| *** sacharya has joined #openstack-ironic | 00:23 | |
| *** gabriel-bezerra has quit IRC | 00:23 | |
| *** gabriel-bezerra has joined #openstack-ironic | 00:24 | |
| *** sacharya has quit IRC | 00:29 | |
| *** Guest460 is now known as mfisch | 00:32 | |
| *** mfisch is now known as Guest81704 | 00:32 | |
| *** elfosardo_ has joined #openstack-ironic | 00:35 | |
| *** baoli has joined #openstack-ironic | 00:38 | |
| *** elfosardo has quit IRC | 00:38 | |
| *** catintheroof has joined #openstack-ironic | 00:55 | |
| *** baoli has quit IRC | 01:43 | |
| *** Nisha has joined #openstack-ironic | 01:52 | |
| *** Nisha has quit IRC | 01:57 | |
| *** Fdaisuke has joined #openstack-ironic | 01:58 | |
| *** hoangcx has joined #openstack-ironic | 02:07 | |
| *** Haomeng has joined #openstack-ironic | 02:08 | |
| *** sacharya has joined #openstack-ironic | 02:27 | |
| *** amit213 has quit IRC | 02:34 | |
| *** amit213 has joined #openstack-ironic | 02:34 | |
| *** smoriya has joined #openstack-ironic | 02:41 | |
| *** baoli has joined #openstack-ironic | 02:44 | |
| *** baoli has quit IRC | 02:49 | |
| *** Guest81704 is now known as mfisch | 03:03 | |
| *** mfisch is now known as Guest30018 | 03:04 | |
| *** amotoki has joined #openstack-ironic | 03:07 | |
| *** Sukhdev has joined #openstack-ironic | 03:09 | |
| *** amotoki has quit IRC | 03:18 | |
| *** amotoki has joined #openstack-ironic | 03:40 | |
| *** links has joined #openstack-ironic | 03:44 | |
| *** Sukhdev has quit IRC | 03:55 | |
| *** catintheroof has quit IRC | 03:56 | |
| *** fragatina has quit IRC | 03:58 | |
| *** appu103_ has joined #openstack-ironic | 04:01 | |
| *** sacharya has quit IRC | 04:02 | |
| *** appu103_ is now known as aparnav | 04:03 | |
| *** itamarl has quit IRC | 04:05 | |
| *** vmud213 has joined #openstack-ironic | 04:10 | |
| *** fragatina has joined #openstack-ironic | 04:13 | |
| *** fragatina has quit IRC | 04:18 | |
| *** fragatina has joined #openstack-ironic | 04:29 | |
| *** fragatina has quit IRC | 04:29 | |
| *** fragatina has joined #openstack-ironic | 04:30 | |
| *** hoangcx has quit IRC | 04:45 | |
| *** hoangcx has joined #openstack-ironic | 04:45 | |
| *** amotoki_ has joined #openstack-ironic | 04:46 | |
| *** amotoki has quit IRC | 04:49 | |
| *** yonglihe has joined #openstack-ironic | 05:04 | |
| *** ChubYann has quit IRC | 05:29 | |
| *** linuxgeek has quit IRC | 05:29 | |
| *** amotoki_ has quit IRC | 05:35 | |
| *** rcernin has joined #openstack-ironic | 05:40 | |
| *** amotoki has joined #openstack-ironic | 05:51 | |
| *** vinbs has joined #openstack-ironic | 05:54 | |
| *** yolanda has joined #openstack-ironic | 05:58 | |
| *** vinbs_ has joined #openstack-ironic | 06:00 | |
| *** vinbs has quit IRC | 06:01 | |
| *** vinbs_ is now known as vinbs | 06:03 | |
| *** sacharya has joined #openstack-ironic | 06:03 | |
| *** sacharya has quit IRC | 06:08 | |
| *** vinbs has quit IRC | 06:16 | |
| *** vinbs has joined #openstack-ironic | 06:20 | |
| *** vinbs has quit IRC | 06:30 | |
| *** vinbs has joined #openstack-ironic | 06:36 | |
| *** itamarl has joined #openstack-ironic | 06:42 | |
| *** vinbs has quit IRC | 06:42 | |
| *** xek__ is now known as xek | 06:43 | |
| *** openstackgerrit has quit IRC | 06:47 | |
| *** openstackgerrit has joined #openstack-ironic | 06:48 | |
| *** chihhsin has quit IRC | 06:50 | |
| *** tesseract has joined #openstack-ironic | 07:01 | |
| *** sacharya has joined #openstack-ironic | 07:04 | |
| *** fragatina has quit IRC | 07:04 | |
| *** irf has joined #openstack-ironic | 07:06 | |
| *** mkovacik__ has quit IRC | 07:06 | |
| *** sacharya has quit IRC | 07:08 | |
| *** ohamada has joined #openstack-ironic | 07:11 | |
| *** yolanda has quit IRC | 07:17 | |
| *** hshiina has joined #openstack-ironic | 07:22 | |
| *** yolanda has joined #openstack-ironic | 07:23 | |
| *** ifarkas has joined #openstack-ironic | 07:29 | |
| *** vinbs has joined #openstack-ironic | 07:33 | |
| *** vinbs has quit IRC | 07:34 | |
| *** daemontool has joined #openstack-ironic | 07:45 | |
| *** Haomeng has quit IRC | 07:58 | |
| *** Haomeng has joined #openstack-ironic | 07:58 | |
| *** derekh has joined #openstack-ironic | 08:07 | |
| *** athomas has joined #openstack-ironic | 08:11 | |
| *** irf has quit IRC | 08:11 | |
| *** irf has joined #openstack-ironic | 08:12 | |
| irf | Morning Ironic !! | 08:12 |
|---|---|---|
| *** elfosardo_ is now known as elfosardo | 08:12 | |
| *** elfosardo has quit IRC | 08:13 | |
| *** lintan has quit IRC | 08:16 | |
| *** lintan has joined #openstack-ironic | 08:16 | |
| *** openstackgerrit has quit IRC | 08:17 | |
| *** openstackgerrit has joined #openstack-ironic | 08:18 | |
| *** lucas-afk has quit IRC | 08:19 | |
| *** daemontool has quit IRC | 08:21 | |
| *** yuanying has quit IRC | 08:23 | |
| *** chihhsin has joined #openstack-ironic | 08:24 | |
| *** lucasagomes has joined #openstack-ironic | 08:24 | |
| *** stendulker has joined #openstack-ironic | 08:25 | |
| *** mkovacik__ has joined #openstack-ironic | 08:26 | |
| lucasagomes | morning all :-) | 08:26 |
| lucasagomes | hey mrda irf | 08:27 |
| *** athomas has quit IRC | 08:29 | |
| *** athomas has joined #openstack-ironic | 08:30 | |
| irf | as continuation to the issue which i was facing last week, i was referring to the portion : http://docs.openstack.org/developer/ironic/deploy/install-guide.html#configure-networking-to-communicate-with-the-bare-metal-server | 08:30 |
| irf | do i need to create the flat network for baremetal ? to launch the instances? | 08:31 |
| vmud213 | morning lucasagomes,irf | 08:31 |
| *** jistr has joined #openstack-ironic | 08:32 | |
| lucasagomes | irf, for the moment, yes | 08:32 |
| lucasagomes | there are patches in the queue for ironic and nova (I believe neutron ones are merged already) that allow you to have network isolation | 08:33 |
