be minimal and focus on interacting with the blackbox of ironic as a complete unknown. | 13:51 |
dtantsur | jroll, vsaienk0, my other thought was to leave only base full-flow tests (1-2) in tempest, then keep a lot more in tree | 13:51 |
*** krtaylor has joined #openstack-ironic | 13:51 | |
jroll | vsaienk0: that doesn't work, does it? | 13:51 |
vsaienk0 | jroll: nova allows to place instance on specific ironic node by adding "--availability-zone ::<ironic-node-uuid>" | 13:51 |
jroll | O_O | 13:51 |
* jroll has never seen that | 13:51 | |
* dtantsur neither | 13:51 | |
TheJulia | I've heard of it | 13:52 |
TheJulia | never tried it | 13:52 |
jroll | that's... wow | 13:52 |
jroll | anyway, breakfast is cooking, bbiab | 13:52 |
dtantsur | this is cool, I know that tripleo has reinvented the same thing essentially | 13:52 |
TheJulia | But truthfully, knowing where to place the node explicitly is kind of breaking the black box in a sense. | 13:52 |
lucasagomes | sounds like a feature :-) | 13:52 |
vsaienk0 | jroll: it nova does: http://paste.openstack.org/show/550163/ | 13:52 |
vsaienk0 | jroll: 06 INFO nova.scheduler.host_manager [req-2d4e8c23-01e0-4839-8c68-00b2cf283ad9 admin admin] Host filter forcing available nodes to f2a411e0-814f-4fe8-9078-9a47cc45c7c5 | 13:53 |
sambetts | I agree with TheJulia, going through nova but forcing a particular node seems like we're not testing things properly | 13:53 |
sambetts | the whole point of having nova there is to do scheduling etc | 13:54 |
vsaienk0 | sambetts: so lets use flavors than | 13:54 |
openstackgerrit | Merged openstack/bifrost: Add instructions to setup ip on provisiong on virsh https://review.openstack.org/395062 | 13:54 |
sambetts | that would be like always testing KVM via nova | 13:54 |
TheJulia | yup, we can just kind of grab one, and if it doesn't work, maybe try 2-3 more times, since test environments generally should be fairly consistent. If NOT... well... someone can propose logic to the test to handle their differences from that perspective | 13:55 |
sambetts | one of the major benefits to testing without nova is that we don't need to spin up a whole devstack | 13:55 |
vsaienk0 | sambetts: we missing main point of ironic testing I think, we don't want to test nova scheduler itself, we need to test Ironic instead, if nova allows to pick hypervisor directly why not use it? | 13:56 |
sambetts | we only need to install Ironic and operate it in standalone | 13:56 |
dtantsur | ++ | 13:58 |
vsaienk0 | sound like we need a spec to discuss all things related to ironic testing :) | 13:58 |
TheJulia | Lets let jim write, and go from there | 13:58 |
*** ohamada has quit IRC | 13:59 | |
*** ohamada has joined #openstack-ironic | 14:00 | |
*** gcb has joined #openstack-ironic | 14:01 | |
*** milan has joined #openstack-ironic | 14:02 | |
*** roeyc has joined #openstack-ironic | 14:02 | |
*** AlexeyAbashkin has quit IRC | 14:02 | |
*** rloo has joined #openstack-ironic | 14:03 | |
dtantsur | vsaienk0, jroll, sambetts, TheJulia, I've started collecting my thoughts, feel free to amend/add to a spec: https://etherpad.openstack.org/p/ironic-testing-future | 14:03 |
mgould | dtantsur: looks reasonable | 14:04 |
*** khamtamtun has joined #openstack-ironic | 14:04 | |
*** clenimar has joined #openstack-ironic | 14:04 | |
TheJulia | I worry we are over-thinking this some. I strongly urge for us to let our fearless leader compose some thoughts, then context set as a group so we are on the same page and can then proceed in the same direction together. | 14:07 |
mgould | +1 | 14:08 |
dtantsur | well, I've been thinking about this for quite some time, hence this thought dump | 14:08 |
TheJulia | ++ | 14:08 |
* dtantsur was thinking about proposing a spec himself, but he'll gladly pass it to jroll :) | 14:09 | |
TheJulia | I'm just trying to stress that we need to get on the same page, thought dumps are good, and I'm sure jroll will read and incorporate it. Just, I worry that discussing in many different directions without starting at the same place just brings us more risk than anything else. | 14:09 |
jroll | less risk and more wasted time :) | 14:10 |
*** charlledaniel has joined #openstack-ironic | 14:10 | |
charlledaniel | morning all | 14:10 |
TheJulia | words are hard :) | 14:10 |
mgould | morning charlledaniel | 14:10 |
*** ricardoas has joined #openstack-ironic | 14:11 | |
jroll | dtantsur: thanks for the pad, I'll pile in on that | 14:11 |
*** ElCoyote_ has quit IRC | 14:11 | |
dtantsur | cool! | 14:11 |
*** lindycoder has joined #openstack-ironic | 14:12 | |
*** priteau has quit IRC | 14:13 | |
*** [1]cdearborn has joined #openstack-ironic | 14:13 | |
dtantsur | bfournie, morning! congrats on getting your spec merged | 14:13 |
dtantsur | also morning charlledaniel | 14:13 |
milan | aarefiev, w/r the not-in operator I'm wondering whether we should support the conjunction of multiple fields as is the case w/ e.g started_at for consistency sake; I'd +0.5 vote for not actually | 14:14 |
*** baoli has joined #openstack-ironic | 14:14 | |
*** [1]cdearborn has quit IRC | 14:14 | |
milan | morning bfournie, charlledaniel, jroll, TheJulia! :) | 14:14 |
bfournie | morning dtantsur! | 14:14 |
*** tuan_luong has joined #openstack-ironic | 14:14 | |
TheJulia | good morning milan o/ | 14:15 |
bfournie | morning milan | 14:15 |
milan | :) | 14:15 |
tuan_luong | morning ironic | 14:15 |
milan | morning tuan_luong! :) | 14:15 |
tuan_luong | morning milan | 14:15 |
*** xavierr has joined #openstack-ironic | 14:15 | |
*** Goneri has joined #openstack-ironic | 14:16 | |
*** priteau has joined #openstack-ironic | 14:16 | |
*** khamtamtun has quit IRC | 14:16 | |
*** pcaruana has quit IRC | 14:16 | |
bfournie | morningmilan | 14:18 |
