| TheJulia | dtantsur|afk: congratulations | 00:11 |
|---|---|---|
| *** milan has joined #openstack-ironic | 00:12 | |
| *** _milan_ has quit IRC | 00:12 | |
| *** zhangfei has joined #openstack-ironic | 00:15 | |
| *** charcol has joined #openstack-ironic | 00:18 | |
| *** crushil has joined #openstack-ironic | 00:25 | |
| TheJulia | i really think we need that devstack back port, but I don't see that happening tonight | 00:32 |
| *** yangyapeng has joined #openstack-ironic | 00:46 | |
| *** yangyapeng has quit IRC | 00:51 | |
| *** dyasny has joined #openstack-ironic | 00:53 | |
| *** litao__ has joined #openstack-ironic | 00:55 | |
| *** phuongnh has joined #openstack-ironic | 01:00 | |
| wong | hi julia,may I ask you a question about drac-driver? | 01:11 |
| *** crushil has quit IRC | 01:15 | |
| *** _milan_ has joined #openstack-ironic | 01:15 | |
| *** milan has quit IRC | 01:16 | |
| *** Sukhdev has quit IRC | 01:18 | |
| *** yangyapeng has joined #openstack-ironic | 01:33 | |
| *** r-daneel_ has joined #openstack-ironic | 01:53 | |
| *** mat128 has joined #openstack-ironic | 01:53 | |
| *** r-daneel has quit IRC | 01:53 | |
| *** r-daneel_ is now known as r-daneel | 01:53 | |
| *** dyasny has quit IRC | 01:58 | |
| *** gcb has joined #openstack-ironic | 02:07 | |
| *** crushil has joined #openstack-ironic | 02:10 | |
| *** r-daneel has quit IRC | 02:17 | |
| *** r-daneel has joined #openstack-ironic | 02:17 | |
| *** mkovacik__ has joined #openstack-ironic | 02:18 | |
| *** _milan_ has quit IRC | 02:18 | |
| *** rloo_afk has quit IRC | 02:22 | |
| *** zhangfei has quit IRC | 02:31 | |
| *** crushil has quit IRC | 02:38 | |
| *** crushil has joined #openstack-ironic | 02:39 | |
| *** zhangfei has joined #openstack-ironic | 02:49 | |
| *** mat128 has quit IRC | 03:03 | |
| *** Sukhdev has joined #openstack-ironic | 03:18 | |
| *** milan has joined #openstack-ironic | 03:21 | |
| *** mkovacik__ has quit IRC | 03:22 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic master: Documetation for 'snmp' hardware type https://review.openstack.org/493468 | 03:56 |
| *** Sukhdev has quit IRC | 03:57 | |
| *** links has joined #openstack-ironic | 04:01 | |
| *** vishwana_ has quit IRC | 04:04 | |
| *** vishwanathj has joined #openstack-ironic | 04:04 | |
| *** Nisha_Agarwal has joined #openstack-ironic | 04:07 | |
| *** udesale has joined #openstack-ironic | 04:15 | |
| *** masber has joined #openstack-ironic | 04:20 | |
| *** _milan_ has joined #openstack-ironic | 04:23 | |
| *** milan has quit IRC | 04:24 | |
| *** yangyapeng has quit IRC | 04:25 | |
| *** yangyapeng has joined #openstack-ironic | 04:27 | |
| *** openstack has quit IRC | 04:42 | |
| *** openstack has joined #openstack-ironic | 04:45 | |
| *** yangyapeng has joined #openstack-ironic | 04:45 | |
| *** openstackstatus has joined #openstack-ironic | 04:45 | |
| *** oanson has joined #openstack-ironic | 04:45 | |
| *** dansmith has joined #openstack-ironic | 04:45 | |
| *** aNuposic has joined #openstack-ironic | 04:45 | |
| *** honza is now known as Guest28838 | 04:45 | |
| *** fbouliane has joined #openstack-ironic | 04:45 | |
| *** JayF has joined #openstack-ironic | 04:45 | |
| *** csmart has quit IRC | 04:45 | |
| *** retr0h has quit IRC | 04:45 | |
| *** phuongnh has joined #openstack-ironic | 04:45 | |
| *** amotoki has joined #openstack-ironic | 04:45 | |
| *** mtreinish has joined #openstack-ironic | 04:45 | |
| *** bradjones has joined #openstack-ironic | 04:45 | |
| *** ChanServ sets mode: +v openstackstatus | 04:45 | |
| *** leyal has joined #openstack-ironic | 04:45 | |
| *** sai has joined #openstack-ironic | 04:45 | |
| *** Dinesh_Bhor has joined #openstack-ironic | 04:45 | |
| *** pc_m has joined #openstack-ironic | 04:45 | |
| *** JayF has quit IRC | 04:46 | |
| *** JayF has joined #openstack-ironic | 04:46 | |
| *** patriciadomin has joined #openstack-ironic | 04:46 | |
| *** dansmith is now known as Guest53762 | 04:46 | |
| *** antonym has joined #openstack-ironic | 04:46 | |
| *** EmilienM has quit IRC | 04:46 | |
| *** EmilienM has joined #openstack-ironic | 04:46 | |
| *** zigo has joined #openstack-ironic | 04:46 | |
| *** antonym has quit IRC | 04:46 | |
| *** antonym has joined #openstack-ironic | 04:46 | |
| *** mmedvede has joined #openstack-ironic | 04:46 | |
| *** mkrai has quit IRC | 04:46 | |
| *** aignatov has joined #openstack-ironic | 04:46 | |
| *** leifmadsen has joined #openstack-ironic | 04:46 | |
| *** weshay has joined #openstack-ironic | 04:46 | |
| *** bradjones is now known as Guest45420 | 04:46 | |
| *** devananda has joined #openstack-ironic | 04:46 | |
| *** SpamapS has joined #openstack-ironic | 04:46 | |
| *** quakers has joined #openstack-ironic | 04:46 | |
| *** bcornec has joined #openstack-ironic | 04:46 | |
| *** lihi has joined #openstack-ironic | 04:46 | |
| *** ptx0 has joined #openstack-ironic | 04:46 | |
| *** dtantsur has joined #openstack-ironic | 04:47 | |
| *** ptx0 has quit IRC | 04:47 | |
| *** ptx0 has joined #openstack-ironic | 04:47 | |
| *** yuanying has joined #openstack-ironic | 04:47 | |
| *** leifmadsen has quit IRC | 04:47 | |
| *** leifmadsen has joined #openstack-ironic | 04:47 | |
| *** zzzeek has joined #openstack-ironic | 04:47 | |
| *** rybridges1 has joined #openstack-ironic | 04:47 | |
| *** rfolco has joined #openstack-ironic | 04:47 | |
| *** kevinbenton has joined #openstack-ironic | 04:47 | |
| *** hwoarang has joined #openstack-ironic | 04:48 | |
| *** hwoarang has quit IRC | 04:48 | |
| *** hwoarang has joined #openstack-ironic | 04:48 | |
| *** aolwas has joined #openstack-ironic | 04:48 | |
| *** radez has joined #openstack-ironic | 04:48 | |
| *** Sukhdev has joined #openstack-ironic | 04:48 | |
| *** edleafe has joined #openstack-ironic | 04:48 | |
| *** vsaienk0 has joined #openstack-ironic | 04:48 | |
| *** lucasagomes has joined #openstack-ironic | 04:49 | |
| *** jistr has joined #openstack-ironic | 04:49 | |
| *** dims has joined #openstack-ironic | 04:49 | |
| *** ianw has joined #openstack-ironic | 04:49 | |
| *** mkrai has joined #openstack-ironic | 04:49 | |
| *** jhesketh has joined #openstack-ironic | 04:49 | |
| *** jaosorior has joined #openstack-ironic | 04:49 | |
| *** dims is now known as Guest58397 | 04:49 | |
| *** adreznec has joined #openstack-ironic | 04:49 | |
| *** jtomasek has joined #openstack-ironic | 04:50 | |
| *** vishwanathj has joined #openstack-ironic | 04:50 | |
| *** trown has joined #openstack-ironic | 04:50 | |
| *** jcook has joined #openstack-ironic | 04:50 | |
| *** marlinc has joined #openstack-ironic | 04:50 | |
| *** karthiks has joined #openstack-ironic | 04:51 | |
| *** Bico_Fino has joined #openstack-ironic | 04:51 | |
| *** bnemec has joined #openstack-ironic | 04:51 | |
| *** srart has joined #openstack-ironic | 04:53 | |
| *** adisky__ has joined #openstack-ironic | 04:54 | |
| *** cinerama has joined #openstack-ironic | 04:54 | |
| *** raginbajin has joined #openstack-ironic | 04:55 | |
| *** ElCoyote_ has joined #openstack-ironic | 04:55 | |
| *** aparnav has joined #openstack-ironic | 05:00 | |
| *** stendulker has joined #openstack-ironic | 05:10 | |
| *** openstackgerrit has joined #openstack-ironic | 05:14 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic master: Documentation for 'snmp' hardware type https://review.openstack.org/493468 | 05:14 |
| *** hshiina has joined #openstack-ironic | 05:17 | |
| *** masuberu has joined #openstack-ironic | 05:18 | |
| *** csmart has joined #openstack-ironic | 05:20 | |
| *** retr0h has joined #openstack-ironic | 05:20 | |
| *** marios has joined #openstack-ironic | 05:20 | |
| *** masber has quit IRC | 05:21 | |
| *** e0ne has joined #openstack-ironic | 05:23 | |
| *** Nisha_Agarwal has joined #openstack-ironic | 05:24 | |
| *** masber has joined #openstack-ironic | 05:25 | |
| *** litao__ has quit IRC | 05:26 | |
| *** milan has quit IRC | 05:27 | |
| *** masuberu has quit IRC | 05:28 | |
| *** milan has joined #openstack-ironic | 05:28 | |
| *** alineb has joined #openstack-ironic | 05:31 | |
| *** jaosorior has quit IRC | 05:32 | |
| *** jaosorior has joined #openstack-ironic | 05:33 | |
| *** udesale has joined #openstack-ironic | 05:34 | |
| *** Nisha_Agarwal has quit IRC | 05:41 | |
| *** e0ne has quit IRC | 05:41 | |
| *** udesale__ has joined #openstack-ironic | 05:43 | |
| *** gcb has joined #openstack-ironic | 05:45 | |
| *** udesale has quit IRC | 05:45 | |
| *** Guest28838 is now known as honza | 05:50 | |
| *** Sukhdev has quit IRC | 05:53 | |
| *** rcernin has joined #openstack-ironic | 05:57 | |
| *** mjura has joined #openstack-ironic | 05:57 | |
| *** e0ne has joined #openstack-ironic | 06:06 | |
| *** mvpnitesh has joined #openstack-ironic | 06:09 | |
| *** ianychoi has quit IRC | 06:23 | |
| *** ianychoi has joined #openstack-ironic | 06:26 | |
| *** milan has quit IRC | 06:30 | |
| *** rbartal_ has joined #openstack-ironic | 06:34 | |
| *** milan has joined #openstack-ironic | 06:34 | |
| *** milan has quit IRC | 06:38 | |
| *** udesale has joined #openstack-ironic | 06:40 | |
| *** udesale__ has quit IRC | 06:42 | |
| *** pcaruana has joined #openstack-ironic | 06:42 | |
| *** jaypipes has joined #openstack-ironic | 06:44 | |
| *** tesseract has joined #openstack-ironic | 06:47 | |
| *** wong has joined #openstack-ironic | 06:50 | |
| wong | hi | 06:51 |
| wong | how to remove driver_internal_info? | 06:51 |
| *** tuanluong has joined #openstack-ironic | 06:53 | |
| *** hoonetorg has joined #openstack-ironic | 06:54 | |
| *** hw_wutianwei has joined #openstack-ironic | 06:58 | |
| *** mgoddard has joined #openstack-ironic | 07:05 | |
| *** Dinesh_Bhor has quit IRC | 07:11 | |
| *** ccamacho has joined #openstack-ironic | 07:12 | |
| *** Dinesh_Bhor has joined #openstack-ironic | 07:18 | |
| *** tuanluong has quit IRC | 07:20 | |
| pas-ha | wong: I'd presume manage/provide cycle should do it. but why do you need it? | 07:20 |
| *** mvpnitesh has quit IRC | 07:26 | |
| *** trungnv has joined #openstack-ironic | 07:38 | |
| wong | there is a mistake after I make the status from manageable to availbale | 07:39 |
| *** mrch has joined #openstack-ironic | 07:39 | |
| wong | I use CLI"clean" to configure RAID,for a long time,the raid configration was success,and the node status was in manageable | 07:42 |
| wong | then I use "provide" to make the node from manageable to availble | 07:44 |
| wong | it was in "clean failed" | 07:44 |
| wong | error is " in node.driver_internal_info.get('clean_steps', []) TypeError: 'NoneType' object is not iterable" | 07:50 |
| *** mikal has joined #openstack-ironic | 07:51 | |
| wong | utill now, I restart the clean-step,and It's OK... | 07:52 |
| wong | I'm afraid it will happen again | 07:52 |
| *** honza has quit IRC | 07:54 | |
| *** honza has joined #openstack-ironic | 07:54 | |
| *** honza is now known as Guest11829 | 07:55 | |
| *** zzzeek has quit IRC | 07:56 | |
| *** tesseract has quit IRC | 07:58 | |
| *** zzzeek has joined #openstack-ironic | 07:59 | |
| *** athomas has quit IRC | 07:59 | |
| *** derekh has joined #openstack-ironic | 08:04 | |
| *** athomas has joined #openstack-ironic | 08:06 | |
| *** priteau has joined #openstack-ironic | 08:12 | |
| *** Nisha_Agarwal has joined #openstack-ironic | 08:14 | |
| *** openstackgerrit has quit IRC | 08:17 | |
| *** priteau has quit IRC | 08:18 | |
| *** priteau has joined #openstack-ironic | 08:19 | |
| *** priteau has quit IRC | 08:23 | |
| *** yuriyz has joined #openstack-ironic | 08:26 | |
| Nisha_Agarwal | morning ironic!!! | 08:27 |
| *** openstackgerrit has joined #openstack-ironic | 08:27 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic master: Add the new capabilities to the iLO InspectInterface https://review.openstack.org/422902 | 08:27 |
| *** karimb has joined #openstack-ironic | 08:28 | |
| *** e0ne has quit IRC | 08:28 | |
| *** priteau has joined #openstack-ironic | 08:35 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic master: Temporary set pin_release_version to ocata https://review.openstack.org/494435 | 08:36 |
| *** logan- has quit IRC | 08:37 | |
| *** kromanenko has joined #openstack-ironic | 08:37 | |
| *** aparnav has quit IRC | 08:39 | |
| *** aparnav has joined #openstack-ironic | 08:40 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic master: Add the new capabilities to the iLO InspectInterface https://review.openstack.org/422902 | 08:40 |
| aparnav | hey vdrok, I have addressed the comments for the patch https://review.openstack.org/#/c/422572/. Take a look when you find some time. | 08:41 |
| *** priteau has quit IRC | 08:41 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic master: Add the new capabilities to the iLO InspectInterface https://review.openstack.org/422902 | 08:44 |
| openstackgerrit | yolanda.robla proposed openstack/bifrost master: DNM: Testing diskimage-builder failures https://review.openstack.org/493033 | 08:44 |
| openstackgerrit | yolanda.robla proposed openstack/bifrost master: DNM: Testing diskimage-builder failures https://review.openstack.org/493033 | 08:52 |
| *** fxpester has joined #openstack-ironic | 09:07 | |
| pas-ha | morning ironic ^) | 09:07 |
| *** Guest11829 is now known as honza | 09:09 | |
| *** priteau has joined #openstack-ironic | 09:09 | |
| *** amarao has joined #openstack-ironic | 09:16 | |