| lucasagomes | still work in progress | 08:33 |
| irf | ok | 08:33 |
| irf | i was thinking of using the existing network for bare metal node | 08:34 |
| openstackgerrit | Aparna proposed openstack/ironic: Adds RAID interface for 'iscsi_ilo' https://review.openstack.org/314016 | 08:34 |
| *** fragatina has joined #openstack-ironic | 08:36 | |
| irf | ok.. moving further , while configuring the drivers for bare metal service , i have followed the steps for PXE setup as i am using pxe_wol driver .. is it enough ? Or do i need to folow the steps for UEFI and iPXE also ? | 08:39 |
| lucasagomes | irf, that depends on what you want, if you want iPXE to boot the machine you should follow the iPXE section | 08:41 |
| lucasagomes | for UEFI, pxe_wol is not able to flip the boot mode at deploy time so, it probably need to be manually configured if you want to have it | 08:41 |
| lucasagomes | (flip by adding the boot_mode:uefi capabilitiy I mean) | 08:42 |
| irf | ok so probably i will use just PXE then | 08:42 |
| *** e0ne has joined #openstack-ironic | 08:42 | |
| *** Nisha has joined #openstack-ironic | 08:43 | |
| irf | i have used the boot options as local | 08:46 |
| irf | using this command : ironic node-update <node-uuid> add properties/capabilities="boot_option:local" | 08:46 |
| irf | i havent use : UEFI | 08:48 |
| lucasagomes | right | 08:50 |
| lucasagomes | that's fine | 08:50 |
| lucasagomes | (as a note, uefi is boot_mode not boot_option) | 08:50 |
| irf | ok | 08:55 |
| *** mkovacik__ has quit IRC | 08:59 | |
| *** milan has joined #openstack-ironic | 09:00 | |
| *** linuxgeek has joined #openstack-ironic | 09:00 | |
| milan | morning ironic! | 09:00 |
| * milan changes nick from mkovacik | 09:00 | |
| *** moshele has joined #openstack-ironic | 09:01 | |
| *** e0ne has quit IRC | 09:03 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/pyghmi: DO NOT REVIEW, TESTING GATE https://review.openstack.org/313638 | 09:03 |
| irf | ia m still getting : No valid host was found. There are not enough hosts available. I think Ironic service and Nova are not able to comunicate properly ... in the nova conductor logs, i am seeing above error but dont know where to look for now ... | 09:03 |
| irf | i referd this as well : http://docs.openstack.org/developer/ironic/deploy/troubleshooting.html#nova-returns-no-valid-host-was-found-error | 09:04 |
| *** sacharya has joined #openstack-ironic | 09:05 | |
| *** sacharya has quit IRC | 09:10 | |
| irf | please help | 09:10 |
| lucasagomes | irf, that's a tricky error, have you checked n-scheduler logs see if there's something there? | 09:17 |
| *** dtantsur|afk is now known as dtantsur | 09:19 | |
| dtantsur | Morning Ironic, happy monday! :-P | 09:19 |
| lucasagomes | dtantsur, hey ya, morning | 09:19 |
| dtantsur | seems like I've finally recovered from Austin \o/ | 09:20 |
| irf | yes .. i have checked the logs and there also same error ..... | 09:20 |
| lucasagomes | JayF, quick updated on the vbmc thingy, apparently disabling the rsp timeout in pyghmi (https://github.com/openstack/pyghmi/blob/master/pyghmi/ipmi/bmc.py#L147) makes it way more reliable | 09:20 |
| lucasagomes | JayF, problem being not configurable by default :-) (/me will submit a patch to pyghmi) | 09:21 |
| lucasagomes | irf, right, but it doesn't say anything else? Can I see the output of nova hypervisor-stats? | 09:21 |
| irf | yes sure ... | 09:21 |
| dtantsur | lucasagomes, is our gate fine? the whiteboard says so, but I've seen strange errors over the weekend | 09:23 |
| irf | here it is : http://pastebin.com/isB1eWmd | 09:23 |
| lucasagomes | dtantsur, I'm not sure to be honest, lemme check | 09:24 |
| dtantsur | no worries, I'll figure out | 09:24 |
| * lucasagomes was looking more into pyghmi's gate now | 09:24 | |
| dtantsur | just thought maybe you already know | 09:24 |
| dtantsur | cause it was nova failed to connect to ironic with http 401 | 09:24 |
| lucasagomes | there's one suspicious failure here http://logs.openstack.org/16/314016/1/check/gate-tempest-dsvm-ironic-pxe_ipa/9a11ac8/logs/devstacklog.txt.gz#_2016-05-09_08_44_02_421 | 09:25 |
| Nisha | dtantsur, ++ even i am facing the same error for ilo ci | 09:25 |
| dtantsur | oooops | 09:25 |
| Nisha | from today | 09:25 |
| * dtantsur updates the whiteboard | 09:25 | |
| Nisha | dtantsur, till friday it was working | 09:26 |
| dtantsur | lucasagomes, this one is fine, I'm rather talking about https://review.openstack.org/313730 | 09:26 |
| dtantsur | Nisha, a link handly? | 09:27 |
| dtantsur | * handy | 09:27 |
| Nisha | http://logs.openstack.org/52/285852/44/check/gate-tempest-dsvm-ironic-pxe_ssh/153769f/logs/screen-n-cpu.txt.gz | 09:28 |
| dtantsur | yeah, the same as in inspector gate. thanks | 09:28 |
| Nisha | dtantsur, yeah looks like nova has pulled off the support from v2 of keystone | 09:29 |
| dtantsur | wonderful | 09:29 |
| Nisha | while ironic uses keystone v3 | 09:29 |
| *** hshiina has quit IRC | 09:29 | |
| Nisha | sorry | 09:29 |
| Nisha | v2 | 09:29 |
| lucasagomes | strange yes failing waiting for hyp-stats hmm | 09:30 |
| Nisha | dtantsur, i was strugglingon same from morning, then i checked even the stable ironci gates are failing jenkins due to this | 09:30 |
| irf | yes. | 09:30 |
| Nisha | yeah because n-cpu doesnt start | 09:30 |
| dtantsur | meh.. | 09:30 |
| lucasagomes | irf, is it a hybrid (vm + baremetal) cloud? | 09:31 |
| dtantsur | https://github.com/openstack-dev/devstack/commit/251b870e859e1a03b57c8fe24c7692a5e2a10fde ? | 09:31 |
| irf | yes | 09:31 |
| *** electrofelix has joined #openstack-ironic | 09:31 | |
| lucasagomes | irf, but apparently you have only 1 nova-compute right? | 09:31 |
| irf | just to mention again ... all openstack services are deployed on VM along with Ironic | 09:31 |
| irf | yes | 09:32 |
| lucasagomes | count=1 there in the hypervisors-stats | 09:32 |
| irf | yes | 09:32 |