*** makowals has quit IRC | 14:18 | |
aarefiev | milan: imo we shouldn't support conjunction at all, as result return empty list | 14:19 |
milan | aarefiev, ack, I feel the same | 14:20 |
* milan puts that into the spec | 14:20 | |
*** shuquan has quit IRC | 14:21 | |
milan | aarefiev, too bad sambetts didn't like the ?finished_at=between:a,b operator :P all parsing would have to consider only a single operator | 14:21 |
*** wajdi_ has quit IRC | 14:22 | |
*** makowals has joined #openstack-ironic | 14:23 | |
sambetts | milan: I liked it ;) I just thought it was redundent :-P | 14:23 |
milan | sambetts then you shouldn't have -1 ed it :P | 14:24 |
*** chas_ has quit IRC | 14:24 | |
*** khamtamtun has joined #openstack-ironic | 14:25 | |
sambetts | -1s aren't always death sentences if there is a valid reason to keep something they can be resovled | 14:25 |
*** chas_ has joined #openstack-ironic | 14:25 | |
aarefiev | yeah, more work :) | 14:25 |
milan | sambetts, yeah, jk | 14:25 |
milan | too bad wg just approved it I'd really like to have between | 14:25 |
milan | it's much clearer to interpret | 14:26 |
milan | not sure whether you guys followed the last revision discussion | 14:26 |
sambetts | unfortunatly I've been tied up with the ironic networking stuff | 14:27 |
milan | sambetts, anyway as an ultimate punishment you'll be reviewing my parser patch I guess :P | 14:27 |
sambetts | hehe | 14:28 |
*** chas_ has quit IRC | 14:29 | |
*** galyna has quit IRC | 14:30 | |
*** galyna has joined #openstack-ironic | 14:30 | |
tuan_luong | morning ironic teams ^:^ | 14:32 |
tuan_luong | I want to discuss: should we use __slots__ in ironic driver class http://book.pythontips.com/en/latest/__slots__magic.html | 14:33 |
*** gcb has quit IRC | 14:35 | |
tuan_luong | hello sambetts dtantsur, aarefiev milan | 14:36 |
tuan_luong | How do you thinks about that? | 14:36 |
milan | tuan_luong, hey, do you have the patch link handy? | 14:37 |
tuan_luong | http://book.pythontips.com/en/latest/__slots__magic.html | 14:37 |
milan | tuan_luong, you're trying to reduce memory footprint correct? | 14:37 |
tuan_luong | Yes right milan | 14:37 |
tuan_luong | and I think dtantsur had mention about that in https://specs.openstack.org/openstack/ironic-specs/specs/not-implemented/driver-composition-reform.html#performance-impact | 14:38 |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: WIP: TEST GRENADE DEV https://review.openstack.org/397126 | 14:38 |
milan | tuan_luong, yeah, I remember now | 14:38 |
milan | tuan_luong, I'd +1 that, maybe worth a spec though | 14:39 |
*** jaosorior has quit IRC | 14:39 | |
milan | what do the cores think? | 14:39 |
dtantsur | tuan_luong, this code does not exist yet, please don't refer to it | 14:40 |
milan | dtantsur, maybe? ^^ | 14:40 |
jroll | I'd say RFE, probably not a spec, but I'd rather just do it as part of driver composition | 14:40 |
tuan_luong | I just commit small in agent driver but actually, I have some unclear in my messages | 14:40 |
*** jaosorior has joined #openstack-ironic | 14:40 | |
dtantsur | I don't care about specs as much as I care about benchmarks that show performance improvements for classic drivers | 14:40 |
jroll | ++ | 14:40 |
lucasagomes | tuan_luong, you mean using slots for the driver interfaces (power, provision etc...) attributes ? | 14:40 |
milan | ah, dtantsur's answer is faster than my question is asked :D | 14:40 |
tuan_luong | hi lucasagomes, yes | 14:40 |
lucasagomes | since we know them beforehand, I don't see why not optimize it | 14:40 |
lucasagomes | so I'm +1 with the idea | 14:40 |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: WIP: TEST GRENADE DEV https://review.openstack.org/397126 | 14:40 |
* dtantsur reminds that only BareDriver objects are not singletons | 14:40 | |
*** cdearborn has joined #openstack-ironic | 14:41 | |
tuan_luong | milan, I thinks when we have 1000 BM | 14:41 |
*** khamtamtun has quit IRC | 14:41 | |
dtantsur | it does not matter how many BM you have, you still have one object per ironic process | 14:41 |
tuan_luong | it will helpfull | 14:41 |
dtantsur | (unless I really misremember how we implement drivers) | 14:42 |
jroll | dtantsur: nope, you're correct | 14:42 |
lucasagomes | yeah they are singletons afaiui | 14:42 |
*** syed__ has joined #openstack-ironic | 14:42 | |
milan | dtantsur, jroll https://specs.openstack.org/openstack/ironic-specs/specs/not-implemented/driver-composition-reform.html#performance-impact | 14:42 |
*** khamtamtun has joined #openstack-ironic | 14:42 | |
dtantsur | so it might be worth adding __slots__ to BareDriver | 14:42 |
dtantsur | milan, I wrote that, but this code does not exist | 14:43 |
jroll | milan: yes, I read it | 14:43 |
*** khamtamtun has quit IRC | 14:43 | |
dtantsur | so yes, BareDriver is already created per node, it might be worth optimizing it | 14:43 |
jroll | +1 for BareDriver.__slots__ | 14:43 |
*** _fortis has joined #openstack-ironic | 14:43 | |
dtantsur | optimizing driver interfaces seems useless both before and after driver composition | 14:43 |
dtantsur | tuan_luong, tl;dr: only this object is created per node: https://github.com/openstack/ironic/blob/master/ironic/drivers/base.py#L158 so you only need __slots__ there | 14:44 |
milan | yeah I'd therefore vote for a spec that proves this is necessary | 14:44 |
* dtantsur is hacking on driver_factor as we speak, so it's all pretty fresh in his mind :) | 14:45 | |
*** yufei has joined #openstack-ironic | 14:45 | |
milan | dtantsur, ack :) | 14:45 |