| *** afazekas is now known as afazekas|sick | 09:19 | |
| *** etingof has joined #openstack-ironic | 09:32 | |
| *** Nisha_Agarwal has quit IRC | 09:33 | |
| *** Nisha_Agarwal has joined #openstack-ironic | 09:34 | |
| *** yangyapeng has quit IRC | 09:39 | |
| Nisha_Agarwal | pas-ha, morning | 09:46 |
| Nisha_Agarwal | dtantsur, hi | 09:46 |
| *** sambetts_ is now known as sambetts | 09:47 | |
| *** afazekas|sick is now known as afazekas | 09:47 | |
| Nisha_Agarwal | sambetts, morning | 09:47 |
| sambetts | Morning Nisha_Agarwal | 09:47 |
| Nisha_Agarwal | o/ | 09:50 |
| vdrok | morning sambetts Nisha_Agarwal aparnav | 10:01 |
| *** ralonsoh has joined #openstack-ironic | 10:08 | |
| *** retr0h has quit IRC | 10:11 | |
| *** mjura has quit IRC | 10:12 | |
| *** mjura has joined #openstack-ironic | 10:14 | |
| sambetts | o/ vdrok | 10:14 |
| *** udesale has quit IRC | 10:18 | |
| sambetts | vdrok: so you want to dig into the CI failures again this morning? | 10:20 |
| vdrok | sambetts: the one from yesterday? | 10:20 |
| vdrok | we actually have a new one as well :) | 10:20 |
| sambetts | oh no... | 10:20 |
| vdrok | https://review.openstack.org/494435 | 10:21 |
| vdrok | but again, I'm still not able to find anyone to merge the devstack patch in ocata | 10:21 |
| vdrok | so multinode is still broken mostly | 10:21 |
| karimb | hello folks, when trying to activate my own driver, i get this error "ERROR oslo_service.service DriverLoadError: Driver, hardware type or interface staging-ovirt could not be loaded. Reason: 'module' object has no attribute 'POWER_DOWN'." | 10:23 |
| karimb | any hints ? | 10:23 |
| vdrok | karimb: seems like the driver issue, it tries to import POWER_DOWN on load | 10:24 |
| vdrok | and does not find it | 10:24 |
| vdrok | is it on the ovirt patch to staging-drivers? | 10:24 |
| karimb | oh right | 10:24 |
| karimb | i can see that now, my bad | 10:24 |
| karimb | vdrok thanks | 10:25 |
| *** retr0h has joined #openstack-ironic | 10:25 | |
| vdrok | yeah karimb we don't have a POWER_DOWN state | 10:25 |
| sambetts | vdrok: I'm unsure about your patch, I think we should be doing it by pinning in the job https://github.com/openstack-infra/devstack-gate/blob/master/devstack-vm-gate-wrap.sh#L263 | 10:25 |
| karimb | right, it s power off, i m drunk i guess | 10:25 |
| vdrok | sambetts: hrm, that might work as well | 10:26 |
| vdrok | sambetts: or, you mean somewhere in project config? | 10:26 |
| *** e0ne has joined #openstack-ironic | 10:27 | |
| sambetts | vdrok: I mean in project-config | 10:27 |
| vdrok | it seems like that one is actually the branch containing the tested patch | 10:27 |
| vdrok | the branch name for selecting grenade branches | 10:27 |
| vdrok | grenade is used from master IIUC | 10:28 |
| sambetts | we want OVERRIDE_ZUUL_BRANCH=stable/ocata in our grenade jos | 10:28 |
| sambetts | jobs* | 10:28 |
| sambetts | and then that becomes GREANDE_BASE_BRACNH | 10:28 |
| vdrok | hrm, but we don't have per-release job definitions in project-config | 10:29 |
| sambetts | errr...we can do https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/ironic.yaml#L189 | 10:30 |
| sambetts | I guess | 10:30 |
| sambetts | le sigh... | 10:30 |
| vdrok | yeah, that does not look pretty :) | 10:30 |
| openstackgerrit | Hironori Shiina proposed openstack/ironic master: iRMC: Follow-up: volume boot for virtual media boot interface https://review.openstack.org/494471 | 10:31 |
| sambetts | soo annoying that we can't set GRENADE_BASE_BRANCH directly | 10:32 |
| sambetts | ... | 10:32 |
| Nisha_Agarwal | hey vdrok | 10:39 |
| vdrok | sambetts: so, multinode has passed on that patch | 10:40 |
| Nisha_Agarwal | vdrok please could u bless https://review.openstack.org/422902 also | 10:40 |
| vdrok | Nisha_Agarwal: I'd prefer to fix the gate first | 10:40 |
| vdrok | sambetts: I'll check what it uses as base release | 10:40 |
| Nisha_Agarwal | vdrok, ok :) | 10:41 |
| vdrok | if it's still ocata, then that patch is the fastest way to go | 10:41 |
| sambetts | vdrok: I'm pretty sure it'll use master as the base release, but I think all that change is doing is pinning master ironic to ocata compatibilty | 10:41 |
| sambetts | :/ | 10:41 |
| vdrok | sambetts: well, yes, that's the intent. we need to set the pin after upgrade, and it needs to be ocata | 10:42 |
| vdrok | because we are upgrading from ocata to master still | 10:42 |
| sambetts | I think that job is now upgrading from master to master | 10:42 |
| sambetts | because base_devstack == stable/pike which doesn't exist | 10:43 |
| sambetts | and so devstack-gate falls back to master if the branch doesn't exist | 10:43 |
| vdrok | sambetts: ah, hrm | 10:43 |
| vdrok | lemme check | 10:43 |
| *** jkilpatr has quit IRC | 10:44 | |
| vdrok | sambetts: indeed | 10:46 |
| vdrok | sambetts: so wdyt about how to proceed? | 10:46 |
| vdrok | we could just cut pike with what we have | 10:47 |
| vdrok | and backport things afterwardfs | 10:47 |
| *** hshiina has quit IRC | 10:48 | |
| sambetts | even that doesn't work exactly as we want because grenade on master will be testing stable/pike -> master (still stable/pike just not in name) and then the stable/pike grenade job will actually do the stable/ocata -> stable/pike testing | 10:48 |
| *** phuongnh has quit IRC | 10:48 | |
| vdrok | sambetts: well, why is this not what we want? :) we'll be backporting things to pike, and pike's grenade doing upgrade from ocata to pike will be testing it | 10:49 |
| sambetts | just because we'll be backporting every change... essentially forcing us into the model that the other OpenStack projects follow where they cut a release branch release candidate and then backport for the whole of feature freeze into that branch | 10:51 |
| vdrok | yeah | 10:51 |
| vdrok | we have this same problem every release | 10:51 |
| sambetts | it just ends up making every commit too and we must not merge anything into master that doesn't get backported | 10:52 |
| sambetts | s/too/two | 10:52 |
| vdrok | previously we didnot have the rolling upgrade, so it just continued to work | 10:52 |
| * vdrok has to leave for a couple of hours | 10:53 | |
| vdrok | will continue when I get back | 10:54 |
| sambetts | no problem :) | 10:54 |
| vdrok | sambetts: if you happen to know someone from devstack-core, would be great if you could get the approve on https://review.openstack.org/#/q/I0874fe3b3628cb3e662ee01f24c4599247fdc82d,n,z :) | 10:54 |
| karimb | vdrok now i get DriverNotFoundInEntrypoint: Could not find the following items in the 'ironic.hardware.types' entrypoint: staging-ovirt. | 10:54 |
| karimb | is there a way i can proceed without using source ironic ? | 10:55 |
| vdrok | karimb: hrm, your entrypoint name is pxe_ovirt | 10:55 |
| vdrok | https://review.openstack.org/#/c/301226/29/setup.cfg | 10:56 |
| *** makowals has quit IRC | 10:58 | |
| karimb | vdrok yeah, but that would be the driver | 10:58 |
| karimb | now, i made it supposedly into a hardware type | 10:59 |
| karimb | so the line in this setup.cfg shouldnt event be there, i believe | 10:59 |
| karimb | instead i m supposed to edit entry points for ironic, which i dont want cause that would mean building entire ironic from source | 10:59 |
| karimb | vdrok does that make sense? | 11:00 |
| sambetts | karimb: in the ironic staging setup.cfg, add the ironic.hardware.types header and add your hardware type | 11:01 |
| karimb | sambetts but i can only do that from source ironic | 11:01 |
| karimb | and i m using rdo with rpms ironic... | 11:01 |
| sambetts | karimb: entry point namespaces can be defined by any python package, ironic itself defines the ones in its tree, but ironic-staging or whereever your code is can also define entries in the ironic.hardware.types namespace | 11:02 |
| *** zhangfei has quit IRC | 11:02 | |
| karimb | right then | 11:03 |
| sambetts | karimb: the same way as ironic-staging defines drivers in setup.cfg and ironic itself defines drivers in setup.cfg | 11:03 |
| pas-ha | karimb: take a look at https://review.openstack.org/#/c/466000/ | 11:03 |
| TheJulia | good morning everyone | 11:03 |
| pas-ha | karimb: there I am adding hw types and driver interfaces for all current drivers in staging | 11:03 |
| pas-ha | you could also use a script I introduce in that patch to list available entrypoints per type/package etc | 11:04 |
| *** fellypefca has joined #openstack-ironic | 11:05 | |
| karimb | pas-ha so actually turns out i have allready created the entrypoints in setup.cfg allready | 11:05 |
| karimb | in the ironic_staging_drivers setup.cfg | 11:05 |
| pas-ha | now you have to pip install it to update the entrypoints and restart ironic conductor | 11:06 |
| fellypefca | Hi everyone, good morning! :) | 11:06 |
| * pas-ha karimb: the error from the last job means there's something wrong with your module too | 11:07 | |
| pas-ha | http://logs.openstack.org/26/301226/29/check/gate-ironic-staging-drivers-dsvm-ubuntu-xenial/9845bf5/logs/screen-ir-cond.txt.gz#_Aug_17_10_24_56_179008 | 11:07 |
| pas-ha | 'module' object has no attribute 'POWER_DOWN'. | 11:07 |
| pas-ha | so ironic found that interface, but failed to load the python code due to error | 11:07 |
| pas-ha | (and crushed) | 11:08 |
| pas-ha | well, not crushed, did not start :) | 11:08 |
| *** hshiina has joined #openstack-ironic | 11:09 | |
| karimb | pas-ha yeah, i did fix that allready | 11:09 |
| sambetts | vdrok: https://review.openstack.org/#/c/494482/ | 11:10 |
| *** stendulker has quit IRC | 11:14 | |
| vsaienk0 | sambetts: we had the same situation when cut ocata please read http://lists.openstack.org/pipermail/openstack-dev/2017-February/111849.html | 11:18 |
| *** jkilpatr has joined #openstack-ironic | 11:24 | |
| sambetts | vsaienk0: hmm.... I guess then we need to make the decision to cut and backport everything until we tag our final release for stable/pike... | 11:28 |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic master: Add the new capabilities to the iLO InspectInterface https://review.openstack.org/422902 | 11:28 |
| sambetts | but during that period grenade tests on master will be testing stable/pike -> master which isn't a valid test case anyway because we should only be testing between major versions and master is still stable/pike technically | 11:29 |
| *** lucasagomes is now known as lucas-hungry | 11:30 | |
| *** aparnav has quit IRC | 11:32 | |
| *** makowals has joined #openstack-ironic | 11:36 | |
| *** jkilpatr has quit IRC | 11:37 | |
| *** jkilpatr has joined #openstack-ironic | 11:37 | |
| *** Guest58397 is now known as dims | 11:37 | |
| vsaienk0 | sambetts: at the moment we do testing from master to master | 11:38 |
| sambetts | we do... but we shouldn't be... and we can change that by https://review.openstack.org/#/c/494482/ | 11:39 |
| *** Guest88899 is now known as rook | 11:39 | |
| TheJulia | I guess if someone wants to edit the prelude, I can review and land it when I get home from my unexpected errand this morning, and then put in for the branch to be created | 11:41 |
| TheJulia | And then we can release 9.0.1 with the other misc fixes/items we've been trying to land for a while | 11:41 |
| TheJulia | Does that work for everyone? | 11:42 |
| fellypefca | TheJulia Hi. I did not quite understand what exactly you asked me to remove here to avoid confusion. :/ https://review.openstack.org/#/c/493983/2/releasenotes/notes/oneviewclient-version-38a674bb1d40671e.yaml | 11:42 |
| sambetts | TheJulia: sure, what should I be editing? and we still have the multinode grenade issue to contend with :( | 11:43 |
| TheJulia | sambetts: let me get the URL, one moment. We need to revise the hardware types statement | 11:44 |
| TheJulia | Or, we could cut the release and back port the Reno.... | 11:44 |
| TheJulia | Possibly | 11:44 |
| sambetts | TheJulia: not sure we can backport renos like that because they're tied to the git tags | 11:45 |
| sambetts | TheJulia: but we can create the branch without a tag | 11:45 |
| TheJulia | Yeah, will need to ask the release team when I get back to the house | 11:45 |
| *** yangyapeng has joined #openstack-ironic | 11:49 | |
| karimb | pas-ha now, i get "no default value found for management interface" | 11:51 |
| pas-ha | do you enable the ipmi hw type too? | 11:52 |
| karimb | pas-ha no | 11:56 |
| *** mat128 has joined #openstack-ironic | 12:00 | |
| *** cdearborn has joined #openstack-ironic | 12:03 | |
| *** xavierr has joined #openstack-ironic | 12:07 | |
| karimb | pas-ha i should ? | 12:07 |
| karimb | how ? | 12:07 |
| *** marios has quit IRC | 12:08 | |
| *** marios has joined #openstack-ironic | 12:08 | |
| *** rbartal_ has quit IRC | 12:10 | |
| *** jlabarre has joined #openstack-ironic | 12:10 | |
| TheJulia | ironic cores, super easy release note fix https://review.openstack.org/#/c/494007/ | 12:11 |
| *** jtomasek_ has joined #openstack-ironic | 12:14 | |
| *** jtomasek has quit IRC | 12:16 | |
| TheJulia | another quick doc patch https://review.openstack.org/#/c/494329 | 12:17 |
| TheJulia | sambetts: vsaienk0: Looks like the release team is okay with us branching, and cutting another release later with moving the tag on master for the base of the branch so we don't run into reno headaches | 12:18 |