| irf | my bare metal server is just one normal desktop machine | 09:32 |
| lucasagomes | irf, so that doesn't work, cause you need > n-computes, at least 1 to manage baremetal (with the ironic virt driver loaded) and the other one for your vms | 09:33 |
| Nisha | dtantsur, lucasagomes so we need to revert back this fix | 09:33 |
| lucasagomes | irf, which driver is loaded on that nova compute? | 09:33 |
| dtantsur | Nisha, on it right now | 09:33 |
| irf | this is writeen in nova.conf file in controller node : compute_driver=ironic.IronicDriver | 09:34 |
| * lucasagomes jumps on a quick meeting | 09:34 | |
| irf | on the compute node as well same line in nova.conf | 09:35 |
| irf | but, compute_manager line is commented .... | 09:35 |
| dtantsur | revert: https://review.openstack.org/#/c/314024/ | 09:36 |
| *** daemontool has joined #openstack-ironic | 09:45 | |
| *** baoli has joined #openstack-ironic | 09:50 | |
| Nisha | dtantsur, i started the ilo gate manually with the above patch. will update by my evening ( next 2-3 hrs) | 09:54 |
| *** baoli has quit IRC | 09:54 | |
| irf | are u back from meeting ? | 09:55 |
| dtantsur | great! we also have a (non-voting unfortunately) gate on devstack, so we'll know if the problem is fixed | 09:56 |
| * Nisha leaving for home right now. will connect from home later | 09:56 | |
| *** sambetts|afk is now known as sambetts | 09:57 | |
| dtantsur | ok folks, get ready for some ugly rant | 09:58 |
| dtantsur | how could they seriously merge https://review.openstack.org/#/c/300151/2/lib/nova_plugins/hypervisor-ironic if these options are not even present in https://github.com/openstack/nova/blob/master/nova/conf/ironic.py ??? | 09:58 |
| sambetts | dtantsur: is that whats blown everything up ?! | 10:00 |
| dtantsur | yep | 10:01 |
| * dtantsur is sad at devstack reviews.. | 10:01 | |
| *** Nisha has quit IRC | 10:01 | |
| sambetts | Don't we co-gate devstack? or are we non-voting? | 10:01 |
| *** hshiina has joined #openstack-ironic | 10:02 | |
| sambetts | where are you seeing the errors from that? The main errors I'm seeing is the nova hypervisor >= 1 one which points at the race condition I've highlighted | 10:03 |
| * lucasagomes is back now | 10:04 | |
| lucasagomes | irf, ok, but I see that you have some used resources over there | 10:05 |
| lucasagomes | memory_mb_used | 6144 | 10:05 |
| irf | ok .. | 10:05 |
| lucasagomes | because for each baremetal machine it will actually count as a hypervisor to nova | 10:05 |
| irf | not gettig sorry | 10:06 |
| lucasagomes | so where does the used resources coming from? | 10:06 |
| dtantsur | sambetts, we're non-voting, and people ignored the failed gate | 10:06 |
| irf | may be the case that VMs are using thise MBs? | 10:06 |
| dtantsur | sambetts, hypervisor stats error is because of http://logs.openstack.org/30/313730/1/check/gate-ironic-inspector-dsvm-ipa/27bc161/logs/screen-n-cpu.txt.gz#_2016-05-08_15_43_49_380 | 10:06 |
| irf | can this be happen ? | 10:06 |
| lucasagomes | irf, it may yes... Because, for baremetal what happens is that | 10:06 |
| lucasagomes | once you spawn a baremetal node, it will take up all the resources for node | 10:07 |
| lucasagomes | and each node is seems as a hypervisor on it's own in nova | 10:07 |
| lucasagomes | that's what I mean | 10:07 |
| lucasagomes | so once you deploy it, it resource tracker 100% of it | 10:07 |
| lucasagomes | since you have some resources used (idk how yet) it may be failing there | 10:07 |
| aparnav | Hey lucasagomes, I have a query about your comment on this patch https://review.openstack.org/#/c/307225 | 10:08 |
| aparnav | Is the check for task.is_take_over() necessary? | 10:08 |
| irf | ok so how shud i rectify that ... any command to check that or sometineg ? | 10:08 |
| lucasagomes | aparnav, lemme read the patch, cause I don't remember my comment | 10:09 |
| aparnav | lucasagomes, sure | 10:09 |
| lucasagomes | irf, hmm not that I know of :-/ | 10:09 |
| irf | i am sure , there is some issue in nova configuration but not sure what is that exactly | 10:09 |
| lucasagomes | perhaps the filters you've been using? | 10:10 |
| lucasagomes | we have some docs about hw to configure nova | 10:10 |
| sambetts | dtantsur: ah, I'm also seeing the hypervisor stats error because of the race condition I identified, results in similar devstack errors | 10:10 |
| lucasagomes | irf, http://docs.openstack.org/developer/ironic/deploy/install-guide.html#configure-compute-to-use-the-bare-metal-service | 10:10 |
| dtantsur | sambetts, ETOOMANYERRORS :D | 10:10 |
| sambetts | quite ... | 10:10 |
| irf | i did this already ... | 10:11 |
| irf | how about this : http://docs.openstack.org/developer/ironic/deploy/install-guide.html#configure-networking-to-communicate-with-the-bare-metal-server | 10:11 |
| *** daemontool_ has joined #openstack-ironic | 10:11 | |
| sambetts | irf: are you running two nova cpus? | 10:11 |
| irf | u mean nova compute ? | 10:12 |
| sambetts | yes | 10:12 |
| *** e0ne has joined #openstack-ironic | 10:12 | |
| irf | no only one nova compute .. as per the command : nova hypervisor-stats | 10:12 |
| *** daemontool has quit IRC | 10:14 | |
| sambetts | irf: if you want to run BM and VMs then you need to run 2 nova computes, 1 on your compute host configured with libvirt for hosting VMs, and another either on your controller or somewhere else configured with the Ironic compute driver etc to communicate with Ironic and provide BMs | 10:14 |
| *** hoangcx has quit IRC | 10:15 | |
| irf | ok so one nova compute is already running ... which is a nova compute node ... | 10:15 |
| irf | nova-compute service is running on this | 10:16 |
| sambetts | right and that is going to host VMs | 10:16 |
| sambetts | now you need a new nova compute, configured to talk to Ironic to provide the BM | 10:16 |