*** chas_ has joined #openstack-ironic | 14:45 | |
*** godp1301 has joined #openstack-ironic | 14:45 | |
tuan_luong | so dtantsur, I will up new patch tomorrow, | 14:48 |
dtantsur | thanks | 14:48 |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Add database fields for all interfaces https://review.openstack.org/396681 | 14:49 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Update iptables rules and services IPs for multinode https://review.openstack.org/368648 | 14:49 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add devstack setup_vxlan_network() https://review.openstack.org/368647 | 14:49 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Skip some steps for multinode case https://review.openstack.org/368646 | 14:49 |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Testing multinode stuff https://review.openstack.org/368173 | 14:49 |
*** chas_ has quit IRC | 14:50 | |
*** ppiela has joined #openstack-ironic | 14:50 | |
jroll | TheJulia: vsaienk0: sambetts: agree with the goals here at line 36? if so, I'll continue https://etherpad.openstack.org/p/ironic-testing-future | 14:51 |
openstackgerrit | Oksana Voshchana proposed openstack/ironic: WIP: TEST GRENADE DEV https://review.openstack.org/397126 | 14:52 |
dtantsur | jroll, btw we should finally start testing root device hints in CI | 14:53 |
jroll | dtantsur: indeed | 14:53 |
dtantsur | two people already told me they don't work for them in newton... | 14:54 |
sambetts | jroll: I agree with them, I think we should also make our devstack environment match up closer to a real cloud | 14:54 |
jroll | sambetts: yeah, I think that's a separate topic, I want to focus on tempest here | 14:54 |
dtantsur | jroll, wdym by "tempest" in this context? both full-flow and features tests (using terms from the etherpad)? | 14:55 |
jroll | dtantsur: yes | 14:55 |
dtantsur | k | 14:55 |
openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector-specs: Endpoint to list introspection statuses https://review.openstack.org/375045 | 14:57 |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] Support defining and loading hardware types https://review.openstack.org/336626 | 14:58 |
lucasagomes | jroll, around ? Mind taking a looking at https://review.openstack.org/#/c/396636/ and https://review.openstack.org/#/c/396640/ (they are backports to newtron, already +2/+1'ed) | 15:03 |
patchbot | patch 396636 - ironic-lib (stable/newton) - Correctly count the number of primary partitions | 15:03 |
patchbot | patch 396640 - ironic-lib (stable/newton) - followup to disk_utils.count_mbr_partitions() | 15:03 |
jroll | lucasagomes: yeah, dmitry pinged me earlier but got distracted | 15:04 |
lucasagomes | jroll, thanks, no worries | 15:05 |
jroll | lucasagomes: done, thanks for the ping | 15:07 |
lucasagomes | jroll, o/ thanks for the +a | 15:07 |
jroll | np | 15:08 |
*** watanabe_isao has joined #openstack-ironic | 15:13 | |
*** mrtenio has joined #openstack-ironic | 15:13 | |
*** links has quit IRC | 15:13 | |
*** watanabe_isao has quit IRC | 15:17 | |
*** causten has joined #openstack-ironic | 15:18 | |
*** ricardoas has quit IRC | 15:21 | |
*** jtomasek has joined #openstack-ironic | 15:21 | |
*** clenimar has quit IRC | 15:22 | |
*** roeyc has quit IRC | 15:23 | |
*** charlledaniel has quit IRC | 15:23 | |
*** mrtenio has quit IRC | 15:23 | |
*** xavierr has quit IRC | 15:23 | |
*** Goneri has quit IRC | 15:25 | |
*** wajdi has joined #openstack-ironic | 15:28 | |
*** rbrud has joined #openstack-ironic | 15:31 | |
*** makowals has quit IRC | 15:33 | |
openstackgerrit | Merged openstack/ironic-lib: Correctly count the number of primary partitions https://review.openstack.org/396636 | 15:35 |
*** gcb has joined #openstack-ironic | 15:35 | |
*** rajinir has joined #openstack-ironic | 15:35 | |
*** Goneri has joined #openstack-ironic | 15:38 | |
dtantsur | sambetts, hey! have you seen such behavior? https://bugzilla.redhat.com/show_bug.cgi?id=1379010 | 15:41 |
openstack | bugzilla.redhat.com bug 1379010 in openstack-ironic "Cisco UCS 200-M3 blades will not properly boot over the network from IPMI" [Unspecified,New] - Assigned to lmartins | 15:41 |
*** makowals has joined #openstack-ironic | 15:44 | |
*** tuan_luong has quit IRC | 15:46 | |
*** jheroux has joined #openstack-ironic | 15:46 | |
sambetts | dtantsur: this bug came up a little while ago I think, as far as I know its a bug in the firmware which is why I told lucasagomes to ask if they had updated it recently, outside of this bug report I've had no information about this bug, I've also had no contact regarding the Cisco drivers, if they are trying to load standalone UCS servers with the UCS(M) driver then they will be sad... (I need | 15:47 |
sambetts | to get that driver renamed, I was going to put up a patch but then driver recomp happened sooo...) | 15:47 |
dtantsur | sambetts, but now you've got a driver for standalone UCS servers, right? | 15:47 |
sambetts | dtantsur: yeah, its the CIMC driver | 15:48 |
dtantsur | sambetts, should we recommend them try it? | 15:48 |
sambetts | yeah probably worth a try | 15:48 |
sambetts | I would love bug reports too | 15:49 |
*** Sukhdev has joined #openstack-ironic | 15:49 | |
openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Bring configurations from tempest to ironic_tempest_plugin https://review.openstack.org/355591 | 15:50 |
*** dtantsur is now known as dtantsur|brb | 15:50 | |
*** moshele has left #openstack-ironic | 15:50 | |
*** hoangcx has quit IRC | 15:52 | |
openstackgerrit | Merged openstack/ironic-lib: followup to disk_utils.count_mbr_partitions() https://review.openstack.org/396640 | 15:53 |