| TheJulia | so if there are no objections, we can cut a release as of right now(), land the stuff we need to master if that fixes our issues, then release the final pike product next week as 9.0.1 | 12:19 |
| *** marios has quit IRC | 12:19 | |
| *** marios has joined #openstack-ironic | 12:20 | |
| sambetts | TheJulia: we'll have to backport all the things to have them tested correctly | 12:20 |
| openstackgerrit | Merged openstack/ironic master: [releasenotes] update irmc's boot-from-volume support https://review.openstack.org/494007 | 12:20 |
| TheJulia | actually, we won't have to | 12:20 |
| TheJulia | we will just change the tag where stable/pike is based | 12:20 |
| *** dprince has joined #openstack-ironic | 12:20 | |
| sambetts | TheJulia: grenade on master doesn't test stable/ocata -> stable/pike any more | 12:20 |
| sambetts | TheJulia: so to have the patches tested for stable/ocata -> stable/pike upgrade we'd have to backport them | 12:21 |
| sambetts | and run them through the system gain | 12:21 |
| sambetts | again* | 12:21 |
| TheJulia | I guess true, and then we could squash it all out of existence to prevent reno headaches | 12:22 |
| TheJulia | alternatively, we edit the prelude reno and change the reference to all hardware types, and then back port idrac hardware type | 12:23 |
| *** karimb has quit IRC | 12:23 | |
| TheJulia | and either use the back ported idrac hw type's release note, and live with the prelude as is. | 12:23 |
| TheJulia | we could change it to "nearly all" or something like that, so it is not specific and not incorrect | 12:24 |
| sambetts | I think if we cut the branch now (no tag) and then backport everything that we merge into master, then we can tag off with no headaches | 12:25 |
| TheJulia | so | 12:25 |
| TheJulia | the model that they built basically forces branches to point at the tag, since all branches are is named tags | 12:25 |
| TheJulia | we don't have repo level access to do that by hand anymore | 12:25 |
| TheJulia | so we're kind of stuck cutting 9.0.0 and then 9.0.1 | 12:26 |
| openstackgerrit | Fellype Cavalcante proposed openstack/ironic master: Updating the minimum version of python-oneviewclient https://review.openstack.org/493983 | 12:27 |
| TheJulia | sambetts: we might be thinking the same thing, just different resulting version numbers | 12:27 |
| sambetts | I thought the projects like nova and neutron created branches without a tag for feature freeze :/ le sigh ... | 12:28 |
| *** bfournie has quit IRC | 12:28 | |
| TheJulia | I think they move their tags, one moment i'll double check | 12:28 |
| sambetts | thats annoying ... I guess we could do 9.0.0rc1 or something | 12:29 |
| *** bfournie has joined #openstack-ironic | 12:29 | |
| sambetts | and then 9.0.0 | 12:29 |
| TheJulia | ohhh, maybe | 12:29 |
| sambetts | I think that might be how the other projects get around it | 12:29 |
| TheJulia | so nova cut their branch off of 16.0.0.0rc1 | 12:30 |
| TheJulia | when they added it | 12:30 |
| *** jcoufal has joined #openstack-ironic | 12:31 | |
| sambetts | so I wonder how that appears in the release notes | 12:31 |
| * TheJulia looks at their last release | 12:31 | |
| TheJulia | they put highlight into their releases repo entry..... | 12:31 |
| sambetts | :/ | 12:31 |
| TheJulia | so nova, looks like they just back ported everything, at least releases repo wise, they never moved their branch base tag | 12:32 |
| sambetts | yeah that how the other projects work | 12:33 |
| TheJulia | yeah, neutron too | 12:33 |
| sambetts | they cut an rc and then backport everything | 12:33 |
| TheJulia | yup | 12:33 |
| TheJulia | lets just do it | 12:33 |
| sambetts | so 9.0.0rc1 | 12:33 |
| *** bfournie has quit IRC | 12:33 | |
| sambetts | and then we'll tag 9.0.0 final when we're happy | 12:33 |
| sambetts | ? | 12:33 |
| TheJulia | yup | 12:33 |
| TheJulia | it is our only option | 12:33 |
| *** mat128 has quit IRC | 12:34 | |
| TheJulia | sambetts: do you know if neutron does like a prelude note? | 12:34 |
| *** karimb has joined #openstack-ironic | 12:35 | |
| sambetts | TheJulia: https://docs.openstack.org/releasenotes/nova/ocata.html#id8 | 12:35 |
| sambetts | nova did | 12:35 |
| sambetts | for their last release | 12:35 |
| TheJulia | yeah, I just found their pike prelude | 12:35 |
| sambetts | and I'm just looking at their tags in their repo and they have 15.0.0rc1 and rc2 and then 15.0.0 final, and 15.0.0 seems to hide the rcs in the release notes | 12:36 |
| *** mat128 has joined #openstack-ironic | 12:36 | |
| TheJulia | so the baremetal filter deprecation got back ported in nova | 12:37 |
| * TheJulia looks at their current notes | 12:37 | |
| TheJulia | seems to be there just fine | 12:37 |
| TheJulia | so, I think we should just make ruby's reno prelude a little less specific, land that, branch it, and we can back port idrac | 12:38 |
| TheJulia | specifically https://review.openstack.org/#/c/493906/ | 12:39 |
| TheJulia | line 17 specifically needs to be edited | 12:40 |
| TheJulia | or ,maybe not? | 12:40 |
| TheJulia | rloo should be on shortly | 12:40 |
| sambetts | TheJulia: just checked, reno strips the pre-release tags like rc1 / b1 etc tags for exactly this perpose | 12:40 |
| sambetts | purpose* | 12:40 |
| sambetts | TheJulia: so that 15.0.0 includes every release note in any 15.0.0 pre-release | 12:41 |
| sambetts | which is pretty cool | 12:41 |
| TheJulia | awesome, well then.... review the link above, I think we can land as is, and we'll cut our rc1 once it hits the repo | 12:42 |
| Nisha_Agarwal | TheJulia, sambetts so the gate would be fixed after we have the above tag done? | 12:42 |
| sambetts | Nisha_Agarwal: partially fixed ... | 12:42 |
| sambetts | ;) | 12:42 |
| Nisha_Agarwal | sambetts, cool :) | 12:43 |
| TheJulia | it will get somewhat better ;) | 12:43 |
| TheJulia | we just can't measure how much | 12:43 |
| Nisha_Agarwal | TheJulia, :) | 12:43 |
| *** lucas-hungry is now known as lucasagomes | 12:44 | |
| *** hw_wutianwei has quit IRC | 12:44 | |
| TheJulia | and now, we wait a little bit. :( | 12:45 |
| *** heroux has joined #openstack-ironic | 12:46 | |
| *** wong has quit IRC | 12:46 | |
| * TheJulia waits for docs jobs to finish running | 12:49 | |
| TheJulia | https://review.openstack.org/#/c/493990/ is in route to land as well | 12:49 |
| *** rloo has joined #openstack-ironic | 12:50 | |
| TheJulia | good morning rloo | 12:50 |
| rloo | good morning TheJulia! | 12:50 |
| rloo | and hi and good morning to everyone else! | 12:50 |
| vdrok | morning TheJulia and rloo | 12:51 |
| vdrok | TheJulia: sambetts so, we're cutting pike now right? | 12:51 |
| rloo | hi vdrok | 12:51 |
| openstackgerrit | Merged openstack/ironic master: [reno] Add prelude for Pike release https://review.openstack.org/493906 | 12:51 |
| TheJulia | once this patch lands, and then we'll backport the things we've been unable to land, and then cut 9.0.0 | 12:52 |
| sambetts | vdrok: we're ctting the branch, and a release candiate, but not the final release | 12:52 |
| vdrok | yup sure | 12:52 |
| TheJulia | https://review.openstack.org/494530 | 12:56 |
| Nisha_Agarwal | so TheJulia 9.0.0 will not be stable pike, correct? | 12:58 |
| sambetts | 9.0.0 will be stable/pike | 12:58 |
| TheJulia | it is not going to work | 12:58 |
| sambetts | ? | 12:58 |
| vdrok | rloo: we also need to add pike in https://github.com/openstack/ironic/blob/master/ironic/common/release_mappings.py right? | 12:58 |
| Nisha_Agarwal | TheJulia, ? | 12:58 |
| TheJulia | ERROR: could not validate version '9.0.0rc1': Version 9.0.0rc1 looks like a pre-release and the release model does not allow for it | 12:59 |
| TheJulia | ERROR: could not validate version '9.0.0rc1': Version '9.0.0rc1' does not match canonical form '9.0.0.0rc1' | 12:59 |
| rloo | vdrok: yes, i thought of it yesterday and decided that it could wait til after we branched/gate was fixed. | 12:59 |
| sambetts | TheJulia: ffs... thats because we're release-with-itermediary right? | 12:59 |
| TheJulia | yup | 12:59 |
| vdrok | ok then we can do just 9.0.0 | 12:59 |
| sambetts | doing 9.0.0 wil screw up our release notes | 13:00 |
| rloo | vdrok: unless you can think of a reason why it needs to be done just before we branch? the intent was to do it before branching but i thought about it and think it can be done after branching. | 13:00 |
| rloo | i think maybe we can't do rc cuz we are using the semver releasing mechanism? | 13:01 |
| rloo | oh yeah, what sambetts sez ^^ | 13:01 |
| sambetts | ffs... why even offer alternative release models when your not going to support it | 13:01 |
| vdrok | rloo: I think the master will be broken immediately after branching because of https://review.openstack.org/494435 | 13:01 |
| sambetts | #rage | 13:01 |
| TheJulia | well, now we need to edit the releae notes | 13:01 |
| vdrok | rloo: namely, the services will fail to start because the pin will be set to pike and pike is not in the release mappings | 13:02 |
| rloo | vdrok: what is that patch for? oh but sambetts -2'd it? | 13:02 |
| vdrok | during upgrade | 13:02 |
| vdrok | yeah | 13:02 |
| TheJulia | I feel like we're in a giant chicken/egg scenario because of the grenade jobs | 13:02 |
| rloo | vdrok: well, the pin is still set to master, code doesn't know about pike? | 13:03 |
| sambetts | basically they've decided for us that master is not queens | 13:03 |
| sambetts | s/not/now | 13:03 |
| vdrok | rloo: here is what's happening http://logs.openstack.org/68/493468/6/check/gate-grenade-dsvm-ironic-ubuntu-xenial/8ea3585/logs/screen-ir-api.txt.gz#_Aug_17_11_35_55_178300 | 13:03 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic-inspector-specs master: Enumerate Inspector errors https://review.openstack.org/494534 | 13:03 |
| sambetts | vdrok, rloo, TheJulia: we could do https://review.openstack.org/#/c/494482/ to force greande to understand the current branch situation correctly | 13:04 |
| vdrok | that's going to be the same on master after we cut pike | 13:04 |
| rloo | vdrok: oh, how'd pike get set there? | 13:04 |
| vdrok | rloo: in the upgrade.sh in the patch that is -2ed | 13:04 |
| TheJulia | sambetts: I say unabandon it at this point | 13:05 |
| *** bfournie has joined #openstack-ironic | 13:05 | |
| TheJulia | I've wf-1'ed my release attempt | 13:05 |
| vdrok | sambetts: i don't really want to do that :( we can't test if it works. and if it does not, the whole gate goes down | 13:05 |
| *** vishwanathj has quit IRC | 13:05 | |
| rloo | vdrok: i'm confused. your 494435 sets it to 'ocata'. what set it to pike? | 13:05 |
| vdrok | rloo: the variable I removed in that patch | 13:06 |
| *** vishwanathj has joined #openstack-ironic | 13:06 | |
| rloo | vdrok: OH, that BASE_DEVSTACK_BRANCH thing. | 13:06 |
| sambetts | vdrok: working through devstack-gate, that override will cause base to be set to stable/ocata, and target to be set to stable/pike which doesn't exist so it falls back to master | 13:06 |
| rloo | vdrok: so you mean w/o that patch, it sets it to pike. gotcha. | 13:06 |
| vdrok | rloo: correct | 13:06 |
| rloo | vdrok: and that is what is breaking our gate? | 13:06 |
| vdrok | rloo: right now, yes | 13:06 |
| vdrok | the devstack patches are being merged right now so the multinode failures we were seeing should go away | 13:07 |
| rloo | vdrok: so what is the problelm with approving that patch now? | 13:07 |
| pas-ha | karimb: sorry, I was out. did you also enabled all the corresponding interfaces your driver / hw type needs? | 13:07 |
| vdrok | rloo: because basically doing that without cutting pike means that the grenade is not testing ocata -> master | 13:08 |
| sambetts | rloo: IMO that patch doesn't fix anything | 13:08 |
| karimb | pas-ha yeah i m doing it right now | 13:08 |
| vdrok | rloo: yeah, it's just a hack | 13:08 |
| sambetts | rloo: grenade is testing master --> master on master right now, not stable/ocata --> master | 13:08 |
| karimb | understood it was just a ironic.conf editing issue | 13:08 |
| sambetts | rloo: we need https://review.openstack.org/#/c/494482/ to make grenade actually test what its supposed to test | 13:08 |
| vdrok | sambetts: rloo but it works because we did not bump any rpc/object versions in pike yet, so backporting to ocata still possibl | 13:09 |
| *** SerenaFeng has joined #openstack-ironic | 13:09 | |
| sambetts | its still not testing our upgrades anymore | 13:09 |
| sambetts | we might as well have grenade == return true | 13:09 |
| rloo | vdrok, sambetts: so it seems like we have a problem that we've had in ocata; grenade testing things that is incorrect for ironic | 13:09 |
| rloo | vdrok, sambetts: so we need to fix it | 13:09 |
| rloo | vdrok, sambetts: is it clear to both of you what that fix might be? not just for now, but future releases too? | 13:10 |
| vdrok | yes, it does not, I get that. just, if we decide to go with project-config change, we need at least one project config core tracking it, so we can land updates quickly if anything goes wrong | 13:10 |
| vdrok | sambetts: rloo ^^ | 13:10 |
| *** SerenaFeng has quit IRC | 13:10 | |
| vdrok | rloo: cut pike at the same time as integrated projects? :) | 13:10 |
| TheJulia | that shouldn't be too hard | 13:10 |
| rloo | vdrok, sambetts: my feeling is that if the 'right' fix is going to take too long, we should hack something to get our release out now. and then fix correctly later. | 13:10 |