| irf | ok .. just to clear ... the current nova compute is running on VM. so it cant provision the bare metal node .. ? | 10:17 |
| sambetts | no, the fact its running in a VM has nothing to do with it, the problem is each nova compute service can only provide VMs or BM not both | 10:18 |
| sambetts | you need multiple nova compute service to do that | 10:18 |
| lucasagomes | aparnav, problem being, how do we implement such is_take_over() if not by listening the states | 10:19 |
| irf | ohh ok | 10:19 |
| irf | and i am trying to provision the both VM and BM as well | 10:19 |
| lucasagomes | aparnav, that shoulds like that, when _do_take_over run in the manager we should somehow pass a flag (via a parameter) to the prepare() function ? | 10:19 |
| lucasagomes | I haven't looked much into it | 10:19 |
| lucasagomes | but wondering how we could implement that | 10:19 |
| irf | so where i shud install new nova compute .. ? | 10:20 |
| sambetts | it can run on the controller or somewhere else | 10:20 |
| lucasagomes | aparnav, or maybe somehow say it when you get the lock? Cause I don't think that checking the purpose string is a good way of doing that | 10:20 |
| irf | and i understand that , controller can have new nova compute even if it is on the VM... | 10:21 |
| sambetts | if your VM is powerful enough to run that extra service | 10:21 |
| aparnav | lucasagomes, My question is do we really need to check for take_over? Cant we just check the provision_state since it the check will be only need for deploying and cleaning, also for inspection once it is added | 10:21 |
| irf | ok | 10:22 |
| lucasagomes | aparnav, yeah I was good with it as well | 10:22 |
| irf | but another dowbt ,,, nova schedulear and nova compute can be on same macbine ? any issue of haveing them on single VM ? | 10:23 |
| sambetts | no | 10:23 |
| irf | ok | 10:23 |
| sambetts | they can be on the same machine the defautl devstack setup runs every service on the same machine | 10:24 |
| irf | so whatever changes i have done in the nova.conf file on the nova compute .. i will remove that and do the same changes in the new nova compute once it is running on the controller node .... | 10:25 |
| sambetts | yup :) | 10:25 |
| aparnav | lucasagomes, the process of making the _do_takeover to send pass a flag or getting the lock seems complex. So I will proceed with checking the states. | 10:25 |
| lucasagomes | aparnav, +1 | 10:25 |
| lucasagomes | yeah, it requires some plumbing | 10:25 |
| irf | awesome !!! thanks much .. will update once i configure that ... | 10:26 |
| aparnav | lucasagomes, thank you :) | 10:26 |
| irf | further, is this by design that single nova compute service provisions only VMs or BMs but not both ? | 10:31 |
| *** daemontool__ has joined #openstack-ironic | 10:31 | |
| irf | can u plz help me in redirecting the documentation ... | 10:31 |
| sambetts | irf: yes, each nova compute can only be configured with one type of hypervisor, e.g. kvm, xen, or baremetal(ironic) | 10:31 |
| lucasagomes | aparnav, thank YOU for the fix | 10:32 |
| lucasagomes | irf, yup, that's how nova works | 10:32 |
| *** daemontool_ has quit IRC | 10:34 | |
| *** e0ne has quit IRC | 10:35 | |
| *** daemontool has joined #openstack-ironic | 10:36 | |
| *** daemontool__ has quit IRC | 10:37 | |
| *** e0ne has joined #openstack-ironic | 10:39 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/pyghmi: Allows BMCs to configure the event loop timeout https://review.openstack.org/314045 | 10:39 |
| *** mjturek1 has quit IRC | 10:41 | |
| *** daemontool_ has joined #openstack-ironic | 10:41 | |
| irf | ok ... thanks much | 10:41 |
| *** MattMan has quit IRC | 10:41 | |
| *** MattMan has joined #openstack-ironic | 10:42 | |
| lucasagomes | JayF, ^ | 10:42 |
| lucasagomes | I can confirm that tunning that timeout makes it way more stable | 10:43 |
| *** daemontool__ has joined #openstack-ironic | 10:43 | |
| irf | after configuration of new nova compute service , hypervisor-stats should show two hypervisor rite ? | 10:43 |
| lucasagomes | yup | 10:43 |
| irf | ok | 10:43 |
| *** daemontool has quit IRC | 10:44 | |
| *** daemontool_ has quit IRC | 10:46 | |
| *** daemontool_ has joined #openstack-ironic | 10:51 | |
| *** baoli has joined #openstack-ironic | 10:51 | |
| *** marios has quit IRC | 10:52 | |
| *** baoli has quit IRC | 10:55 | |
| *** daemontool__ has quit IRC | 10:55 | |
| *** mgould has joined #openstack-ironic | 10:59 | |
| *** daemontool__ has joined #openstack-ironic | 10:59 | |
| milan | hey guys, we've got +2x2 on the Inspector HA spec, should we possibly workflow+1 it? https://review.openstack.org/#/c/253675/20 | 10:59 |
| sambetts | milan: done ;) | 10:59 |
| mgould | morning Ironicers! | 11:00 |
| sambetts | Hey mgould! | 11:00 |
| * mgould checks the clock | 11:00 | |
| milan | sambetts, that's fast! thanks :D | 11:00 |
| mgould | damn, afternoon :-) | 11:00 |
| mgould | sambetts, milan hi! | 11:00 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/virtualbmc: Restructure the repository according to OpenStack https://review.openstack.org/313025 | 11:00 |
| milan | morning sambetts and moguld :) | 11:00 |
| sambetts | mgould: according to my clock you were 1 minute still in the morning haha | 11:00 |
| mgould | \o/ | 11:01 |
| dtantsur | milan, heh, forgot about i, sorry | 11:01 |
| dtantsur | afternoon, mgould :) | 11:01 |
| milan | dtantsur, no prob | 11:01 |
| openstackgerrit | Merged openstack/ironic-inspector-specs: High Availability for Ironic Inspector https://review.openstack.org/253675 | 11:02 |
| mgould | dtantsur, afternoon :-) | 11:02 |
| mgould | milan, high-five! | 11:02 |
| milan | mgould, yay :D | 11:03 |
| milan | hig-five! | 11:03 |
| *** daemontool_ has quit IRC | 11:03 | |
| milan | mgould, now the fun will start :D | 11:04 |
| *** daemontool has joined #openstack-ironic | 11:04 | |
| mgould | wait, you mean that was the easy bit? | 11:05 |