*** logan- has quit IRC | 15:54 | |
*** UForgotten has quit IRC | 15:58 | |
*** UForgotten has joined #openstack-ironic | 15:58 | |
*** hshiina has joined #openstack-ironic | 16:00 | |
*** sabeen has joined #openstack-ironic | 16:00 | |
*** logan- has joined #openstack-ironic | 16:01 | |
*** sabeen has quit IRC | 16:05 | |
*** mjura has quit IRC | 16:09 | |
*** zackf has joined #openstack-ironic | 16:10 | |
jlvillal | heh: Can't Stop the SUSE https://youtu.be/A-Rn0iQEpc8 :) | 16:10 |
*** makowals has quit IRC | 16:11 | |
*** sabeen has joined #openstack-ironic | 16:13 | |
*** jrcloud has joined #openstack-ironic | 16:15 | |
openstackgerrit | Michael Turek proposed openstack/ironic-specs: Add spec for OpenBMC driver https://review.openstack.org/388218 | 16:17 |
*** zackf has quit IRC | 16:17 | |
*** yufei has quit IRC | 16:20 | |
*** makowals has joined #openstack-ironic | 16:20 | |
lucasagomes | jlvillal, heh pretty cool | 16:21 |
jlvillal | lucasagomes: :) | 16:21 |
*** khamtamtun has joined #openstack-ironic | 16:23 | |
*** Goneri has quit IRC | 16:24 | |
*** absubram has joined #openstack-ironic | 16:24 | |
*** pcaruana has joined #openstack-ironic | 16:24 | |
*** dtantsur|brb is now known as dtantsur | 16:27 | |
*** chas_ has joined #openstack-ironic | 16:27 | |
*** mrtenio has joined #openstack-ironic | 16:29 | |
*** milan has quit IRC | 16:30 | |
*** ChubYann has joined #openstack-ironic | 16:31 | |
*** gcb has quit IRC | 16:34 | |
*** Goneri has joined #openstack-ironic | 16:36 | |
*** fragatina has quit IRC | 16:37 | |
*** xavierr has joined #openstack-ironic | 16:38 | |
*** amotoki has quit IRC | 16:39 | |
*** rpioso has joined #openstack-ironic | 16:39 | |
*** beekneemech is now known as bnemec | 16:40 | |
openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Add portgroups RFE to ml2 integration spec https://review.openstack.org/397290 | 16:43 |
jroll | ^ super easy one | 16:43 |
*** khamtamtun has quit IRC | 16:43 | |
openstackgerrit | Grzegorz Grasza (xek) proposed openstack/ironic-specs: Support rolling upgrades https://review.openstack.org/299245 | 16:46 |
*** rloo has quit IRC | 16:47 | |
*** rloo has joined #openstack-ironic | 16:48 | |
*** vsaienk0 has left #openstack-ironic | 16:48 | |
*** vsaienk0 has joined #openstack-ironic | 16:48 | |
*** jrcloud has quit IRC | 16:48 | |
*** stendulker has joined #openstack-ironic | 16:49 | |
*** khamtamtun has joined #openstack-ironic | 16:49 | |
vsaienk0 | jroll: are you going to upload a new patchset/reply here? https://review.openstack.org/#/c/387534/ ? | 16:49 |
patchbot | patch 387534 - nova-specs - Proposal for ironic static portgroups support | 16:49 |
jroll | vsaienk0: literally just did :) | 16:49 |
vsaienk0 | :) | 16:50 |
rloo | hi and happy Monday morning ironickers! | 16:51 |
jroll | vsaienk0: I'll do back-and-forth with nova today, if there's work to do in your morning go ahead and do it :) | 16:51 |
jroll | rloo: morning! | 16:51 |
jlvillal | vsaienk0: Any ideas on the failure here? http://logs.openstack.org/68/389268/3/experimental/gate-grenade-dsvm-ironic-multitenant-nv/7818fa9/ | 16:51 |
rloo | < 10 minutes to update subteam reports for all you leads out there :D | 16:51 |
vdrok | morning jlvillal and rloo ! | 16:51 |
jlvillal | vsaienk0: In console.html | 16:51 |
dtantsur | morning rloo, jlvillal | 16:52 |
jlvillal | vdrok: dtantsur rloo vsaienk0 lucasagomes sambetts jroll: Good morning :) | 16:53 |
openstackgerrit | Merged openstack/ironic-specs: Add portgroups RFE to ml2 integration spec https://review.openstack.org/397290 | 16:53 |
*** KennethWilke has quit IRC | 16:53 | |
lucasagomes | hi there | 16:53 |
* dtantsur is confused by having both "CI refactoring" and "Gate improvements" in the whiteboard | 16:53 | |
rloo | dtantsur: the CI refactoring was a 'thing' in ocata priorities. | 16:54 |
rloo | dtantsur: 'gate improvements' seemed to be reporting on any gate improvements. | 16:54 |
rloo | dtantsur: should i delete 'gate improvements'? | 16:54 |
jroll | rloo: 'gate improvements' was a 'thing' in newton priorities :) | 16:55 |
*** diablo_rojo has joined #openstack-ironic | 16:55 | |
rloo | jroll: OH. i will delete it then. jlvillal used it to mention pep8 jobs. jlvillal -- what do you think? | 16:55 |
jroll | I guess I could go either way, just pointing that out | 16:56 |
jlvillal | rloo: I just picked a section which looked like it applied. | 16:56 |
jlvillal | rloo: Not sure if I noticed every section that could apply. | 16:56 |
*** rbartal has quit IRC | 16:56 | |
rloo | jroll: well, it is confusing so we need to clarify somehow. | 16:56 |
lucasagomes | dtantsur, oh I probably did that too | 16:56 |
jlvillal | rloo: If you want to move it to a different section, good with me. | 16:56 |
dtantsur | rloo, please merge one into the other :) | 16:56 |
lucasagomes | ++ | 16:56 |
rloo | ok, i'll move them all to CI... | 16:57 |
vdrok | jlvillal: the failure is in http://logs.openstack.org/68/389268/3/experimental/gate-grenade-dsvm-ironic-multitenant-nv/7818fa9/logs/grenade.sh.txt.gz#_2016-11-12_19_13_56_877, instance can't be pinged | 16:57 |
vdrok | jlvillal: http://logs.openstack.org/68/389268/3/experimental/gate-grenade-dsvm-ironic-multitenant-nv/7818fa9/logs/old/ironic-bm-logs/node-6_console_12-11-2016-19:13:24.txt.gz seems it was not able to start up | 16:57 |
jlvillal | vdrok: Thanks. Okay. I'll dig in to figure out why adding multitenant to the grenade job make it not work. | 16:58 |
jlvillal | vdrok: I wonder if the current multinode changes might already have a fix ???? | 16:58 |
* jlvillal might want to base his patches on the multinode stuff.... | 16:59 | |