| rloo | TheJulia: is it true, we have to cut a stable/pike branch today? | 13:11 |
| vdrok | rloo: i don't think we have to | 13:11 |
| vdrok | it's just it seems it will make our lifes easier | 13:11 |
| rloo | vdrok: oh. sec, there was email about that if i can find it... | 13:11 |
| *** baoli has joined #openstack-ironic | 13:11 | |
| *** ElCoyote_ has quit IRC | 13:12 | |
| sambetts | rloo: hack something in and get a potentally broken release out... because we're not testing if it can upgrade from stable/ocata | 13:12 |
| TheJulia | rloo: we were advised to cut it as early as possible, drop dead I think was next week | 13:12 |
| rloo | I just looked. Deadline for last release candidates / intermediary releases: August 24 | 13:13 |
| rloo | Final Pike release: August 30 | 13:13 |
| rloo | so we wanted to cut the branch this week. let's sit back a few minutes and ponder this. | 13:13 |
| sambetts | can we cut stable/pike using the last itermediary release we made? | 13:14 |
| sambetts | the branch I mean | 13:14 |
| rloo | i'm not quick enough to know what the right fix is wrt the grenade issue. vdrok & sambetts, do you know what the right solution is? cuz we're going to encounter this again in the future. | 13:14 |
| TheJulia | sambetts: the last release was earlier in the cycle | 13:14 |
| rloo | we have time to get cores from other projects to approve, as long as we know what the solution is. | 13:14 |
| TheJulia | so... not exactly | 13:14 |
| sambetts | rloo: unfortunatly the whole system appears to be built for the projects that can cut release candidates | 13:14 |
| vdrok | rloo: as I said, I think the right thing is to cut release at the same time as integrated projects | 13:15 |
| rloo | vdrok: OH. so you're saying that the way grenade etc is set up, it would be diff to change to accommodate us? | 13:15 |
| rloo | s/diff/difficult/ | 13:15 |
| vdrok | yeah | 13:15 |
| rloo | or impossible | 13:15 |
| rloo | i think it is something to bring up in ptg. don't we already have something to bring up wrt grenade? | 13:16 |
| *** udesale has joined #openstack-ironic | 13:16 | |
| sambetts | well... actually I think the easiest long term fix would be to allow projects to define what their own master release mapping is | 13:16 |
| TheJulia | okay, lets take a step back | 13:16 |
| sambetts | and not make assumtions that master is the next release after a certain point in time | 13:16 |
| sambetts | ... | 13:16 |
| TheJulia | what can we do now, in the short term, | 13:16 |
| sambetts | I think https://review.openstack.org/#/c/494482/ is our best bet | 13:16 |
| vdrok | rloo: sambetts TheJulia well, vsaienk0 already posted what sdague thought about that last release http://lists.openstack.org/pipermail/openstack-dev/2017-February/111849.html | 13:16 |
| TheJulia | because without stable/pike, things will continue to land on other master branches that will break us further | 13:17 |
| sambetts | TheJulia: there is also that... | 13:17 |
| sambetts | *head meet desk* | 13:17 |
| rloo | vdrok: ah, thx for reminding me about that email | 13:18 |
| rloo | ok, so i think we should cut a stable/pike branch | 13:18 |
| sambetts | it really feels like we're just going to need to bite the bullet and cut 9.0.0 | 13:18 |
| * TheJulia begins pouring glasses of whisky for everyone | 13:18 | |
| rloo | yeah. | 13:18 |
| TheJulia | sambetts: agreed | 13:18 |
| rloo | just wondering if it can be numbered something else but we don't have -1 | 13:18 |
| vdrok | on top of it all, project-config is broken now too :) | 13:18 |
| sambetts | yup | 13:19 |
| TheJulia | we can add a hardware type, and make it 9.1.0! | 13:19 |
| rloo | vdrok: oh geez. what does that mean? how broken is it? | 13:19 |
| rloo | TheJulia: yes, hw type + ilo/capabilities if we end up doing that. | 13:19 |
| sambetts | TheJulia: I guess we should save rloo's prelude until whatever the final version is we cut right? | 13:19 |
| sambetts | TheJulia: so its at the top of the Pike series | 13:19 |
| rloo | yeah, my preference is not to put the prelude until the *final* release | 13:20 |
| rloo | did the prelude get merged? | 13:20 |
| sambetts | +100 | 13:20 |
| TheJulia | sambetts: we need to edit it real quick | 13:20 |
| TheJulia | +10, i'll go click revert | 13:20 |
| rloo | oh yeah, it got merged | 13:20 |
| rloo | but we can edit it | 13:20 |
| rloo | i mean revert :) | 13:20 |
| * sambetts says revert it | 13:20 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Revert "[reno] Add prelude for Pike release" https://review.openstack.org/494542 | 13:20 |
| vdrok | rloo: I thibnk they need to merge smth like https://review.openstack.org/494456 | 13:21 |
| TheJulia | so, i'll cut 9.0.0 from the merge _before_ the prelude landed | 13:21 |
| sambetts | TheJulia: do we need to revert it then? | 13:21 |
| TheJulia | and we can add the prelude with the hw types, and call it 9.1.0 and not feel horrible. | 13:21 |
| rloo | TheJulia: yup. although did that irmc reno get updated/merged? | 13:21 |
| TheJulia | rloo: it merged right before that one I believe | 13:22 |
| rloo | TheJulia: just checked; it merged | 13:22 |
| rloo | let's revert just to keep things clean/easy. | 13:22 |
| sambetts | +1 | 13:22 |
| TheJulia | I was typing the same thing out :) | 13:23 |
| sambetts | TheJulia: we should get the docs landed for it too | 13:23 |
| sambetts | TheJulia: https://review.openstack.org/#/c/494329/1 | 13:23 |
| rloo | ok I +A'd the revert. | 13:23 |
| TheJulia | we can backport them | 13:23 |
| rloo | it'll need to be updated anyway, to 9.1.0, right? | 13:23 |
| TheJulia | rloo: yeah | 13:24 |
| * rloo wonders what this means for rolling upgrades, 7.0.0 -> 9.1.0 but it should work, heh. | 13:24 | |
| sambetts | http://s2.quickmeme.com/img/12/1227c32e4c9eccafd355175bcf6174adbb35b9e9120c627d0251b44532e50bf2.jpg | 13:24 |
| rloo | spoken by a canadian :) | 13:24 |
| TheJulia | heh | 13:25 |
| rloo | so what's the plan of action. 1. revert prelude patch; 2. cut stable/pike branch. ? | 13:25 |
| sambetts | 3. backport everything we merge into master | 13:25 |
| rloo | after cutting, the gate isn't going to miraculously work, will it? | 13:26 |
| TheJulia | https://review.openstack.org/494530 | 13:26 |
| sambetts | 4. cut 9.1.0 when we're happy everything is in stable/pike | 13:26 |
| TheJulia | yup, that is the plan | 13:26 |
| rloo | we should add this to the etherpad so people know what the plan is. or send email. not sure. | 13:26 |
| rloo | if we are going to backport everything we merge into master, we need to be careful about what we merge. | 13:27 |
| sambetts | +1 feature freeze still applys | 13:27 |
| sambetts | we should only be merging the things that we were going to merge into pike anywher | 13:27 |
| sambetts | anyway* | 13:27 |
| TheJulia | agreed | 13:28 |
| rloo | i will need to push a patch up to update common/release_mappings for the new intermediate release. i may include the final release there too. | 13:28 |
| TheJulia | rloo: good idea | 13:29 |
| rloo | take a look at whiteboard, L120 | 13:31 |
| karimb | pas-ha do i need a driver entrypoint anyway if i enabled the hardware type code ? | 13:31 |
| rloo | sambetts, vdrok, TheJulia ^^ cuz I put your names down :) | 13:32 |
| *** rbartal has joined #openstack-ironic | 13:32 | |
| vdrok | yeah, lgtm | 13:32 |
| openstackgerrit | yolanda.robla proposed openstack/bifrost master: DNM: Change path for the vm logs https://review.openstack.org/494549 | 13:35 |
| *** Guest53762 is now known as dansmith | 13:36 | |
| *** zhangfei has joined #openstack-ironic | 13:37 | |
| rloo | jlvillal: if you haven't yet, please remove the rechecks because the gate is broken. | 13:38 |
| rloo | TheJulia, sambetts, vdrok: we *still* need to discuss/agree on a process for this or we'll have the same problem with every release. not sure how to proceed here. guess i'll leave a note in whiteboard for dmitry to read... | 13:40 |
| rloo | all hail our vacationing PTL dtantsur | 13:40 |
| *** amarao1 has joined #openstack-ironic | 13:41 | |
| *** amarao1 has quit IRC | 13:42 | |
| TheJulia | I think the post-mortem discussion and figuring out how to prevent it in the future is a whisky discussion | 13:42 |
| *** amarao1 has joined #openstack-ironic | 13:42 | |
| sambetts | rloo: yeah... PTG topic I think... | 13:42 |
| TheJulia | because there are many contributing factors | 13:42 |
| sambetts | my thought are that projects need to be able to declare what openstack version their master branch is | 13:43 |
| vdrok | we can write our own grenade :) | 13:43 |
| * rloo wonders how successful whisky discussions are, vs PTG discussions | 13:43 | |
| *** amarao1 has quit IRC | 13:43 | |
| TheJulia | rloo: they could be the same thing | 13:43 |
| rloo | :) | 13:43 |
| rloo | and have whisky provided by the foundation! win-win | 13:44 |
| *** amarao1 has joined #openstack-ironic | 13:44 | |
| sambetts | if projects could declare what openstack version their master branch is then none of this would be an issue | 13:44 |
| rloo | i guess i am wondering if it should be a xproject discussion | 13:44 |
| TheJulia | In my mind, we are like a round peg trying to pass through a square hole | 13:44 |
| openstackgerrit | Merged openstack/ironic master: [docs] update irmc boot-from-volume https://review.openstack.org/494329 | 13:44 |
| TheJulia | with releases that is | 13:44 |
| rloo | and if so, it needs to be scheduled in on mon-tues, whenever | 13:44 |
| openstackgerrit | Merged openstack/ironic master: Revert "[reno] Add prelude for Pike release" https://review.openstack.org/494542 | 13:44 |
| sambetts | TheJulia: goverance provides this release model... they should support it | 13:44 |
| rloo | yay, we reverted quickly. (on the bright side, heh) | 13:45 |
| TheJulia | sambetts: I totally agree, although all the other projects want to be very water like | 13:45 |
| TheJulia | err, waterfall like | 13:45 |
| vdrok | TheJulia: so do we need to pour some +1s on the releases patch? | 13:45 |
| rloo | ^^ vdrok beat me to it | 13:46 |
| TheJulia | vdrok: sure! I already let ttx know | 13:46 |
| TheJulia | we should keep a list of things to back port.... | 13:46 |
| rloo | TheJulia: i think the default should be to backport everything? | 13:47 |
| TheJulia | we kind of already do I see | 13:47 |
| rloo | TheJulia: otherwise, too much book keeping | 13:47 |
| TheJulia | not sure we need to backport https://review.openstack.org/494329 tbh, since the docs will point to master branch | 13:47 |
| rloo | TheJulia: i only wanted to list on etherpad, the things that *must* be backported | 13:47 |
| TheJulia | yeah | 13:48 |
| rloo | TheJulia: i think we should backport docs. at least in the past, we had docs/release. | 13:48 |
| sambetts | TheJulia: I thought we would need to because there will be docs.openstack.org/ironic/pike right? | 13:48 |
| vdrok | ++ | 13:48 |
| rloo | yeah, that ^^, i hope we still do anyway | 13:48 |
| TheJulia | okay, I was just thinking one less patch :) | 13:48 |
| rloo | TheJulia: I believe when i asked dtantsur last week, he said it was ok to continue to backport docs to stable branch, which is why i didn't worry too much about getting docs merged | 13:49 |
| TheJulia | yeah, I was only thinking of saving the gate some excess cycles. Anyway, now we get to wait once again | 13:50 |
| sambetts | TheJulia: I think we need to be careful and should only merge the patches onto master when their backport equivolents pass all CI | 13:51 |
| sambetts | rloo: ^ | 13:51 |
| rloo | TheJulia: wrt https://review.openstack.org/#/c/494530/, do you need to update the hash, i don't know how it works | 13:51 |
| sambetts | because the backports are the only patches that are going to be running the correct grenade | 13:51 |
| vdrok | rloo: nope, we can release from older commits as it seems | 13:51 |
| vdrok | that hash is before the prelude was merged | 13:52 |
| rloo | vdrok: ok, so why have a hash then? (just wondering but don't worry, I'll just sign.) | 13:52 |
| sambetts | rloo: tags are tied to specific commits, not to any point in time | 13:53 |
| vdrok | rloo: you mean why did we revert? :) without hash we could not select a specific commit | 13:53 |
| TheJulia | sambetts: In that case, we need to remove all approvals, or -2 all approved patches | 13:53 |
| * TheJulia will go do that | 13:53 | |
| sambetts | TheJulia: yeah :'( | 13:53 |
| rloo | sambetts: we can't backport until *after* the patch lands in master. or do you mean, we should make sure the backport patches exist/pass, before landing the master patch? | 13:54 |
| vdrok | rloo: yup | 13:55 |
| TheJulia | we can't land it, we can propose it | 13:55 |
| rloo | gotcha. | 13:55 |
| *** jaypipes has quit IRC | 13:55 | |
| rloo | which brings up something else. who can land patches to stable/pike. me, TheJulia, and dtantsur? | 13:55 |
| TheJulia | just stable cores | 13:56 |
| rloo | TheJulia: right, so us three + the other cores that aren't in ironic. | 13:56 |
| TheJulia | yup | 13:57 |
| rloo | TheJulia: I'll discuss with you later today, whether I ought to keep an eye out for things tomorrow. | 13:57 |
| *** jlabarre has quit IRC | 13:57 | |
| vdrok | if these are just backports, I guess we can do with just one +2+A? | 13:58 |