| * mgould cowers in a corner | 11:05 | |
| *** sacharya has joined #openstack-ironic | 11:05 | |
| *** e0ne has quit IRC | 11:06 | |
| *** daemontool_ has joined #openstack-ironic | 11:07 | |
| *** daemontool__ has quit IRC | 11:07 | |
| milan | mgould, seems I've got to hack it now :D | 11:08 |
| dtantsur | happy hacking :D | 11:08 |
| *** daemontool__ has joined #openstack-ironic | 11:08 | |
| milan | so, I'd say that was the easy bit :D | 11:08 |
| dtantsur | that reminds me to submit a release request for inspector | 11:08 |
| milan | dtantsur, thanks :D | 11:08 |
| milan | dtantsur, submit a preservation request and a solid fork maybe too ;) | 11:09 |
| dtantsur | hmm, I'd love to land https://review.openstack.org/#/c/276331/ before the release, so I'll update it today | 11:10 |
| *** daemontool has quit IRC | 11:10 | |
| *** sacharya has quit IRC | 11:10 | |
| sambetts | +1 | 11:11 |
| *** daemontool has joined #openstack-ironic | 11:11 | |
| *** daemontool_ has quit IRC | 11:12 | |
| lucasagomes | mgould, mat128 if you guys are happy with https://review.openstack.org/#/c/313025/ I will go ahead and merge it | 11:12 |
| mgould | lucasagomes, go for it | 11:12 |
| lucasagomes | I want to start doing some work on adding unittests for vbmc and things like that | 11:12 |
| lucasagomes | mgould, cool, thanks a lot for all the reviews! Very appreciated | 11:13 |
| lucasagomes | mgould, fyi, I've uncluded the umask + chdir when detaching the process | 11:13 |
| mgould | lucasagomes, cool | 11:14 |
| *** daemontool_ has joined #openstack-ironic | 11:14 | |
| dtantsur | lucasagomes, checking now | 11:15 |
| *** daemontool__ has quit IRC | 11:15 | |
| lucasagomes | dtantsur, ack | 11:15 |
| *** thrash|wknd is now known as thrash | 11:15 | |
| * lucasagomes waits | 11:15 | |
| dtantsur | done | 11:16 |
| *** smoriya has quit IRC | 11:17 | |
| *** daemontool has quit IRC | 11:18 | |
| jroll | morning everyone | 11:19 |
| jroll | YAY GATE | 11:20 |
| lucasagomes | jroll, heh down again :-/ | 11:20 |
| dtantsur | morning jroll, yes :( | 11:20 |
| dtantsur | happy monday | 11:20 |
| jroll | yah, pinged some folks | 11:20 |
| lucasagomes | jroll, if that makes you more comfortable, the new job on pyghmi's gate detected its first failure: http://logs.openstack.org/12/313012/4/check/gate-tempest-dsvm-ironic-pxe_ipmitool-pyghmi-src-nv/c5be74f/logs/screen-ir-cond.txt.gz#_2016-05-09_08_46_37_073 | 11:20 |
| lucasagomes | (which happens prior to the waiting for hypervisor-stats thingy) | 11:21 |
| jroll | lucasagomes: woot | 11:21 |
| jroll | so, is there any change we need to make in nova for keystone v3 to work with our client? | 11:25 |
| jroll | pas-ha: vdrok ^ I think you know things about this | 11:25 |
| odyssey4me | jroll I think we're already using Keystone v3 for the OSA implementation - let me check | 11:26 |
| jroll | yeah, I think it should just work, but it's unclear to me if it actually does :) | 11:27 |
| odyssey4me | jroll hmm, ok so we have done it in the neutron section: https://github.com/openstack/openstack-ansible-ironic/blob/master/templates/ironic.conf.j2#L100-L113 | 11:28 |
| odyssey4me | but I'm not seeing a nova section, unless I'm blind | 11:28 |
| odyssey4me | should there be a nova section? | 11:29 |
| jroll | odyssey4me: well, there's ironic configuration done in nova.conf | 11:29 |
| jroll | so the virt driver can talk to ironic | 11:30 |
| *** stendulker has quit IRC | 11:30 | |
| odyssey4me | ah yes | 11:30 |
| odyssey4me | yuk, this is keystone v2 https://github.com/openstack/openstack-ansible-os_nova/blob/master/templates/nova.conf.j2#L245-L253 | 11:31 |
| jroll | eh, I'm still not so sure. the options look like v2, however I think our client can handle v2 or v3 | 11:31 |
| jroll | but that todo makes me unsure | 11:31 |
| odyssey4me | mrda ^ I see your note - please test whether changing that up to Keystone v3 will work? | 11:31 |
| odyssey4me | jroll yeah, if your client is doing that, then as I understand it you're including logic that you shouldn't be... the keystone auth library already has that logic | 11:32 |
| vmud213 | lucasagomes:I didn't get your comment on https://review.openstack.org/#/c/311613 | 11:32 |
| jroll | odyssey4me: yeah, it's hard to tell, I'm not super familiar | 11:32 |
| odyssey4me | if you're shimming additional logic in front of that then you're walking into tech debt and causing an inconsistent operator experience | 11:32 |
| vmud213 | lucasagomes: bios local boot support for GPT partition table | 11:32 |
| jroll | odyssey4me: changing those nova.conf things directly to v3 options certainly doesn't work, devstack did that and broke our gate :) | 11:33 |
| odyssey4me | not you personally, of course | 11:33 |
| vmud213 | should i keep the Bios boot partition configurable or hardcoded? | 11:33 |
| lucasagomes | vmud213, on the release note? | 11:33 |
| lucasagomes | vmud213, I'm good with it being configurable (I added a comment in the patch-set, not per line about it) | 11:33 |
| jroll | odyssey4me: we do have quite a mess here, though https://github.com/openstack/python-ironicclient/blob/master/ironicclient/client.py#L20 | 11:33 |
| lucasagomes | vmud213, my -1 was about the release note | 11:33 |
| vmud213 | I think it'd always be better to keep it as configurable parameters | 11:33 |
| lucasagomes | vmud213, me too | 11:33 |
| lucasagomes | vmud213, my -1 was because of the release note | 11:34 |
| vmud213 | ok..so i see some commnet even on that..No problem. | 11:34 |
| lucasagomes | vmud213, see my comment at the bottom of the patch | 11:34 |
| vmud213 | i just make changes to release note and other nits. | 11:34 |
| lucasagomes | vmud213, (at "history") | 11:34 |
| odyssey4me | jroll yeah, I've seen those a few times - what a mess | 11:34 |
| lucasagomes | vmud213, yup | 11:34 |
| vmud213 | lucasagomes, got it.thanks | 11:35 |