vdrok | jlvillal: going to look after the meeting :) | 16:59 |
jroll | meeting tiiiime | 17:00 |
*** zackf has joined #openstack-ironic | 17:03 | |
*** Sukhdev has quit IRC | 17:04 | |
*** e0ne has quit IRC | 17:05 | |
*** jaosorior has quit IRC | 17:05 | |
*** jaosorior has joined #openstack-ironic | 17:08 | |
*** chas__ has joined #openstack-ironic | 17:10 | |
openstackgerrit | Vladyslav Drok proposed openstack/ironic-specs: Add bonding configuration fields to portgroups https://review.openstack.org/396610 | 17:11 |
*** fragatina has joined #openstack-ironic | 17:12 | |
*** milan has joined #openstack-ironic | 17:12 | |
*** chas_ has quit IRC | 17:12 | |
*** hshiina has left #openstack-ironic | 17:19 | |
*** clsacramento has quit IRC | 17:21 | |
*** xek has joined #openstack-ironic | 17:25 | |
*** rbudden has quit IRC | 17:25 | |
*** aNuposic has joined #openstack-ironic | 17:28 | |
*** khamtamtun has quit IRC | 17:29 | |
*** khamtamtun has joined #openstack-ironic | 17:29 | |
*** ohamada has quit IRC | 17:30 | |
*** causten has quit IRC | 17:30 | |
*** causten has joined #openstack-ironic | 17:32 | |
*** KennethWilke has joined #openstack-ironic | 17:34 | |
*** chas__ has quit IRC | 17:34 | |
openstackgerrit | Merged openstack/python-ironic-inspector-client: UUID, started_at, finished_at in the status https://review.openstack.org/395609 | 17:35 |
*** chas_ has joined #openstack-ironic | 17:35 | |
* jroll hopes fiancee brings lunch home before his 1800 meeting, steps away anyway | 17:36 | |
*** jpich has quit IRC | 17:36 | |
openstackgerrit | Debayan Ray proposed openstack/ironic-specs: EFI iSCSI boot for ProLiant hardware https://review.openstack.org/207337 | 17:37 |
*** mgoddard has quit IRC | 17:37 | |
*** mgoddard has joined #openstack-ironic | 17:38 | |
*** [1]rpioso has joined #openstack-ironic | 17:39 | |
*** chas_ has quit IRC | 17:39 | |
*** rama_y has joined #openstack-ironic | 17:39 | |
*** khamtamtun has quit IRC | 17:40 | |
vdrok | jlvillal: https://review.openstack.org/368647 this one might be the fix for your issue, at least it looks related :) | 17:40 |
patchbot | patch 368647 - ironic - Add devstack setup_vxlan_network() | 17:40 |
jlvillal | vdrok: Thanks! :) | 17:40 |
openstackgerrit | Debayan Ray proposed openstack/ironic-specs: EFI iSCSI boot for ProLiant hardware https://review.openstack.org/207337 | 17:41 |
*** rpioso has quit IRC | 17:41 | |
lucasagomes | calling it a day folks, have a great evening all! | 17:41 |
vdrok | night lucasagomes | 17:42 |
lucasagomes | o/ | 17:42 |
xavierr | bye lucasagomes | 17:42 |
*** lucasagomes is now known as lucas-afk | 17:42 | |
*** athomas has quit IRC | 17:42 | |
vdrok | jlvillal: ah, you're trying multitenant, not multinode, then I'm not sure what's the problem | 17:42 |
jlvillal | vdrok: Yeah. I'm trying to figure it out. | 17:43 |
jlvillal | vdrok: I know the 'multiNODE' patches are using multiTENANT | 17:43 |
jlvillal | vdrok: So likely something related to what I am doing. I'll dig more into the error and look at the patches. | 17:43 |
jlvillal | vdrok: Thanks. | 17:43 |
*** khamtamtun has joined #openstack-ironic | 17:46 | |
NobodyCam | Good Morning Ironic'ers | 17:49 |
NobodyCam | night lucas-afk | 17:50 |
mgould | morning NobodyCam | 17:50 |
NobodyCam | :) | 17:50 |
mgould | and good night Ironicers :-) | 17:50 |
*** mgould is now known as mgould|afk | 17:50 | |
NobodyCam | night mgould|afk | 17:50 |
NobodyCam | :p | 17:50 |
*** jaosorior has quit IRC | 17:51 | |
*** devananda|away is now known as devananda | 17:52 | |
dtantsur | morning NobodyCam | 17:52 |
NobodyCam | hey hey dtantsur :) good morning :) | 17:52 |
vdrok | good morning NobodyCam ! | 17:53 |
vdrok | night mgould|afk | 17:54 |
NobodyCam | morning vdrok :) | 17:54 |
vdrok | NobodyCam: heard you're back to upstream? | 17:54 |
NobodyCam | prob around 10 to 15 more days then !! :) | 17:54 |
vdrok | heh | 17:55 |
NobodyCam | :) | 17:55 |
*** trown is now known as trown|lunch | 17:56 | |
*** khamtamtun has quit IRC | 17:57 | |
*** khamtamtun has joined #openstack-ironic | 17:57 | |
*** bcornec1 has joined #openstack-ironic | 17:58 | |
*** stendulker has quit IRC | 17:59 | |
*** causten has quit IRC | 17:59 | |
dtantsur | wow, my patch makes random unrelated variables in project be set to None | 18:00 |
dtantsur | Oo | 18:00 |
dtantsur | oO | 18:00 |
dtantsur | O_O | 18:00 |
*** derekh has quit IRC | 18:01 | |
sambetts | night all | 18:01 |
*** sambetts is now known as sambetts|afk | 18:01 | |
* NobodyCam adjusts his meeting alarm | 18:01 | |
NobodyCam | night sambetts|afk :) | 18:01 |
*** fragatin_ has joined #openstack-ironic | 18:01 | |
*** causten has joined #openstack-ironic | 18:02 | |
*** ralonsoh has quit IRC | 18:02 | |
*** rbudden has joined #openstack-ironic | 18:03 | |
*** fragatina has quit IRC | 18:05 | |
dtantsur | importutils.try_import('proliantutils.exception') returns None with my patch, wonderful... | 18:05 |
JayF | dtantsur: if you wanna link it up I'll take a look | 18:05 |
dtantsur | JayF, thanks, I'll post it as it is now | 18:06 |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] Support defining and loading hardware types https://review.openstack.org/336626 | 18:06 |
dtantsur | JayF, this one ^^^ | 18:06 |
* dtantsur calls it a day now | 18:06 | |
*** causten has quit IRC | 18:06 | |
openstackgerrit | Debayan Ray proposed openstack/ironic-specs: EFI iSCSI boot for ProLiant hardware https://review.openstack.org/207337 | 18:07 |
jroll | dtantsur: isn't that correct? | 18:07 |
jroll | if not importutils.try_import(...): raise exception | 18:07 |