| openstackgerrit | OpenStack Release Bot proposed openstack/ironic-inspector stable/pike: Update .gitreview for stable/pike https://review.openstack.org/494560 | 13:58 |
| openstackgerrit | OpenStack Release Bot proposed openstack/ironic-inspector stable/pike: Update UPPER_CONSTRAINTS_FILE for stable/pike https://review.openstack.org/494561 | 13:58 |
| openstackgerrit | OpenStack Release Bot proposed openstack/ironic-inspector master: Update reno for stable/pike https://review.openstack.org/494562 | 13:58 |
| TheJulia | sweet | 13:59 |
| rloo | vdrok: almost all patches are backports :) I think we can if emergency, i don't think we should in general. | 13:59 |
| vdrok | rloo: yeah, except in this case, pike and master would be identical :) | 14:00 |
| *** jlabarre has joined #openstack-ironic | 14:00 | |
| vdrok | s/would/will | 14:00 |
| rloo | vdrok: true, but i'm not going to make that call :) | 14:00 |
| TheJulia | vdrok: I agree with you there, but I'm with rloo, in that I prefer things be as proper as possible | 14:01 |
| rloo | vdrok, sambetts: wrt grenade/devstack-related patches, etc, if there are some that we still need/want to land, maybe list them under the release plan so folks know | 14:01 |
| TheJulia | I _htink I've got the stuff that would just land blocked now | 14:01 |
| TheJulia | if someone would double check openstack/ironic | 14:01 |
| vdrok | rloo: I don;t think so. only ironic things | 14:01 |
| rloo | vdrok: ok, so we don't need to backport that 999999 change? | 14:01 |
| vdrok | rloo: we did already, it's the bacports that are currently merging | 14:02 |
| *** mrch has quit IRC | 14:02 | |
| rloo | vdrok: oh, they got approved then. good. | 14:02 |
| vdrok | TheJulia: also, can we get the release without dmitry? ttx wants his vote on that patch | 14:03 |
| TheJulia | ugh, I'll comment | 14:03 |
| rloo | heh, i just replied | 14:04 |
| rloo | i don't know if dmitry explicitly mentioned in writing anywhere, that julia would take over? | 14:04 |
| rloo | TheJulia: ok, found a link to dtantsur passing release to you | 14:06 |
| rloo | TheJulia: commented | 14:06 |
| TheJulia | okay, I was still looking for it | 14:06 |
| TheJulia | in the text | 14:06 |
| *** jaypipes has joined #openstack-ironic | 14:08 | |
| jlvillal | rloo, I stopped my recheck/watcher script | 14:12 |
| *** trown is now known as trown|brb | 14:12 | |
| *** rbartal has quit IRC | 14:12 | |
| rloo | jlvillal: thx! | 14:13 |
| vdrok | morning jlvillal | 14:14 |
| rloo | and good morning jlvillal :) | 14:15 |
| *** trown|brb is now known as trown | 14:17 | |
| rloo | vdrok, vsaienk0: is this something that we want backported? https://review.openstack.org/#/c/489418/ | 14:18 |
| rloo | to stable/pike? ^^ | 14:18 |
| vdrok | rloo: yes I think so | 14:18 |
| openstackgerrit | Xavier proposed openstack/ironic master: Introduce hpOneView and ilorest to OneView https://review.openstack.org/468129 | 14:18 |
| openstackgerrit | Xavier proposed openstack/ironic master: Migrate oneview driver to use python-hpOneView https://review.openstack.org/466872 | 14:18 |
| openstackgerrit | Xavier proposed openstack/ironic master: Fix persistent information when getting boot device https://review.openstack.org/491983 | 14:18 |
| rloo | vdrok: so we should create a backport patch first etc before landing it? | 14:18 |
| vdrok | rloo: I'd also like to get this one https://review.openstack.org/480905 with backports in | 14:18 |
| vdrok | rloo: correct | 14:19 |
| vdrok | pike is not created yet :) will do after that | 14:19 |
| rloo | vdrok: wrt 480905, why does it depend on the stable/ocata one? | 14:20 |
| rloo | vdrok: it isn't a backport. | 14:20 |
| vdrok | rloo: yes, it is a chain of patches across releases. kind of backport. | 14:20 |
| rloo | vdrok: but we're not supposed to approve patches to branches, until the patch lands in master. | 14:21 |
| *** karimb has quit IRC | 14:21 | |
| rloo | vdrok: but you need th patch to land in stable/ocata first? | 14:21 |
| vdrok | rloo: this one is an exception :( | 14:21 |
| vdrok | yup | 14:21 |
| rloo | vdrok: and once stable/pike exists, you'll need it to land in stable/pike before master? | 14:21 |
| vdrok | yes | 14:21 |
| vdrok | the reason to have it is, right now basically every plugin installed after ironic will be installed from the old release during upgrade, while we need it installed from new | 14:22 |
| vdrok | that chain would fix it | 14:22 |
| vdrok | s/old release/old upgrade script | 14:23 |
| *** karimb has joined #openstack-ironic | 14:23 | |
| sambetts | le sigh... I hate grenade... | 14:23 |
| *** jlabarre has quit IRC | 14:23 | |
| rloo | vdrok: i wonder if we should start opening bugs against these so we can find/remember them later... or put more info in the commit msgs so people will understand more context... | 14:24 |
| rloo | sambetts: love/hate relationship :( and :) | 14:24 |
| *** hshiina has quit IRC | 14:24 | |
| vdrok | rloo: hrm, I wrote this in the message, maybe not clearly enough tho. I can create a bug if you think it would be better | 14:25 |
| TheJulia | I think I'm leaning more towards hate, personally | 14:25 |
| sambetts | yup... and I hate devstack-gate... | 14:25 |
| rloo | vdrok: i don't know what the right thing is, i think the problem here is that it doesn't follow the norm. so when i look at https://review.openstack.org/#/c/480905/, i don't know why it depends on the other. | 14:26 |
| TheJulia | okay, so who is going to handle back port creation? | 14:26 |
| vdrok | rloo: ok, i'll clarify that bit when updating | 14:26 |
| *** jlabarre has joined #openstack-ironic | 14:26 | |
| rloo | vdrok: having said that, i'm not that familiar with devstack, so it may be obvious to others :) | 14:26 |
| rloo | TheJulia: I can do it, but will wait a few min to see if anyone else wants to volunteer. (Call me Tom Sawyer...) | 14:27 |
| TheJulia | I guess since they will all be clean one of us can do back ports without any issues | 14:27 |
| * TheJulia awaits for proposal bot messages | 14:28 | |
| vdrok | TheJulia: rloo well it's just a matter of hitting a button, i can help with that :D | 14:28 |
| TheJulia | vdrok: excellent :) | 14:28 |
| rloo | thx vdrok! :) | 14:28 |
| * TheJulia ponders more coffee | 14:29 | |
| vdrok | ok, I see 9.0.0 branch, but not a tag uet | 14:30 |
| vdrok | yet | 14:30 |
| openstackgerrit | OpenStack Release Bot proposed openstack/ironic stable/pike: Update .gitreview for stable/pike https://review.openstack.org/494578 | 14:31 |
| openstackgerrit | OpenStack Release Bot proposed openstack/ironic stable/pike: Update UPPER_CONSTRAINTS_FILE for stable/pike https://review.openstack.org/494579 | 14:31 |
| openstackgerrit | OpenStack Release Bot proposed openstack/ironic master: Update reno for stable/pike https://review.openstack.org/494580 | 14:31 |
| TheJulia | \o/ | 14:31 |
| vdrok | whee | 14:31 |
| Nisha_Agarwal | :) | 14:31 |
| Nisha_Agarwal | so we have stable pike fo rnow | 14:32 |
| * rloo passes small glass of champagne to everyone; big celebration will be later after the official release... | 14:32 | |
| vdrok | the devstack backport to ocata is not merged yet :) | 14:32 |
| * Nisha_Agarwal joins rloo | 14:32 | |
| TheJulia | rloo: :) | 14:32 |
| fellypefca | \o/ | 14:32 |
| rloo | vdrok: so things will still fail? | 14:32 |
| vdrok | yeah | 14:33 |
| TheJulia | vdrok: who do we need to ping? | 14:33 |
| TheJulia | or is it in flight? | 14:33 |
| rloo | vdrok: well, it shouldn't take very long... | 14:33 |
| vdrok | TheJulia: noone, it's merging | 14:33 |
| TheJulia | ok | 14:33 |
| vdrok | pike backport failed, but i've rechecked already | 14:33 |
| rloo | vdrok: yesterday, it took about 2 hours for the master patch to merge, and it is about 1.5 hours so far for the ocata patch... | 14:35 |
| openstackgerrit | yolanda.robla proposed openstack/bifrost master: DNM: Change path for the vm logs https://review.openstack.org/494549 | 14:35 |
| *** cdearborn has quit IRC | 14:35 | |
| rloo | vdrok: so the ocata patch is needed for master CI to work, the pike patch is needed for pike CI to work, right? | 14:36 |
| *** cdearborn has joined #openstack-ironic | 14:36 | |
| *** Nisha_Agarwal has quit IRC | 14:36 | |
| vdrok | rloo: vice versa, ocata patch for pike ci, pike patch for master ci | 14:36 |
| rloo | vdrok: ok. our bad, sorry. i wasn't sure yesterday if those were needed; we should have gotten them approved yest too. | 14:37 |
| *** Nisha_Agarwal has joined #openstack-ironic | 14:37 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic master: Get rid of sourcing stackrc in grenade settings https://review.openstack.org/494582 | 14:37 |
| TheJulia | rloo: don't feel bad, I didn't completely get that was the case until it was a bit too late last night to get anyone to act upon them. | 14:37 |
| * rloo is reminded of those competitions where the baton is being passed and drops... | 14:37 | |
| TheJulia | so, earliest we can begin back porting is what, ~2 hours? | 14:38 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic master: Get rid of sourcing stackrc in grenade settings https://review.openstack.org/480905 | 14:39 |
| rloo | TheJulia: i suspect so. | 14:39 |
| *** ralonsoh has quit IRC | 14:39 | |
| rloo | vdrok: we won't need this any more? https://review.openstack.org/#/c/494435/ | 14:39 |
| vdrok | rloo: I think we don't | 14:40 |
| vdrok | I'll abandon it when I see things are fine :) | 14:40 |
| rloo | vdrok: ok, thx. | 14:40 |
| TheJulia | I put a big "DO NOT RECHECK" note on the whiteboard | 14:42 |
| TheJulia | We can recheck the patches on master once we're read to let everything begin merging once it looks like things are fixed | 14:42 |
| *** zhangfei has quit IRC | 14:44 | |
| rloo | TheJulia: yup and thx! | 14:45 |
| TheJulia | vdrok: your back port lacks the cherry picked message | 14:45 |
| *** baoli has quit IRC | 14:45 | |
| * TheJulia is sure someone actually cares, but that person is not her right now | 14:46 | |
| vdrok | TheJulia: I'm not sure it needs to have one, as we'll have to merge prior release before consequent one | 14:46 |
| TheJulia | I think the depends-on flag might throw it off then | 14:46 |
| TheJulia | if stable/pike needs to land before master | 14:47 |
| * TheJulia goes back and looks | 14:47 | |
| vdrok | throw off? | 14:47 |
| vdrok | this depends on across branches does work | 14:47 |
| TheJulia | oh, no, I thought it was a different branch | 14:47 |
| TheJulia | https://review.openstack.org/#/c/480904/ <-- rloo | 14:47 |
| *** baoli has joined #openstack-ironic | 14:48 | |
| vdrok | yeah, ocata first, then pike, then master | 14:48 |
| rloo | TheJulia: huh? | 14:48 |
| *** huats has joined #openstack-ironic | 14:48 | |
| TheJulia | okay | 14:48 |
| rloo | TheJulia: oh yeah, see above, i already had that discussion with vdrok. | 14:48 |
| vdrok | but anyway, we could do this after release | 14:48 |
| rloo | vdrok: see, not just me that is confused :) | 14:48 |
| TheJulia | I thought that was just master and stable/pike | 14:49 |
| TheJulia | i didn't realize ocata as well | 14:49 |
| vdrok | yup :( | 14:49 |
| TheJulia | so... I might as well go have a beer and a long lunch because we're going to be waiting at least 2 hours then :( | 14:49 |
| vdrok | zuul says an hour to get the pike ci back | 14:50 |
| vdrok | with master, more like 3 hours | 14:50 |
| rloo | TheJulia: yup. honestly, i don't think things are *that* urgent yet. we can land today, tomorrow, or monday. after that, dmitry can deal with the stress/urgency. ha ha. | 14:50 |
| vdrok | :D | 14:50 |
| TheJulia | rloo: true, the important thing is the branch | 14:50 |
| rloo | well, important thing is getting our ci to work :) | 14:51 |
| TheJulia | well, that too | 14:51 |
| *** amarao1 has quit IRC | 14:52 | |
| *** karimb has quit IRC | 14:58 | |
| openstackgerrit | Merged openstack/ironic-inspector master: Update reno for stable/pike https://review.openstack.org/494562 | 14:58 |
| *** udesale has quit IRC | 15:00 | |
| *** mjura has quit IRC | 15:05 | |
| TheJulia | So one thing worth noting for next week. I need to drive like 6 hours on monday through what is likely to be complete gridlock do to the eclipse, so I'll likely miss the meeting unless I'm at full stop on the road (which could be the case)... and if so, I should likely consider having all the supplies to start grilling or something :) | 15:05 |
| TheJulia | I only mention it now because my plans went sideways and everything got delayed a day | 15:06 |
| rloo | TheJulia: sweet (about the eclipse trip) | 15:06 |
| TheJulia | I know in North Carolina, they are expecting 12-16 hours of gridlock | 15:07 |
| *** Nisha_Agarwal has quit IRC | 15:10 | |
| *** srobert has joined #openstack-ironic | 15:11 | |
| *** amarao1 has joined #openstack-ironic | 15:13 | |
| rloo | TheJulia: wow, maybe you want to get there the day before? | 15:15 |
| TheJulia | vdrok: http://logs.openstack.org/05/480905/14/check/gate-grenade-dsvm-ironic-multinode-multitenant-ubuntu-xenial/4e1e9f0/logs/grenade.sh.txt.gz :( | 15:15 |
| TheJulia | I'm definitely not going through NC this weekend | 15:16 |
| *** ccamacho has quit IRC | 15:16 | |
| *** rama_y has joined #openstack-ironic | 15:17 | |
| vdrok | TheJulia: hrm. /me looks | 15:17 |
| *** rpioso has joined #openstack-ironic | 15:19 | |
| rama_y | Hi sambetts, I was able to deploy the bm node via nova | 15:21 |