| lucasagomes | cool, thanks for the patch | 11:35 |
| odyssey4me | jroll I would recommend having a chat with dolphm / lbragstad (or generally in #openstack-keystone) about what needs to be done to make it work right... and perhaps they can point you at a client that's doing it the right way | 11:35 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/virtualbmc: Restructure the repository according to OpenStack https://review.openstack.org/313025 | 11:35 |
| *** e0ne has joined #openstack-ironic | 11:36 | |
| jroll | odyssey4me: sure, I agree | 11:36 |
| *** amotoki has quit IRC | 11:36 | |
| *** aparnav has quit IRC | 11:36 | |
| *** frontrunner has quit IRC | 11:37 | |
| jroll | dtantsur: congrats on EOL :) | 11:39 |
| *** appu103_ has joined #openstack-ironic | 11:39 | |
| mgould | vmud213, not sure about this specific parameter, but in general: hardcoded if possible, to reduce code and ops complexity, but configurable if every possible answer will break some subset of users | 11:41 |
| *** e0ne has quit IRC | 11:42 | |
| vmud213 | mgould, i understand. But in this context i hope better to leave it as a configurable parameter. | 11:42 |
| * lucasagomes goes grab some food | 11:42 | |
| *** lucasagomes is now known as lucas-hungry | 11:43 | |
| mgould | vmud213, yeah, sounds like you're in situation #2 here | 11:43 |
| *** e0ne has joined #openstack-ironic | 11:46 | |
| openstackgerrit | Merged openstack/virtualbmc: Restructure the repository according to OpenStack https://review.openstack.org/313025 | 11:49 |
| *** nathan-h has joined #openstack-ironic | 11:53 | |
| *** fragatina has quit IRC | 11:54 | |
| jroll | 11:55:09 openstackgerrit | Jim Rollenhagen proposed openstack-infra/project-config: Give ironic grenade jobs 7 "nodes" running tinyipa https://review.openstack.org/314066 | 11:55 |
| jroll | should be useful | 11:55 |
| sambetts | sweet :D | 11:55 |
| jroll | I realized smoke tests must be passing for people if they are getting to the resource create bits | 11:56 |
| jroll | I thought the order was the other way for some reason | 11:56 |
| sambetts | I think it is the other way around :/ | 11:57 |
| *** yolanda has quit IRC | 11:57 | |
| sambetts | at least thats how I understood it | 11:57 |
| *** e0ne has quit IRC | 11:58 | |
| jroll | looking at the gate it runs smoke tests first | 11:58 |
| *** baoli has joined #openstack-ironic | 11:58 | |
| jroll | well, and the code https://github.com/openstack-dev/grenade/blob/master/grenade.sh#L241-L262 | 11:58 |
| *** dprince has joined #openstack-ironic | 12:00 | |
| sambetts | :/ oh ... well thats pretty good then | 12:01 |
| dtantsur | jroll, heh, thanks :) | 12:01 |
| *** yolanda has joined #openstack-ironic | 12:02 | |
| *** baoli has quit IRC | 12:03 | |
| openstackgerrit | vinay kumar muddu proposed openstack/ironic-lib: Add support for BIOS local boot for GPT label https://review.openstack.org/311613 | 12:06 |
| dtantsur | jroll, lucas-hungry, I think we should keep an eye on https://review.openstack.org/#/c/300154/ | 12:08 |
| jroll | dtantsur: heh, so a missing depends-on is what broke us, eh | 12:09 |
| dtantsur | yep | 12:09 |
| jroll | cool | 12:11 |
| sambetts | \0\ | 12:12 |
| sambetts | \-/ | 12:12 |
| sambetts | ffs... | 12:12 |
| sambetts | \0/ | 12:12 |
| jroll | lol | 12:13 |
| *** early` has quit IRC | 12:14 | |
| *** baoli has joined #openstack-ironic | 12:14 | |
| *** mjturek1 has joined #openstack-ironic | 12:24 | |
| openstackgerrit | Mathieu Mitchell proposed openstack/ironic: Allow configuring shred's final overwrite with zeros https://review.openstack.org/304101 | 12:24 |
| *** appu103_ has quit IRC | 12:25 | |
| irf | hello , i am back .. | 12:27 |
| *** jjohnson2_ has joined #openstack-ironic | 12:27 | |
| irf | configured the nova compute service on the controoler node and now hypervisor-stats command shoes two compute nodes | 12:27 |
| irf | do i need to foloow any specific steps to use the baremetal hypervisor for the second compute service ? | 12:28 |
| sambetts | irf: the ones in the install guide, e.g. setting the right virt driver etc | 12:28 |
| *** mgould has quit IRC | 12:29 | |
| irf | yes i did that .. | 12:30 |
| irf | i set virt_type=qemu in the nova-compute conf file | 12:30 |
| irf | full path of the file : /etc/nova/nova-compute.conf | 12:30 |
| irf | both the hypervisors are enabled and up | 12:31 |
| sambetts | qemu? thats a virtual machine hypervisor, you need to set it to the ironic virt driver, like in the install guide http://docs.openstack.org/developer/ironic/deploy/install-guide.html#configure-compute-to-use-the-bare-metal-service compute_driver=ironic.IronicDriver | 12:31 |
| irf | ohhh | 12:32 |
| irf | sorrry | 12:32 |
| *** daemontool__ has joined #openstack-ironic | 12:41 | |
| *** early has joined #openstack-ironic | 12:43 | |
| *** daemontool_ has quit IRC | 12:45 | |
| openstackgerrit | Merged openstack/pyghmi: Add Hardware inventory from Lenovo Agentless https://review.openstack.org/312568 | 12:46 |
| *** Nisha has joined #openstack-ironic | 12:47 | |
| irf | ok .. i have added compute_driver=ironic.IronicDriver in the default section of nova-compute file | 12:47 |
| irf | removed the entry : [libvert] | 12:48 |
| *** marios has joined #openstack-ironic | 12:50 | |
| irf | ok now my Ironic node is being listed as hypervisor host for the command : nova hypervisor-list is it rite ? | 12:52 |
| sambetts | Yes :D | 12:52 |
| Nisha | dtantsur, your patch fixed the stack issue | 12:52 |
| Nisha | :) | 12:52 |
| irf | ok gre8 | 12:52 |
| dtantsur | Nisha, awesome, thanks for checkingf | 12:52 |
| *** lucas-hungry is now known as lucasagomes | 12:54 | |
| *** mgould has joined #openstack-ironic | 12:54 | |
| lucasagomes | dtantsur, ++ will take a look | 12:55 |
| *** vmud213 has quit IRC | 12:56 | |