*** khamtamtun has quit IRC | 18:07 | |
dtantsur | jroll, well, without my patch is does NOT return None and tests pass... so I suspect it's not correct | 18:07 |
* dtantsur is confused like hell | 18:07 | |
dtantsur | I'm probably missing something obvious, but right now it looks like black magic | 18:08 |
jroll | dtantsur: ah, weird, maybe we break the mocks | 18:08 |
jroll | (the mocks make it return not None) | 18:08 |
jroll | anyway, go have a beer :) | 18:08 |
dtantsur | yep :) | 18:08 |
*** khamtamtun has joined #openstack-ironic | 18:08 | |
*** dtantsur is now known as dtantsur|afk | 18:09 | |
*** [1]rpioso has quit IRC | 18:13 | |
*** rbrud has quit IRC | 18:13 | |
*** jkilpatr has quit IRC | 18:14 | |
*** _milan_ has joined #openstack-ironic | 18:14 | |
NobodyCam | night dtantsur|afk | 18:14 |
*** betherly has quit IRC | 18:14 | |
*** milan has quit IRC | 18:15 | |
*** _vsaienko has quit IRC | 18:16 | |
*** khamtamtun has quit IRC | 18:18 | |
openstackgerrit | Michael Turek proposed openstack/ironic-specs: Add spec for OpenBMC driver https://review.openstack.org/388218 | 18:20 |
*** jkilpatr has joined #openstack-ironic | 18:21 | |
*** causten has joined #openstack-ironic | 18:29 | |
*** chas_ has joined #openstack-ironic | 18:30 | |
*** chuckC_ has quit IRC | 18:33 | |
*** rcernin has joined #openstack-ironic | 18:33 | |
*** chas__ has joined #openstack-ironic | 18:34 | |
*** bdeschenes has joined #openstack-ironic | 18:34 | |
*** chas_ has quit IRC | 18:34 | |
*** chas__ has quit IRC | 18:38 | |
*** khamtamtun has joined #openstack-ironic | 18:40 | |
mariojv | has anyone successfully used the discovery feature of inspector for enrolling new nodes in ironic? http://docs.openstack.org/developer/ironic-inspector/usage.html#discovery | 18:49 |
mariojv | aNuposic is figuring out whether that feature works and is useful | 18:49 |
*** krtaylor has quit IRC | 18:49 | |
aNuposic | Hi Folks, I am trying to understand ironic inspector and specifically how it will discover the node that is not already enrolled with Ironic? Do I need to run "inspection start" with UUID of the new node and will that automatically enroll it as "node_not_found_hook" would be set to enroll | 18:49 |
*** fxpester has quit IRC | 18:50 | |
*** khamtamtun has quit IRC | 18:50 | |
jlvillal | mariojv: aNuposic: I believe aarefiev is the one who implemented that code. | 18:53 |
jlvillal | Not sure if he is still online as it is getting late where he is. | 18:53 |
xavierr | aNuposic: ironic inspector will run an IPA agent in the ramdisk image, then will discover properties from the hardware and update the node in Ironic accordingly | 18:55 |
xavierr | aNuposic: you will need a node in manageable state and then start inspection by doing 'inspection start' | 18:56 |
*** [1]rpioso has joined #openstack-ironic | 18:57 | |
*** _vsaienko has joined #openstack-ironic | 18:59 | |
aNuposic | xavierr: So after putting it into manageable state and running "inspection start" will it enroll the new unidentified node and set ipmi properties on node? | 18:59 |
*** trown|lunch is now known as trown | 18:59 | |
vdrok | good night everyone! | 19:00 |
aNuposic | xavierr: I am reading at this http://docs.openstack.org/developer/ironic-inspector/usage.html#discovery and it says 'node_not_found_hook' will enroll for us. I am not clear will this be done on running "inspection start" or any other command | 19:01 |
* xavierr reading | 19:02 | |
xavierr | aNuposic: I've never used inspection like that. by that I mean discover new hardware and create new nodes based on that | 19:05 |
*** _vsaienko has quit IRC | 19:05 | |
*** khamtamtun has joined #openstack-ironic | 19:06 | |
xavierr | aNuposic: my flow is: have a node missing properties info in manageable state -> run inspection -> node is updated with the new properties | 19:06 |
jroll | there's no UUID of the new node to run 'inspection start' with, in this case | 19:07 |
jroll | I think the idea is, you (or inspector?) run a dhcp server that has a catch-all pxe response that will boot IPA | 19:07 |
aNuposic | jroll: Yeah I agree, there will be no UUID. | 19:09 |
aNuposic | jroll: Let me try to do that. | 19:09 |
aNuposic | xavierr: I believe your flow is perfect if node had UUID | 19:10 |
jroll | aNuposic: have you seen http://docs.openstack.org/developer/ironic-inspector/install.html ? | 19:10 |
jroll | it has some things about configuring pxe and such | 19:10 |
xavierr | aNuposic: yeah, it is. sorry :/ | 19:10 |
jroll | and dnsmasq as well | 19:10 |
aNuposic | jroll: where do i get the pxe response, it it in some file? | 19:11 |
*** _vsaienko has joined #openstack-ironic | 19:11 | |
jroll | aNuposic: that link above seems to have a decent dhcp response and pxe config you can use | 19:11 |
*** Sukhdev has joined #openstack-ironic | 19:12 | |
aNuposic | jroll: I see. I am reading that. It do seem to have some dnsmasq and pxe configurations | 19:12 |
aNuposic | jroll: great | 19:12 |
jroll | :) | 19:13 |
xavierr | use ipxe is also an option | 19:13 |
*** fragatin_ has quit IRC | 19:15 | |
*** _milan_ has quit IRC | 19:15 | |
*** fragatina has joined #openstack-ironic | 19:15 | |
openstackgerrit | Merged openstack/python-ironicclient: OSC add capability to remove node/chassis_uuid https://review.openstack.org/391812 | 19:17 |