| rloo | so dumb question, after a patch is reverted, do you re-revert-the-reverted-patch? maybe better if i just submit a new patch? | 15:22 |
| rama_y | sambetts: Ran $nova-manage cell_v2 discover_hosts for the scheduler to discover the node | 15:22 |
| vdrok | TheJulia: I have no idea what's going on there :( this worked a week ago | 15:22 |
| *** bfournie has quit IRC | 15:22 | |
| aNuposic | Hi sambetts, dtantsur: I was able to build tinyipa with py3. Now the final image will have python3 already installed | 15:25 |
| *** aNuposic is now known as anupn | 15:25 | |
| sambetts | vdrok: I keep seeing that on my downstream CI! | 15:25 |
| vdrok | sambetts: what exactly? | 15:25 |
| sambetts | vdrok: /bin/sh: 1: brctl: not found | 15:25 |
| vdrok | sambetts: ah. I was talking about http://logs.openstack.org/82/494582/1/check/gate-grenade-dsvm-ironic-multinode-multitenant-ubuntu-xenial/d072aab/logs/grenade.sh.txt.gz#_2017-08-17_14_54_42_498 | 15:26 |
| sambetts | vdrok: ah :/ | 15:26 |
| vdrok | this patch depends on the one introducing the common_settings, hence I'm not sure what's going on | 15:26 |
| *** jlabarre has quit IRC | 15:29 | |
| *** jlabarre has joined #openstack-ironic | 15:31 | |
| *** logan- has joined #openstack-ironic | 15:31 | |
| *** rcernin has quit IRC | 15:32 | |
| sambetts | vdrok: grenade sources everything from /old/ so unless common_settings is backported your screwed | 15:32 |
| sambetts | :P | 15:32 |
| vdrok | sambetts: here is the pike change https://review.openstack.org/494582. it depends on the one in ocata https://review.openstack.org/480904 that adds common_settings | 15:33 |
| *** bfournie has joined #openstack-ironic | 15:33 | |
| *** e0ne has quit IRC | 15:34 | |
| vdrok | sambetts: another thing I see on that pike grenade job is this http://logs.openstack.org/82/494582/1/check/gate-grenade-dsvm-ironic-multinode-multitenant-ubuntu-xenial/d072aab/logs/devstack-gate-setup-workspace-old.txt.gz#_2017-08-17_14_51_21_666 in setup-workspace-old and this | 15:34 |
| vdrok | http://logs.openstack.org/82/494582/1/check/gate-grenade-dsvm-ironic-multinode-multitenant-ubuntu-xenial/d072aab/logs/devstack-gate-setup-workspace-new.txt.gz#_2017-08-17_14_49_23_705 in setup-workspace-new | 15:34 |
| vdrok | does it mean that old is setup from pike and new from master? | 15:34 |
| openstackgerrit | Ruby Loo proposed openstack/ironic master: [reno] Add prelude for Pike (9.1.0) release https://review.openstack.org/494608 | 15:35 |
| vdrok | that would explain the error, as it does not depend on master change | 15:35 |
| sambetts | errr.... | 15:36 |
| *** jaypipes is now known as jaypipes-afk | 15:37 | |
| vdrok | do we need to update something in project config after we cut the branch? | 15:37 |
| rloo | vdrok: that reminde me, we should look at https://docs.openstack.org/ironic/latest/contributor/releasing.html#things-to-do-after-releasing | 15:38 |
| sambetts | vdrok: I didn't think so... | 15:38 |
| rloo | vdrok: update the ironic devstack plugin to point at the branched tarball for IPA? | 15:38 |
| vdrok | rloo: in that bit it's not about ipa. devstack itself is setup from pike as old release and master in new | 15:39 |
| vdrok | while it should be ocata for old and pike for new | 15:39 |
| vdrok | I'll ask in infra I guess | 15:39 |
| sambetts | dafuq... why is grenade on stable/pike base devstack == stable/pike... | 15:39 |
| rloo | oh :-( | 15:39 |
| sambetts | http://logs.openstack.org/82/494582/1/check/gate-grenade-dsvm-ironic-multinode-multitenant-ubuntu-xenial/d072aab/logs/grenade_localrc.txt.gz | 15:39 |
| sambetts | that is wrong | 15:39 |
| rloo | vdrok: so in stable/ocata, we have devstack with newton as old and ocata as new? | 15:39 |
| vdrok | rloo: yes | 15:40 |
| vdrok | sambetts: rloo maybe the .gitreview is the reason? | 15:40 |
| *** logan- has quit IRC | 15:40 | |
| sambetts | vdrok: it seems like this check is failing https://github.com/openstack-infra/devstack-gate/blob/master/devstack-vm-gate-wrap.sh#L334 | 15:40 |
| *** logan- has joined #openstack-ironic | 15:40 | |
| rloo | yeah, i think i just +A'd the .gitreview change | 15:41 |
| *** Nisha_Agarwal has joined #openstack-ironic | 15:41 | |
| *** wong has joined #openstack-ironic | 15:43 | |
| vdrok | rloo: ok, let's wait for logs from the grenade on that change | 15:44 |
| *** ccamacho has joined #openstack-ironic | 15:44 | |
| *** ccamacho has quit IRC | 15:44 | |
| vdrok | because in my patch I indeed see http://logs.openstack.org/82/494582/1/check/gate-grenade-dsvm-ironic-multinode-multitenant-ubuntu-xenial/d072aab/logs/devstack-gate-setup-host.txt.gz, GRENADE_NEW_BRANCH: master and GRENADE_OLD_BRANCH: stable/pike | 15:45 |
| rloo | vdrok: here's the patch, looks like it is running the grenade one now... https://review.openstack.org/#/c/494578/ | 15:46 |
| vdrok | rloo: yeah I know. but you can't see the logs you need until the run finishes | 15:46 |
| rloo | vdrok: right. also, if this is the reason, we should remember to document that | 15:47 |
| sambetts | yeah.... it appears that so https://github.com/openstack-infra/devstack-gate/blob/master/devstack-vm-gate-wrap.sh#L334 should use ZUUL_BRANCH as the GRENADE_BASE_BRACNH and ZUUL_BRANCH should equal the patch branch | 15:47 |
| rloo | i mean, document that after creating branch, need .gitreview and whatever, before ci works. | 15:47 |
| vdrok | rloo: yup. I'll do a patch if that's the case | 15:47 |
| TheJulia | .gitreview is only for the git-review plugin | 15:47 |
| *** wong has quit IRC | 15:47 | |
| sambetts | vdrok: errr... https://review.openstack.org/#/c/494582/ this patch is against master ... not stable/pike | 15:48 |
| TheJulia | upper-constraints on the other hand, is kind of important | 15:48 |
| vdrok | sambetts: aaah | 15:48 |
| vdrok | omg I'm dumb | 15:48 |
| *** rama_y has quit IRC | 15:48 | |
| vdrok | topic stable/pike though, close enough :D | 15:48 |
| sambetts | hahahaha yup | 15:48 |
| sambetts | I was like how can the gate be doing something so wrong | 15:49 |
| sambetts | :-P | 15:49 |
| rloo | TheJulia: wrt the stuff 'todo' after creating a branch, were you going to do any of those? i don't know the urgency of them. | 15:49 |
| vdrok | glad that I have not asked infra to look into it :D | 15:49 |
| rloo | TheJulia: wrt https://docs.openstack.org/ironic/latest/contributor/releasing.html#things-to-do-after-releasing | 15:49 |
| rloo | vdrok: :) | 15:49 |
| TheJulia | rloo: some of that is now automagical, some of the rest of it seems non-critical with exception to the sen-ver flat which I was thinking could go on the reno | 15:51 |
| TheJulia | since we have to bump the minor on the stable branch as well | 15:51 |
| rloo | TheJulia: ok, will leave it for you to deal with :) | 15:52 |
| rloo | TheJulia: or to pass the buck to dmitry | 15:52 |
| TheJulia | Yeah, first a working ci | 15:52 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic stable/pike: Get rid of sourcing stackrc in grenade settings https://review.openstack.org/494617 | 15:53 |
| vdrok | now, that's pike | 15:53 |
| sambetts | \o/ | 15:53 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic master: Get rid of sourcing stackrc in grenade settings https://review.openstack.org/480905 | 15:54 |
| *** udesale has joined #openstack-ironic | 15:55 | |
| *** udesale has quit IRC | 15:58 | |
| *** udesale has joined #openstack-ironic | 15:58 | |
| *** trown is now known as trown|lunch | 16:00 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic master: Adds 9.0 to release_mappings https://review.openstack.org/494620 | 16:01 |
| *** patchbot has joined #openstack-ironic | 16:02 | |
| *** kromanenko has quit IRC | 16:05 | |
| *** srobert has quit IRC | 16:05 | |
| rpioso | Will changes that couldn't get beyond the gate breakage be considered for inclusion in Pike after the gate is fixed? | 16:06 |
| *** marios has quit IRC | 16:06 | |
| sambetts | rpioso: if they were already considered for the pike release yes | 16:07 |
| rloo | rpioso: we're planning on the official pike release being 9.1, not 9.0 | 16:08 |
| sambetts | rpioso: the patches we were trying to get in hasn't changed, it just we've had to branch and do an initial release in order to make the CI work as intended | 16:08 |
| *** rcernin has joined #openstack-ironic | 16:08 | |
| sambetts | also what rloo said | 16:08 |
| vdrok | rloo: I'm talking about that pin_release_version set to pike in grenade upgrade.sh | 16:09 |
| rpioso | rloo, sambetts: Awesome! Thank you. | 16:09 |
| vdrok | it's not number like 9.0 :( | 16:09 |
| rloo | vdrok: yup, just saw your comment. will do another patch for that. | 16:09 |
| TheJulia | vdrok: c'est la vie. | 16:09 |
| vdrok | :) | 16:09 |
| TheJulia | the fact that 494578 has passed the two multinode jobs gives me hope | 16:13 |
| TheJulia | one of which being multinode grenade | 16:13 |
| NobodyCam | Good Morning Ironic folks... | 16:13 |
| sambetts | TheJulia: \o/ | 16:13 |
| TheJulia | good morning NobodyCam | 16:14 |
| NobodyCam | Morning TheJulia :) | 16:14 |
| vdrok | morning NobodyCam | 16:15 |
| *** srobert has joined #openstack-ironic | 16:16 | |
| vdrok | TheJulia: we're lucky, as it started before ocata devstack change was merged :) but when it gets to gate, it should be all ok | 16:16 |
| NobodyCam | Good morning vdrok :) | 16:16 |
| * TheJulia ponders going to the store | 16:19 | |
| rloo | TheJulia: go! | 16:19 |
| openstackgerrit | Ruby Loo proposed openstack/ironic master: Adds 9.0 to release_mappings https://review.openstack.org/494620 | 16:21 |
| rloo | vdrok: ^^ am I correct that master CI will fail w/o that? | 16:21 |
| vdrok | rloo: I think so, yes | 16:21 |
| sambetts | rloo: +1 | 16:22 |
| rloo | vdrok: so we need to merge that first, if it passes... do we have other dependencies, hope there isn't a chicken/egg somewhere... | 16:22 |
| * jlvillal reading backscroll and whiteboard to figure out status of the gate and why | 16:22 | |
| sambetts | jlvillal: https://etherpad.openstack.org/p/IronicWhiteBoard L120 | 16:23 |
| vdrok | rloo: we need https://review.openstack.org/493990 for master ci | 16:23 |
| patchbot | patch 493990 - openstack-dev/devstack (stable/pike) - Increase host_subset_size for ironic | 16:23 |
| * jlvillal is basically getting the idea that Ironic got hosed by changes to other projects... | 16:23 | |
| vdrok | but we may be lucky and won't need it :) | 16:23 |
| rloo | jlvillal: more like, when others created stable/pike and we didn't. | 16:23 |
| jlvillal | sambetts: Thanks | 16:23 |
| sambetts | jlvillal: basically we got hosed because we havn't cut a branch yet | 16:23 |
| *** rama_y has joined #openstack-ironic | 16:23 | |
| rloo | jlvillal: that link somewhere to the dev email from ocata cycle summarizes it i think. | 16:23 |
| jlvillal | sambetts: rloo: Ah okay. So for future we should go before everyone else ;) | 16:23 |
| sambetts | +1 we need to cut an inital release at the same time as all the other projects cut a branch | 16:24 |
| sambetts | and then backport everything we merge into master into it | 16:24 |
| sambetts | keeping them in sync for a period of time until we're ready to cut our offical stable/pike release | 16:25 |
| vdrok | in this case I'm not sure what's the benefit of having the release model we have. just to be able to do intermediate releases? | 16:25 |
| vdrok | does anyone use them? | 16:25 |
| sambetts | I don't know if anyone does tbh | 16:25 |
| jlvillal | vdrok: I think one of the other benefits was that we wouldn't need to feature freeze as long. | 16:26 |
| jlvillal | Or I vaguely think that was one of the reasons. | 16:26 |
| vdrok | jlvillal: I think even if we have integrated release, we ourselves decide freezes and stuff | 16:26 |
| sambetts | I'm not sure I thought it was dictated | 16:26 |
| vdrok | https://releases.openstack.org/pike/schedule.html | 16:27 |
| rloo | i didn't like the release model we have, didn't think it'd really work but anyway... | 16:27 |
| vdrok | here for example nova spec freeze and cinder spec freeze are different | 16:27 |
| rloo | i updated the white board with our dependencies for this release... | 16:28 |
| sambetts | tbh I think we have the release model we have because we went sem-ver before everyone else | 16:28 |
| sambetts | now the whole problem is semver it might make sense to revisit it | 16:28 |
| vdrok | oh, right, feature freeze is for all it seems | 16:28 |
| sambetts | s/prolem/project | 16:28 |
| vdrok | but anyway, there are ffes | 16:29 |
| rloo | sambetts: if i recall, there were two main reasons: wanted to release more frequently so folks wouldn't have to wait 6+ months to get their feature 'in' a release; reduce stress on us at release time, since features can land any time in cycle. | 16:29 |
| *** udesale__ has joined #openstack-ironic | 16:29 | |
| vdrok | rloo: the stress does not seem to leave :) | 16:29 |
| rloo | sambetts: i don't think in reality, that has happened yet ^^. although i would say there seems to be less stress at final release time. | 16:29 |
| *** udesale has quit IRC | 16:29 | |
| jlvillal | vdrok: Xanax will help ;) | 16:30 |
| rloo | vdrok: stress seems more spread out though. or maybe we are better at managing expectations, saying earlier that feature X won't make it. | 16:30 |
| *** udesale__ has quit IRC | 16:30 | |