| *** irf has quit IRC | 12:56 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: [inspection] wait for the PXE DHCP by default and remove the carrier check https://review.openstack.org/313511 | 12:57 |
| dtantsur | sambetts, mat128, this one should be fine by you hopefully :) ^^^ | 12:57 |
| *** zzzeek has joined #openstack-ironic | 12:58 | |
| *** yolanda has quit IRC | 12:58 | |
| *** ppiela has joined #openstack-ironic | 13:00 | |
| Nisha | dtantsur, do we have any code patches for https://review.openstack.org/#/c/305864/ | 13:00 |
| mat128 | dtantsur: works for me :) I thought we had discussed options tho (c.f. "folks not proposing anything") | 13:01 |
| sambetts | dtantsur: https://review.openstack.org/#/c/314086/ I've just pushed this btw, could you maybe try it and see if it helps too? my concern is that dhcp-all-interfaces won't actually configure an interface for dhcp if carrier != 1 so I don't know how the node in your bug zilla is getting an IP if carrier is wrong | 13:01 |
| dtantsur | Nisha, not yet. it was on my radar, but I got distracted by bugs. I want a buy-in from at least one core though | 13:01 |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Support Ironic node names in our API https://review.openstack.org/276331 | 13:01 |
| dtantsur | mat128, well, not something from the top of my head :( | 13:01 |
| Nisha | dtantsur, :) I was trying to use the code patches for my inband inspection support in ilo drivers | 13:02 |
| mat128 | dtantsur: no problem, I remember suggesting something like "trust_sysfs=false" or something, but don't worry about it :) your current patch makes it even better | 13:02 |
| *** links has quit IRC | 13:02 | |
| dtantsur | sambetts, DIB magic, yeah... it definitely gets an IP *somehow* otherwise IPA would not come back | 13:02 |
| Nisha | dtantsur, as this is the base for any code changes i do for inband inspection supprot there | 13:03 |
| dtantsur | mat128, yeah, sorry. My problem with trust_sysfs is that we're shifting the decision to a user, and a user/operator might not actually know | 13:03 |
| dtantsur | mat128, so instead I suggest them to specify what they expect us to return *for sure*. I think it's more fair to them. | 13:03 |
| *** rloo has joined #openstack-ironic | 13:03 | |
| mat128 | yep, your patch is perfect as is and works for every use case I can think of :) | 13:04 |
| *** yolanda has joined #openstack-ironic | 13:04 | |
| dtantsur | great! | 13:04 |
| dtantsur | Nisha, I'll come to it soon. Please review the spec in the meanwhile. | 13:04 |
| Nisha | Yeah thats in my list... | 13:04 |
| Nisha | dtantsur, lucasagomes regarding comments on https://review.openstack.org/#/c/302403/ | 13:06 |
| dtantsur | sambetts, I need to say that building tinyipa is substantially faster than the coreos one! | 13:06 |
| * dtantsur uses it in his testing now | 13:06 | |
| sambetts | dtantsur: :D | 13:06 |
| *** sacharya has joined #openstack-ironic | 13:06 | |
| * sambetts so happy he's made a significant difference to Ironic with tinyipa hehe | 13:07 | |
| *** jcoufal has joined #openstack-ironic | 13:07 | |
| dtantsur | yeah, cheers :) | 13:07 |
| * dtantsur is looking forward to dropping coreos here | 13:08 | |
| * dtantsur wishes he could avoid DIB just as easily..... | 13:08 | |
| Nisha | dtantsur, ilo drivers need to use DIB | 13:10 |
| sambetts | that DIB patch I added will hopfully make it better for you, not sure though, are you stuck on DIB because you have to build a RH based IPA ramdisk? | 13:10 |
| sambetts | Nisha: for making an ISO? | 13:10 |
| dtantsur | Nisha, I know... I'd prefer we use something different to build images, e.g. libguestfs | 13:10 |
| Nisha | sambetts, yeah as DIB only supports uefi | 13:10 |
| Nisha | and secureboot | 13:10 |
| *** sacharya has quit IRC | 13:11 | |
| Nisha | and frankly building images thru coreos/tinyipa doesnt work here in the servers i am trying...:( couldnt make that work suceesfully...anyway i need DIB for testing it on gate | 13:12 |
| sambetts | hmm, I wonder what changes I need to make to tinyipa to make it support those :/ there is "make all" or "make then make iso" in tinyipa that'll spit out an iso btw :) | 13:12 |
| *** krtaylor has quit IRC | 13:12 | |
| Nisha | sambetts, its not those steps, generally it fails at puuling the coreos/tinyipa initially step only | 13:13 |
| Nisha | it doesnt go thru the process i would say | 13:13 |
| Nisha | so i dont say that it doesnt work, but i would i am not able to make it work | 13:13 |
| mat128 | Nisha: could you post a log by any chance? | 13:13 |
| sambetts | Nisha: I'd love to know whats missing / what error you see with tinyipa on your end | 13:13 |
| sambetts | so we can get it working for you | 13:14 |
| Nisha | mat128, i would do once i get chance to do, i need to make ilo CI public asap | 13:14 |
| Nisha | sambetts, ilo CI needs to use DIB for UEFI and secureboot | 13:14 |
| Nisha | those are not supported by coreos and tinyipa yet | 13:15 |
| *** jtomasek_ has joined #openstack-ironic | 13:15 | |
| Nisha | dtantsur, could you look into https://review.openstack.org/#/c/302403/ and let me know ur opinion on the comments | 13:15 |
| *** zzzeek has quit IRC | 13:16 | |
| sambetts | Nisha: can I ask what elements you use to build IPA through DIB so I can check out the code and work out what needs adding to tinyipa to make it work with those | 13:16 |
| *** amotoki has joined #openstack-ironic | 13:17 | |
| *** Nisha_away has joined #openstack-ironic | 13:17 | |
| *** zzzeek has joined #openstack-ironic | 13:18 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/virtualbmc: "vbmc list" to sort the result by domain name https://review.openstack.org/314095 | 13:18 |
| *** krtaylor has joined #openstack-ironic | 13:19 | |
| *** Nisha_brb has joined #openstack-ironic | 13:20 | |
| *** Nisha has quit IRC | 13:20 | |
| *** KennethWilke has quit IRC | 13:21 | |
| *** Nisha_away has quit IRC | 13:22 | |