*** jkilpatr_ has joined #openstack-ironic | 19:18 | |
*** jkilpatr has quit IRC | 19:18 | |
*** milan has joined #openstack-ironic | 19:19 | |
*** krtaylor has joined #openstack-ironic | 19:21 | |
*** bfournie has left #openstack-ironic | 19:21 | |
*** dprince has quit IRC | 19:22 | |
*** rpioso has joined #openstack-ironic | 19:22 | |
*** [1]rpioso has quit IRC | 19:23 | |
*** pc_m has quit IRC | 19:23 | |
*** pc_m has joined #openstack-ironic | 19:24 | |
*** jkilpatr_ has quit IRC | 19:25 | |
*** khamtamtun has quit IRC | 19:26 | |
*** bfournie has joined #openstack-ironic | 19:26 | |
*** khamtamtun has joined #openstack-ironic | 19:27 | |
*** chas_ has joined #openstack-ironic | 19:28 | |
*** milan has quit IRC | 19:30 | |
*** _vsaienko has quit IRC | 19:30 | |
*** priteau has quit IRC | 19:30 | |
*** milan has joined #openstack-ironic | 19:34 | |
*** _vsaienko has joined #openstack-ironic | 19:34 | |
*** rbudden has quit IRC | 19:34 | |
*** jkilpatr_ has joined #openstack-ironic | 19:39 | |
*** priteau has joined #openstack-ironic | 19:43 | |
*** sabeen has quit IRC | 19:45 | |
*** rbudden has joined #openstack-ironic | 19:58 | |
openstackgerrit | Annie Lezil proposed openstack/ironic-python-agent: Trying to create a package for biosdevname https://review.openstack.org/396406 | 19:59 |
*** [1]rpioso has joined #openstack-ironic | 20:05 | |
*** priteau has quit IRC | 20:07 | |
*** rpioso has quit IRC | 20:09 | |
*** e0ne has joined #openstack-ironic | 20:09 | |
*** ijw has quit IRC | 20:18 | |
aNuposic | jroll: I was going over http://docs.openstack.org/developer/ironic-inspector/install.html but not clear on how pxe will also get the info about new node? Is it because the new node will boot at the same time? | 20:18 |
jroll | aNuposic: the pxe server doesn't need any info about the new node, that will cause everything on the network to pxe into the agent | 20:19 |
jroll | so you just power the node on and it ends up booting IPA and talking to inspector | 20:20 |
jroll | if that makes sense | 20:20 |
aNuposic | jroll: Hmm, okay. So, from my side i just need to boot a node. | 20:22 |
jroll | aNuposic: indeed | 20:22 |
jroll | aNuposic: in devstack, that's something like virsh start node-0 | 20:22 |
aNuposic | jroll: let me check that option. So basically, i will boot the node and then inspector will get to know that this node is not already enrolled and then it will automatically enroll it? as i will set "node_not_found_hook=enroll" | 20:24 |
*** _vsaienko has quit IRC | 20:24 | |
aNuposic | am i going right? | 20:24 |
jroll | aNuposic: yep, inspector uses mac addresses and uuids to decide if the node is already in ironic's db, if it can't find it, that hook will cause inspector to enroll it with ironic | 20:25 |
aNuposic | jroll: Ah makes sense to me now. | 20:26 |
*** khamtamtun has quit IRC | 20:27 | |
*** bfournie has left #openstack-ironic | 20:28 | |
*** khamtamtun has joined #openstack-ironic | 20:31 | |
*** rcernin has quit IRC | 20:35 | |
*** e0ne has quit IRC | 20:40 | |
*** dprince has joined #openstack-ironic | 20:41 | |
*** godp1301_ has joined #openstack-ironic | 20:44 | |
*** pcaruana has quit IRC | 20:45 | |
*** bfournie has joined #openstack-ironic | 20:46 | |
*** godp1301 has quit IRC | 20:47 | |
*** priteau has joined #openstack-ironic | 20:47 | |
*** flaviodsr has joined #openstack-ironic | 20:47 | |
*** khamtamtun has quit IRC | 20:51 | |
*** causten has quit IRC | 20:52 | |
*** causten has joined #openstack-ironic | 20:53 | |
*** priteau has quit IRC | 20:53 | |
*** khamtamtun has joined #openstack-ironic | 20:53 | |
*** _vsaienko has joined #openstack-ironic | 20:53 | |
*** khamtamtun has quit IRC | 20:55 | |
*** ijw has joined #openstack-ironic | 20:58 | |
*** penick has joined #openstack-ironic | 20:58 | |
*** rcernin has joined #openstack-ironic | 21:00 | |
*** causten has quit IRC | 21:01 | |
*** causten has joined #openstack-ironic | 21:02 | |
*** MattMan has quit IRC | 21:02 | |
*** ijw has quit IRC | 21:02 | |
*** MattMan has joined #openstack-ironic | 21:03 | |
*** rbudden has quit IRC | 21:05 | |
rloo | jroll: what do you think about me emailing this week's priorities along with the subteam reports? (or not, or as a separate email)? | 21:09 |
jroll | rloo: seems fine to me :) | 21:10 |
rloo | jroll: ok. | 21:10 |
*** sabeen has joined #openstack-ironic | 21:12 | |
*** godp1301_ has quit IRC | 21:14 | |
*** rbudden has joined #openstack-ironic | 21:25 | |
*** rama_y has quit IRC | 21:25 | |
rloo | jroll, devananda: did we decide to cancel the weekly ironic-api meetings? | 21:27 |
*** jkilpatr_ has quit IRC | 21:27 | |
jroll | rloo: no, we never decided that, but I'm inclined to | 21:27 |
rloo | jroll: me too. guess we should hear from devananda then? i'm updating our meetings wiki now, which reminded me to ask :) | 21:28 |
jroll | rloo: indeed | 21:28 |
rloo | jroll: if we don't hear from devananda by tomorrow's meeting time, let's cancel. | 21:28 |
jroll | rloo: sounds good | 21:28 |
*** rbudden has quit IRC | 21:29 | |
*** rama_y has joined #openstack-ironic | 21:30 | |
*** ijw has joined #openstack-ironic | 21:38 | |
*** Goneri has quit IRC | 21:38 | |
*** chuckC_ has joined #openstack-ironic | 21:39 | |
milan | guys, I'm wondering: having deployed a full-disk image via ironic, how do I change console kernel CLI parameter on the instance? | 21:49 |
jroll | milan: change grub (or whatever bootloader your image uses) config :) | 21:50 |
milan | jroll, yeah that's what I was afraid of :D | 21:51 |
milan | thx | 21:51 |
jroll | heh, np | 21:51 |
milan | btw ipmi tool SOL via socat show-off: http://paste.openstack.org/show/589186/ | 21:51 |
milan | IPA worked just OK :) | 21:51 |
*** trown is now known as trown|outtypewww | 21:53 | |
jroll | milan: lol, nice | 21:55 |
milan | :) it's even interactive (telnet) | 21:55 |