| *** udesale has joined #openstack-ironic | 16:30 | |
| sambetts | rloo: yeah... I think we need to revisit it ... I think we're just better at managing expectations | 16:30 |
| rloo | gate issues though seem to follow the other release model; backed up gates etc. | 16:30 |
| TheJulia | I'd love to see us release more often, I really feel like this release crunch is just made so much worse because it is not a regular like clockwork thing for us, but out dependencies are so intertwined with testing... i dunno. | 16:30 |
| jlvillal | I'm not sure if the gate has been worse for us this cycle, or at least worse for us these last four weeks. | 16:31 |
| vdrok | jlvillal: :) | 16:31 |
| TheJulia | jlvillal: worst since it now has less capacity :( | 16:31 |
| rloo | jlvillal: i think gate gets worse at crunch time for other projects cuz they want to get their features in, and some of that breaks our stuff. | 16:31 |
| * sambetts has to run | 16:32 | |
| rloo | bye sambetts, thx for helping out today! | 16:32 |
| vdrok | good night sambetts | 16:32 |
| jlvillal | Thanks sambetts ! | 16:32 |
| TheJulia | goodnight sambetts | 16:32 |
| sambetts | no problem :) night all, cya tomorrow for another rollercoaster | 16:32 |
| *** r-daneel has joined #openstack-ironic | 16:32 | |
| rloo | TheJulia: do you have time/want to send out email on devlist about the status of the release? | 16:32 |
| rloo | TheJulia: err why we branched etc? | 16:33 |
| *** sambetts is now known as sambetts|afk | 16:33 | |
| rloo | TheJulia: or I can do it, cuz you're going to the store, right? :) | 16:33 |
| TheJulia | I kind of see sending out such an email as pointless at the moment I'd rather start a postmortem discussion | 16:34 |
| rloo | TheJulia: ok. just that people that don't read irc won't know what's going on. | 16:34 |
| TheJulia | or maybe an email once we're done, and then detail why. | 16:34 |
| TheJulia | also, many don't read the list | 16:34 |
| priteau | Hello. I am researching how to use the multitenant network support in Ocata and I am puzzled by step 5 of https://docs.openstack.org/project-install-guide/baremetal/draft/configure-tenant-networks.html. How can I find out which ML2 drivers support bare metal? | 16:35 |
| rloo | TheJulia: eg, someone asked earlier about what was going on. | 16:35 |
| TheJulia | Okay, I'll get a beer and start drafting an email in a minute | 16:36 |
| * TheJulia never got a chance to leave for the store... and it can wait | 16:36 | |
| rloo | TheJulia: thx. i think it can be short, we were going to do what we hoped was official pike release today but due to grenade/CI issues had to cut the 9.0 branch... | 16:38 |
| TheJulia | it will be fairly short :) | 16:38 |
| rloo | TheJulia: and link to whiteboard :) | 16:40 |
| *** jtomasek_ has quit IRC | 16:41 | |
| *** Nisha_Agarwal has quit IRC | 16:42 | |
| priteau | The most interesting information I found so far is this mailing list post which says to use implement support in networking_generic_switch: http://lists.openstack.org/pipermail/openstack-operators/2017-January/012467.html | 16:47 |
| *** udesale has quit IRC | 16:50 | |
| *** rbartal has joined #openstack-ironic | 16:51 | |
| *** derekh has quit IRC | 16:52 | |
| priteau | Oh, there is support for various switches in networking-generic-switch | 16:53 |
| *** lucasagomes is now known as lucas-afk | 16:54 | |
| TheJulia | priteau: I think it really depends on your situation and needs | 16:56 |
| TheJulia | for example, i used networking_generic_switch for a live demo, and used it in my home lab for a while | 16:56 |
| *** amarao has quit IRC | 16:57 | |
| TheJulia | some of the ml2 drivers that exist also work using some sort of SDN fabric in between the switch and neutron, so that may also complicate things, so you need to evaluate your network architecture, and evaluate what ml2 driver options exist for the hardware you have. | 16:57 |
| priteau | Our switches are Dell S6000. I am finding various bits of ML2 driver for Dell hardware, but nothing that seems mature / up to date. Maybe the S4810 support of networking_generic_switch is close enough for the S6000 | 17:00 |
| priteau | Does networking_generic_switch support multiple ToR switches? I see the switch_id/switch_info values in the ironic port config, should I just declare multiple "[genericswitch:dell-hostname]" entries in /etc/neutron/plugins/ml2/ml2_conf_genericswitch.ini? | 17:03 |
| *** slagle has joined #openstack-ironic | 17:09 | |
| vdrok | good night everyone! | 17:11 |
| fellypefca | night vdrok | 17:11 |
| *** rbartal has quit IRC | 17:14 | |
| TheJulia | priteau: should be support wise, multiple ToR... I think but I've honestly never tried it. | 17:14 |
| TheJulia | Now, if dell would let me visit a lab again with a bunch of switches and available baremetal servers, I could verify it. :) | 17:15 |
| rloo | bye vdrok, thx for helping with release! | 17:16 |
| TheJulia | goodnight vdrok | 17:16 |
| anupn | night vdrok | 17:16 |
| rpioso | TheJulia: Would you like access? I can try to arrange that. | 17:16 |
| rpioso | nite vdrok | 17:17 |
| TheJulia | rpioso: I wouldn't mind stopping by austin for a day or two at some point, it was kind of fun before the last summit :) Sadly, my schedule is booked until summit timeframe | 17:18 |
| rpioso | TheJulia: ty for the ML announcement. Very helpful. | 17:18 |
| rpioso | TheJulia: Just let us know. | 17:19 |
| *** trown|lunch is now known as trown | 17:20 | |
| TheJulia | Yeah, as much as I dislike the mailing list, I guess people are a little nervous that we cut a stable/pike branch at the moment | 17:20 |
| priteau | TheJulia: Thanks for the info, that's very helpful | 17:20 |
| TheJulia | rpioso: will do | 17:20 |
| rpioso | TheJulia: yep, and awesome :) | 17:21 |
| *** rpioso is now known as rpioso|eat | 17:23 | |
| *** srobert has quit IRC | 17:41 | |
| *** edleafe has quit IRC | 17:55 | |
| *** edleafe has joined #openstack-ironic | 17:56 | |
| *** srobert has joined #openstack-ironic | 17:56 | |
| *** dprince has quit IRC | 18:00 | |
| *** jkilpatr has quit IRC | 18:02 | |
| *** jkilpatr has joined #openstack-ironic | 18:03 | |
| *** jcoufal has quit IRC | 18:05 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/pike: Adds 9.0 to release_mappings https://review.openstack.org/494662 | 18:08 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic master: Finalize migration to keystoneauth adapters https://review.openstack.org/478825 | 18:10 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic master: Use adapters for cinderclient https://review.openstack.org/476171 | 18:10 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic master: Use adapters for neutronclient https://review.openstack.org/476170 | 18:10 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic master: Rework keystone auth for glance https://review.openstack.org/467728 | 18:10 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic master: Fix swiftclient creation https://review.openstack.org/476173 | 18:10 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic master: Use adapters for inspectorclient https://review.openstack.org/476172 | 18:10 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic master: Introduce keystoneauth adapters for clients https://review.openstack.org/469627 | 18:10 |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/pike: Update UPPER_CONSTRAINTS_FILE for stable/pike https://review.openstack.org/494579 | 18:11 |
| *** etingof has quit IRC | 18:11 | |
| TheJulia | rloo: https://review.openstack.org/494662 when you have a minute, you might want to bless the back port. :) | 18:11 |
| patchbot | patch 494662 - ironic (stable/pike) - Adds 9.0 to release_mappings | 18:11 |
| rloo | TheJulia: already did, thx! :) | 18:12 |
| TheJulia | rloo: reason for it is the coverage test on 494579 failed | 18:12 |
| rloo | TheJulia: yeah, my bad; I should have gotten that in before the branching, but there was no way to get that in even if i had had pushed that patch yesterday :-( | 18:13 |
| rloo | TheJulia: at least, *now* I know why we need to make those changes before doing a release! | 18:14 |
| TheJulia | doh :) | 18:15 |
| TheJulia | yeah, complete horrible catch-22 kind of situation | 18:15 |
| *** dprince has joined #openstack-ironic | 18:16 | |
| TheJulia | I suspect the grenade setting changes will fail but on unit tests for the release mapping | 18:16 |
| rloo | TheJulia: still need those devstack backports to merge | 18:16 |
| TheJulia | those back ports won't merge without working unit tests | 18:16 |
| TheJulia | at least on stable/pike | 18:16 |
| TheJulia | stable/ocata should be fine | 18:17 |
| TheJulia | we should review that change | 18:17 |
| rloo | TheJulia: oh. argh. | 18:17 |
| *** cdearborn has quit IRC | 18:17 | |
| *** jcoufal has joined #openstack-ironic | 18:17 | |
| rloo | TheJulia: the backport to stable/ocata will let ci on stable/pike work. we need the stable/pick backport to work, to get master CI working. | 18:17 |
| rloo | TheJulia: i see a circle... | 18:17 |
| TheJulia | wait | 18:18 |
| TheJulia | https://review.openstack.org/#/c/480904/ | 18:18 |
| patchbot | patch 480904 - ironic (stable/ocata) - Get rid of sourcing stackrc in grenade settings | 18:18 |
| TheJulia | Already has a +1 from jenkins | 18:18 |
| TheJulia | err | 18:18 |
| TheJulia | CI | 18:18 |
| rloo | TheJulia: that's ocata... | 18:19 |
| TheJulia | yes, we should be able to see if that helps the grenade stuff in | 18:19 |
| TheJulia | https://review.openstack.org/#/c/494617/ | 18:20 |
| patchbot | patch 494617 - ironic (stable/pike) - Get rid of sourcing stackrc in grenade settings | 18:20 |
| rloo | TheJulia: I should review 480904 too... | 18:21 |
| * TheJulia wonders if we should make a chart of landing order of operations | 18:21 | |
| rloo | TheJulia: i tried in the whiteboard... | 18:21 |
| *** dtrainor has joined #openstack-ironic | 18:27 | |
| TheJulia | rloo: I expanded the list a little bit based upon the discussions and all :( | 18:34 |
| TheJulia | rloo: but I did highlight the unknowns as "possibly" | 18:35 |
| *** etingof has joined #openstack-ironic | 18:35 | |
| rloo | TheJulia: thx! | 18:36 |
| rloo | TheJulia: we can update as we find out more info... | 18:36 |
| rloo | TheJulia: i suspect that we will have to approve 494662 before 494620 (adds 9.0 to release mappings, stable/pike before master) | 18:37 |
| *** rpioso|eat has quit IRC | 18:37 | |
| TheJulia | rloo: indeed :( | 18:38 |
| *** ChubYann has joined #openstack-ironic | 18:39 | |
| rloo | TheJulia: i am not sure what to do about bumping the semver. wondering if we need to do it twice on master. maybe a question for infra. i mean we def need to do it in branch (and I hope they allow this, what if they don't...) | 18:40 |
| TheJulia | oh, we can definitely do it in branch | 18:41 |
| *** jlabarre has quit IRC | 18:41 | |
| TheJulia | I can look at pbr in the morning since I think thats where that headache originates | 18:41 |
| rloo | TheJulia: hey, look what vdrok said in https://review.openstack.org/#/c/494620/!!! | 18:42 |
| patchbot | patch 494620 - ironic - Adds 9.0 to release_mappings | 18:42 |
| *** jlabarre has joined #openstack-ironic | 18:42 | |
| * rloo hopes... | 18:42 | |
| TheJulia | that is true, it passed and wouldn't have otherwise | 18:43 |
| TheJulia | i have status.o.o/zuul up tracking 494662 which is the back port of that rev | 18:45 |
| TheJulia | 494579,2 should tell us as its coverage and unit tests will implode without that patch | 18:46 |
| TheJulia | the other one to watch is 494617,1 as it's grenade job tells us if 480904 can be merged | 18:49 |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/pike: Get rid of sourcing stackrc in grenade settings https://review.openstack.org/494617 | 18:56 |
| TheJulia | ^^^ to rebase the job for the release mapping | 18:56 |
| TheJulia | since it was failing on that | 18:56 |
| TheJulia | but 1 out of 2 grenade tests passed, so good sign for the stable/ocata revision | 18:57 |
| rloo | TheJulia: thx for keeping an eye out... | 18:58 |
| *** rpioso|eat has joined #openstack-ironic | 18:59 | |
| *** baoli has quit IRC | 19:00 | |
| TheJulia | i really wish there was kill job functionality.... | 19:02 |
| dtrainor | Howdy. I'm having issue introspecting nodes. I can import them just fine (openstack overcloud node import instackenv.json), power status gets returned (off), but when I try to actually start the introspection (openstack overcloud node introspect --all-manageable --provide), I don't see any ipmi commands made by ironic to my bmc (virtual ipmi device, virtualized network). I can run the same ipmitool command agains the nodes | 19:02 |
| dtrainor | that ironic-conductor.log reports sending, and the power state then gets modified and introspection happens as it normally would, but trying to power the nodes on during introspection is not successful. | 19:02 |
| dtrainor | I can run 'ironic set-node-power-state' and the node does power on, too. Just... not part of introspection | 19:03 |
| dtrainor | fwiw I'm using a very recent osp puddle for 12, figured i need to disclose that | 19:03 |
| TheJulia | ops puddle? | 19:03 |
| TheJulia | err, osp | 19:03 |
| TheJulia | since autocorrect hates me | 19:04 |