kvcobb | mgould: Wanted to drop a quick update to you and jayF regarding something you helped me with last week (attempting to get bare metal hardware working with Devstack), the issue I was having was not devstack related though - the issue was caused by my cloud sitting behind a proxy. When we left off we figured out that machines on my small private network which accessed the web via uplink from company VPN were not able to talk to | 21:57 |
kvcobb | freshly booted nodes on the private network after PXE boot. So I've got everything working now (or at least I'm past these connectivity issues). It turned out once my conducter had been configured for web access via my intranet proxy, all the machines on the private network were invisible to it because it was looking for them out on the main intranet network. My solution was to setup Squid on a machine on my private network and I | 21:57 |
kvcobb | plugged in the following settings to squid.conf: | 21:57 |
kvcobb | cache_peer <intranet_proxy_url_here:port> parent 911 0 default no-query | 21:57 |
kvcobb | acl local-intranet dst 192.168.1.0/24 | 21:57 |
openstackgerrit | Michael Turek proposed openstack/ironic-specs: Add spec for OpenBMC driver https://review.openstack.org/388218 | 21:57 |
kvcobb | never_direct allow all | 21:58 |
kvcobb | always_direct allow local-intranet | 21:58 |
kvcobb | 21:58 | |
kvcobb | This allowed machines on my private network access to the web, but for local addresses it now looks locally rather than to the proxy. This solved the problem, now communication between conductor and node is great, now I just need to figure out how to properly configure my wholedisk image (instance starts successfully but after it gets cleaned and the image is deployed it won't boot up. And on to this next problem... :) | 21:58 |
JayF | kvcobb: congrats on fixing it, I thought it was a proxy. FWIW for future reference, if you're going to paste in more than a couple of lines please use a pastebin :) | 21:58 |
kvcobb | will do, apologies :) | 21:58 |
*** aNuposic has quit IRC | 21:59 | |
jroll | I'm off for the night, see y'all tomorrow | 21:59 |
*** _vsaienko has quit IRC | 21:59 | |
*** causten has quit IRC | 22:01 | |
*** e0ne has joined #openstack-ironic | 22:04 | |
*** e0ne has quit IRC | 22:10 | |
milan | good night Ironic! /me gone :) | 22:14 |
*** milan has quit IRC | 22:14 | |
*** krtaylor has quit IRC | 22:20 | |
*** priteau has joined #openstack-ironic | 22:30 | |
*** jcoufal has quit IRC | 22:33 | |
*** jheroux has quit IRC | 22:38 | |
*** priteau has quit IRC | 22:40 | |
*** baoli has quit IRC | 22:40 | |
tonyb | jroll: Can you vote on this https://review.openstack.org/#/c/395517/ ? | 22:41 |
patchbot | patch 395517 - openstack-infra/irc-meetings - Remove unused Ironic v2 API meeting | 22:41 |
*** jkilpatr has joined #openstack-ironic | 22:46 | |
*** ccamacho has quit IRC | 22:53 | |
*** rbudden has joined #openstack-ironic | 22:56 | |
*** penick has quit IRC | 23:00 | |
*** penick has joined #openstack-ironic | 23:03 | |
*** lindycoder has quit IRC | 23:04 | |
*** dprince has quit IRC | 23:06 | |
*** krtaylor has joined #openstack-ironic | 23:10 | |
*** chas_ has quit IRC | 23:17 | |
*** chas_ has joined #openstack-ironic | 23:18 | |
*** chas_ has quit IRC | 23:20 | |
*** chas_ has joined #openstack-ironic | 23:20 | |
*** absubram has quit IRC | 23:21 | |
*** ijw_ has joined #openstack-ironic | 23:22 | |
*** rbudden has quit IRC | 23:22 | |
*** ijw has quit IRC | 23:24 | |
*** chas_ has quit IRC | 23:25 | |
*** chuckC_ has quit IRC | 23:28 | |
*** cdearborn has quit IRC | 23:28 | |
*** rbudden has joined #openstack-ironic | 23:28 | |
jlvillal | rloo: Was/is there a spec for multi-tenant networking? | 23:31 |
jlvillal | rloo: I looked at the specs and it didn't jump out at me. | 23:31 |
*** hoangcx has joined #openstack-ironic | 23:32 | |
*** hoangcx has quit IRC | 23:33 | |
JayF | jlvillal: is it just me, or does https://specs.openstack.org have no stylesheet? | 23:36 |
jlvillal | JayF: I'm not sure. I'm looking at: https://specs.openstack.org/openstack/ironic-specs/ | 23:37 |
JayF | jlvillal: I'm looking for the sspec you're asking about | 23:37 |
JayF | but I just noticed that and was curious if it was local-only | 23:37 |
jlvillal | JayF: Thanks for looking. I was looking there and searching for 'network', 'tenant', etc... | 23:38 |
jlvillal | With no luck and getting confused as I would think for sure it was a spec. | 23:38 |
JayF | https://specs.openstack.org/openstack/ironic-specs/specs/6.1/network-provider.html | 23:38 |
JayF | there you go | 23:38 |
*** penick has quit IRC | 23:38 | |
jlvillal | JayF: That is an awful name ;) | 23:38 |
jlvillal | JayF: It would be nice if the specs were that name we used to talk about the feature. I think 'Multi-tenant networking' | 23:39 |
jlvillal | JayF: Thanks!!!! | 23:39 |
*** diablo_rojo has quit IRC | 23:40 | |
JayF | eh, naming things is hard | 23:41 |
*** penick has joined #openstack-ironic | 23:41 | |
JayF | I'd rather us save our bikeshedding effort for where it matters; in the code :P | 23:41 |
JayF | jroll: I officially volunteer to be on stable core team. Basically just tired of seeing a lot of unapproved requirements bumps and docs :D | 23:43 |
*** [1]rpioso has quit IRC | 23:48 | |
*** penick has quit IRC | 23:48 | |
openstackgerrit | John L. Villalovos proposed openstack/ironic-specs: Rename 'Pluggable network providers' to 'Multi-tenant networking' https://review.openstack.org/397442 | 23:50 |
openstackgerrit | John L. Villalovos proposed openstack/ironic-specs: Rename 'Pluggable network providers' to 'Multi-tenant networking' https://review.openstack.org/397442 | 23:51 |
*** aNuposic has joined #openstack-ironic | 23:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!