| dtrainor | sorry, red hat openstack platform engineering build, product of a few days old rod build | 19:04 |
| *** srobert has quit IRC | 19:04 | |
| TheJulia | off of master? | 19:04 |
| dtrainor | it's cut from pike | 19:07 |
| TheJulia | so... that is a problem | 19:08 |
| *** baoli has joined #openstack-ironic | 19:08 | |
| TheJulia | ironic-inspector's pike release was literally cut this morning. ironic.... while we have created a pike branch just hours ago, because our release model is different, and a whole host of other issues, our stable/pike branch is broken. On a plus side, we're working to fix it as quickly as CI will allow us to. | 19:10 |
| TheJulia | so, that being said, depending on how that build is built, it might not actually be correct yet | 19:10 |
| dtrainor | oh, you know what, i saw that email this morning | 19:10 |
| dtrainor | understood. | 19:10 |
| dtrainor | i'll chase that thread, thanks for the reminder | 19:10 |
| TheJulia | if ironic and ironic-inspector were built from master, then they should work | 19:10 |
| dtrainor | duh. i'm sorry. too much context switching. | 19:11 |
| TheJulia | dtrainor: no worries | 19:11 |
| dtrainor | because, you know, you sent the email, so.... | 19:11 |
| dtrainor | heh | 19:11 |
| dtrainor | i might just take a ci build and slip it in by hand and see how far i get | 19:12 |
| dtrainor | hmm zuul says it might be a while | 19:13 |
| TheJulia | yeah | 19:13 |
| dtrainor | cool. well again i really appreciate it, thanks | 19:14 |
| TheJulia | Speaking of zuul, looks like 5 hours to go | 19:14 |
| dtrainor | woohoo | 19:15 |
| dtrainor | i'll find something else to dig in to, in the mean time | 19:15 |
| TheJulia | cool | 19:15 |
| *** efried has quit IRC | 19:18 | |
| *** srobert has joined #openstack-ironic | 19:24 | |
| *** efried has joined #openstack-ironic | 19:24 | |
| *** mgoddard has quit IRC | 19:32 | |
| TheJulia | rloo, going and running an errand (and finally going to the store. I should be back in ~1.5 hours. if everything touching 494662 goes well, then we should be able to begin back porting all the stuff for master branch *crosses fingers* | 19:38 |
| rloo | :) | 19:38 |
| TheJulia | we may want to consider a 9.0.1 release with "don't use 9.0" as the highlight :) | 19:39 |
| rloo | TheJulia: oh... with 'Fixes... ' :) Will think about it... | 19:40 |
| rloo | TheJulia: I think it'll work for people; just that it fails our tests... | 19:40 |
| TheJulia | I guess yeah | 19:42 |
| *** athomas has quit IRC | 19:42 | |
| rloo | TheJulia: oh, wait, to package it, it has to pass tests, so... | 19:43 |
| rloo | TheJulia: won't hurt, so I'll push something up. | 19:43 |
| *** baoli has quit IRC | 19:43 | |
| TheJulia | ok, bbl | 19:44 |
| *** baoli has joined #openstack-ironic | 19:45 | |
| *** baoli has quit IRC | 19:45 | |
| *** yangyape_ has joined #openstack-ironic | 19:46 | |
| *** trown is now known as trown|brb | 19:46 | |
| *** yangyapeng has quit IRC | 19:47 | |
| *** e0ne has joined #openstack-ironic | 19:47 | |
| *** baoli has joined #openstack-ironic | 19:53 | |
| *** srobert has quit IRC | 19:56 | |
| *** trown|brb is now known as trown | 19:59 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic master: devstack: Comment variables related to multi-tenant networking https://review.openstack.org/494690 | 20:06 |
| *** jkilpatr has quit IRC | 20:09 | |
| *** e0ne has quit IRC | 20:12 | |
| *** rcernin has quit IRC | 20:16 | |
| TheJulia | rloo: take a look at 494579 | 20:19 |
| rloo | TheJulia: hey! | 20:20 |
| *** adisky__ has quit IRC | 20:20 | |
| rloo | TheJulia: should we just +A 494662 now, instead of waiting? | 20:20 |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/pike: Update .gitreview for stable/pike https://review.openstack.org/494578 | 20:21 |
| rloo | TheJulia: are you back from your errands, cuz... I have some questions about reno | 20:21 |
| TheJulia | rloo: ^^^ rebase to get around same test failures, I think we should +A 494662 at this point | 20:22 |
| TheJulia | go ahead and ask, my other half delayed departure | 20:22 |
| rloo | TheJulia: ok | 20:22 |
| rloo | TheJulia: sec | 20:22 |
| openstackgerrit | Ruby Loo proposed openstack/ironic master: [reno] Add prelude for intermediate Pike 9.0.1 release https://review.openstack.org/494693 | 20:24 |
| rloo | TheJulia: ^^ | 20:24 |
| rloo | TheJulia: so... I don't know how the renos are done. just that they are published from master, right? | 20:25 |
| rloo | TheJulia: so how do we add a reno for something in stable/pike only? | 20:25 |
| rloo | TheJulia: I mean, we have a patch that fails now, in master, to add reno for pike. sec, let me find it. | 20:25 |
| TheJulia | so I believe the fact that it gets tagged will cause it to fall out, but I was going to check the code tomorrow | 20:26 |
| rloo | TheJulia: https://review.openstack.org/#/c/494580/ | 20:26 |
| patchbot | patch 494580 - ironic - Update reno for stable/pike | 20:26 |
| TheJulia | it may need to land in the branch first, since the id on the slug name is hunted across branches | 20:26 |
| rloo | TheJulia: where will the reno with the prelude for 9.1.0 get put? | 20:26 |
| rloo | TheJulia: under 'pike' or under 'current'? | 20:27 |
| rloo | TheJulia: I guess same question for the 9.0.1 patch i just added, but I'm more concerned with the official 9.1.0 patch-prelude. | 20:28 |
| *** jkilpatr has joined #openstack-ironic | 20:30 | |
| TheJulia | i would cherry-pick it down to have the update reno patch, and see what it does, I won't +2 or +a anything until I can dig through reno's logic | 20:30 |
| rloo | TheJulia: ok, i can try that. but not cherry pick, i'll need to make it dependent on the stable/pike 9.0-to-release-mapping patch. i'll just WIP it. | 20:31 |
| TheJulia | one moment | 20:31 |
| TheJulia | oh right, the reno change is on master | 20:32 |
| rloo | TheJulia: yeah. i think it ties in with my muddy understanding of not changing renos once we've branched. | 20:33 |
| * TheJulia pulls up the reno source code | 20:33 | |
| *** hoonetorg has quit IRC | 20:36 | |
| TheJulia | hmmmmmmmm | 20:39 |
| *** rama_y has quit IRC | 20:42 | |
| *** srobert has joined #openstack-ironic | 20:43 | |
| TheJulia | well I'm going cross eyed... | 20:43 |
| TheJulia | do you have the rev number for the reno index changes? | 20:44 |
| rloo | TheJulia: you mean this thing? https://review.openstack.org/#/c/494580/ | 20:44 |
| patchbot | patch 494580 - ironic - Update reno for stable/pike | 20:44 |
| TheJulia | yeah, building locally | 20:45 |
| TheJulia | okay, if it is on the branch, it shows up under pike as well. | 20:50 |
| TheJulia | cutting a tag locally | 20:50 |
| TheJulia | rebuilding | 20:51 |
| rloo | TheJulia: I just looked at our release notes, in current, there is the same bug fix as in ocata 7.0.1. | 20:51 |
| TheJulia | ugh, i bet because we've mucked with the history | 20:51 |
| rloo | TheJulia: current 8.0.0 release | 20:51 |
| rloo | TheJulia: so i think that we should *only* add the prelude reno to pike, not to master... | 20:52 |
| TheJulia | or a tag is missing | 20:52 |
| rloo | or yeah, something else. but it makes sense. if we fix a bug in master, we add a reno to that. and if we backport&release, you'd want to see the bug reno there too. | 20:52 |
| TheJulia | maybe not as a prelude for 9.0.1? | 20:53 |
| rloo | TheJulia: i am assuming that things are consistent, same for preludes or anything in a reno file, but dunno. | 20:53 |
| *** heroux has quit IRC | 20:53 | |
| TheJulia | yeah, backporting it definitely shows up as expected | 20:53 |
| rloo | TheJulia: all I know is that we don't want those preludes to show up in 'current'... | 20:54 |
| TheJulia | rebuilding my release notes again | 20:55 |
| TheJulia | hmmm shows up on my main unreleased... | 20:58 |
| *** trown is now known as trown|outtypewww | 21:02 | |
| TheJulia | rloo: lets hold off on landing them, I do actually need to head out now. but I'm working on it locally to figure it out with reno | 21:03 |
| rloo | TheJulia: ok, i'll probably be gone after you get back. might want to ping dougH. | 21:03 |
| rloo | TheJulia: I am thinking now, that maybe we wanted that prelude thing when we branched, but sigh. | 21:04 |
| TheJulia | crcrossing versions, they preludes are unique, it is just we will have to bump master's semver flag 2x I think | 21:04 |
| TheJulia | to get us to 9.2.x | 21:04 |
| TheJulia | I'll try when I get back | 21:04 |
| rloo | so far, yay: https://review.openstack.org/#/c/494620/ (on master) | 21:05 |
| patchbot | patch 494620 - ironic - Adds 9.0 to release_mappings | 21:05 |
| *** rama_y has joined #openstack-ironic | 21:05 | |
| *** mat128 has quit IRC | 21:05 | |
| TheJulia | woot | 21:06 |
| *** srobert has quit IRC | 21:06 | |
| *** dprince has quit IRC | 21:19 | |
| *** baoli has quit IRC | 21:22 | |
| *** slagle has quit IRC | 21:28 | |
| *** jcoufal has quit IRC | 21:32 | |
| rloo | so far, yay: https://review.openstack.org/#/c/494662/ (on stable/pike) | 21:32 |
| patchbot | patch 494662 - ironic (stable/pike) - Adds 9.0 to release_mappings | 21:32 |
| *** priteau has quit IRC | 21:38 | |
| *** oanson has quit IRC | 21:42 | |
| *** oanson has joined #openstack-ironic | 21:44 | |
| *** bfournie has quit IRC | 22:04 | |
| openstackgerrit | Merged openstack/ironic-inspector stable/pike: Update .gitreview for stable/pike https://review.openstack.org/494560 | 22:04 |
| openstackgerrit | Merged openstack/ironic-inspector stable/pike: Update UPPER_CONSTRAINTS_FILE for stable/pike https://review.openstack.org/494561 | 22:05 |
| rloo | nice that some things are working... ^^ | 22:06 |
| *** jlabarre has quit IRC | 22:11 | |
| *** dyasny has joined #openstack-ironic | 22:21 | |
| *** mtanino has joined #openstack-ironic | 22:33 | |
| *** openstackgerrit has quit IRC | 22:33 | |
| *** bfournie has joined #openstack-ironic | 22:36 | |
| *** bfournie has quit IRC | 22:37 | |
| *** Nisha_Agarwal has joined #openstack-ironic | 22:38 | |
| *** Nisha_Agarwal has quit IRC | 22:43 | |
| TheJulia | :) | 22:47 |
| *** openstackgerrit has joined #openstack-ironic | 22:49 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic master: devstack: Remove unused variable IRONIC_VM_NETWORK_RANGE https://review.openstack.org/494731 | 22:49 |
| *** yangyape_ has quit IRC | 22:52 | |
| TheJulia | rloo: so, if we propose the two release notes exclusively to stable/pike, they will render and appear for pike | 23:07 |
| *** yangyapeng has joined #openstack-ironic | 23:07 | |
| TheJulia | we _will_ want to still "Sem-Ver: feature" master for pbr builds, but for some unknown reason, reno doesn't pickup on it | 23:08 |
| TheJulia | that is the big unknown, how to get it to pickup on that | 23:08 |
| *** dyasny has quit IRC | 23:10 | |
| *** yangyapeng has quit IRC | 23:12 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/pike: Fix DRAC classic driver double manage/provide https://review.openstack.org/494736 | 23:14 |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/pike: Add 'hardware type' for Dell EMC iDRACs https://review.openstack.org/494737 | 23:14 |
| rloo | TheJulia: ok, the releasenotes on stable/pike make sense. just means we'll never see those files on master but i don't care. | 23:14 |
| rloo | TheJulia: thx for pressing those cherrypick buttons :) | 23:14 |
| TheJulia | I pressed those two, I'll press more after dinner and a movie | 23:14 |
| TheJulia | actually, I could do two more | 23:15 |
| rloo | TheJulia: did you 'only' press the cherry-pick button. i see you also rebased them. | 23:15 |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/pike: Update proliantutils version for Pike release https://review.openstack.org/494738 | 23:17 |
| TheJulia | eh, one more | 23:17 |
| openstackgerrit | Merged openstack/ironic master: Adds 9.0 to release_mappings https://review.openstack.org/494620 | 23:17 |
| TheJulia | rloo: no, a bit more complex. git review -d the upper constraints patch ; git review -X masterbranchchangeid; git review -R stable/pike | 23:18 |
| rloo | TheJulia: ahh, I thought it wasn't that simple | 23:18 |
| rloo | TheJulia: look ^^!! | 23:18 |
| rloo | TheJulia: master CI is good now | 23:18 |
| TheJulia | i know! | 23:18 |
| TheJulia | weI'm tempted to wave a "please don't merge anything" sign until we get the back ports sorted | 23:19 |
| TheJulia | s/weI/I/ | 23:19 |
| rloo | TheJulia: you've -2'd everything didn't you? | 23:19 |
| TheJulia | not _everything_, only the 10 patches or so that we have been struggling to merge | 23:19 |
| rloo | TheJulia: well, who's approving the other patches that haven't yet been approved? ;) | 23:20 |
| TheJulia | true | 23:20 |
| TheJulia | anyway, dinner, movie, I'll fire off some more cherry picks down to the stable branch later tonight | 23:20 |
| TheJulia | hopefully these should all pass just fine | 23:20 |
| rloo | TheJulia: thx! | 23:21 |
| rloo | TheJulia: or wait til the upper constraints patch lands before cherry-picking-with-button :) | 23:22 |
| TheJulia | heh | 23:22 |
| *** rpioso|eat has quit IRC | 23:25 | |
| *** pc_m has quit IRC | 23:27 | |
| *** pc_m has joined #openstack-ironic | 23:28 | |
| *** rama_y has quit IRC | 23:32 | |
| openstackgerrit | Merged openstack/ironic stable/pike: Adds 9.0 to release_mappings https://review.openstack.org/494662 | 23:32 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