| *** jtomasek_ has quit IRC | 13:22 | |
| dtantsur | sambetts, I guess tinyipa does not have journalctl? | 13:23 |
| sambetts | dtantsur: no, its not systemd | 13:23 |
| mat128 | dtantsur: CoreOS does | 13:23 |
| mat128 | if thats the only thing thats blocking you | 13:23 |
| lucasagomes | jjohnson2_, around? when you have some time, mind taking a look at https://review.openstack.org/#/c/314045/ ? | 13:23 |
| dtantsur | just checking, not important right now | 13:23 |
| dtantsur | thnx | 13:23 |
| *** rbudden has joined #openstack-ironic | 13:24 | |
| jjohnson2_ | lucasagomes, so weird, I still don't get notified about pyghmi reviews naturally | 13:24 |
| lucasagomes | jjohnson2_, btw, it would be great to have unittests in pyghmi (I'm filling a bug about it) | 13:25 |
| *** marios has quit IRC | 13:25 | |
| lucasagomes | jjohnson2_, oh, yeah notifications are handy | 13:25 |
| lucasagomes | there's also a problem with the python 3 patch, I left a link to the error in the review | 13:25 |
| jjohnson2_ | lucasagomes, yeah, they don't even appear on incoming reviews... | 13:25 |
| *** marios has joined #openstack-ironic | 13:25 | |
| jjohnson2_ | lucasagomes, yeah, I mentioned it don't work... ;) | 13:25 |
| lucasagomes | oh heh | 13:25 |
| lucasagomes | jjohnson2_, at least we can detect it on gate now :-) | 13:25 |
| dtantsur | sambetts, aha. maybe we need to fix the 'logs' inspection collector to support whatever is for logging there | 13:26 |
| jjohnson2_ | lucasagomes, the virtual bmc code is broken by the changes | 13:26 |
| jjohnson2_ | lucasagomes, so it just hangs out | 13:26 |
| jjohnson2_ | I mean the review just hangs out | 13:26 |
| lucasagomes | right on | 13:26 |
| jjohnson2_ | the error is an error | 13:26 |
| lucasagomes | I can take a look if needed | 13:26 |
| jjohnson2_ | it's too early on a monday for me to speak good | 13:26 |
| openstackgerrit | Merged openstack/pyghmi: Allows BMCs to configure the event loop timeout https://review.openstack.org/314045 | 13:27 |
| sambetts | dtantsur: yeah, might be a plan, I think by default even syslog is disabled because it takes more RAM XD | 13:27 |
| *** keedya has joined #openstack-ironic | 13:27 | |
| jjohnson2_ | lucasagomes, if you don't mind, it's a matter of the whole bytes/bytearray/string stuff being fixed | 13:27 |
| jjohnson2_ | lucasagomes, actually, I didn't look at the gate failure yet, just based on my recollection of tests before even running git review | 13:27 |
| dtantsur | sambetts, ouch :D | 13:27 |
| *** vishwanathj has joined #openstack-ironic | 13:28 | |
| *** Nisha_away has joined #openstack-ironic | 13:28 | |
| lucasagomes | jjohnson2_, yeah no problems! | 13:28 |
| jjohnson2_ | lucasagomes, one would think integration tests by starting a fake bmc and then running commands against the fake bmc would be handy | 13:29 |
| lucasagomes | jjohnson2_, ++ totally | 13:29 |
| lucasagomes | and in parallel we should start working on unittests for pyghmi | 13:30 |
| jjohnson2_ | of course, I need to upload some sample FRU, SPD, and so on and so forth for the parser fuctions to hit | 13:30 |
| jlvillal | lucasagomes, I was wondering if you had a chance to test Grenade anymore? | 13:30 |
| jjohnson2_ | and now to work my other job... | 13:30 |
| jlvillal | mgould, Also wondering if you had made any progress on Grenade? :) | 13:31 |
| *** KennethWilke has joined #openstack-ironic | 13:31 | |
| mgould | jlvillal, nope | 13:31 |
| *** Nisha_brb has quit IRC | 13:31 | |
| mgould | you? | 13:31 |
| lucasagomes | jlvillal, I haven't runned it today, on friday I passed the network problem using jroll's patch | 13:31 |
| lucasagomes | then hit one about not having enough loop devices | 13:32 |
| lucasagomes | so I manually created 200 of 'em heh now I'm hitting something different, lemme run it again | 13:32 |
| lucasagomes | see what I get | 13:32 |
| jlvillal | lucasagomes, Thanks. I think you got farther than me. I am dying elsewhere. I will update information when I get to work. | 13:32 |
| *** rloo has quit IRC | 13:32 | |
| *** mtanino has joined #openstack-ironic | 13:32 | |
| *** rloo has joined #openstack-ironic | 13:33 | |
| jlvillal | mgould, Thanks. I've been running it. Getting an error. I will update etherpad in next few hours | 13:33 |
| mgould | jlvillal, cool | 13:33 |
| mgould | are you going to be around for the meeting today? | 13:33 |
| jlvillal | Yep | 13:33 |
| lucasagomes | jlvillal, cool, I'm trying to split the time between vbmc/pyghmi's and grenade cause both needs a lot of attention at least for now | 13:33 |
| jlvillal | lucasagomes, Thanks! | 13:33 |
| *** yolanda has quit IRC | 13:36 | |
| dtantsur | sambetts, you are fine with backport https://review.openstack.org/#/c/313932/ ? | 13:37 |
| sambetts | dtantsur: yup +1 | 13:37 |
| dtantsur | sambetts, please +2 then ;) | 13:39 |
| sambetts | sure :F | 13:41 |
| sambetts | :D | 13:41 |
| *** yolanda has joined #openstack-ironic | 13:43 | |
| dtantsur | jroll, wdyt about https://bugs.launchpad.net/ironic/+bug/1578947 | 13:45 |
| openstack | Launchpad bug 1578947 in Ironic "No need to do node clean if it deleted from a deploy failed state" [Undecided,Opinion] - Assigned to Zhenguo Niu (niu-zglinux) | 13:45 |
| jroll | dtantsur: totally agree with you, posted as such | 13:46 |
| sambetts | we about to respond we exactly the same thing | 13:47 |
| NobodyCam | good morning Ironic :) | 13:47 |
| jroll | I for one know that my deployment needs a reclean if a deploy fails | 13:47 |
| dtantsur | devananda, hey, I've see your (expired) bug about us not fully wiping the GPT backup. I suspect https://bugs.launchpad.net/ironic-python-agent/+bug/1576526 might be a case of it. | 13:48 |
| openstack | Launchpad bug 1576526 in ironic-python-agent "Issue in reteieving UUID of the root partition during boot loader installation" [Undecided,Incomplete] | 13:48 |
| *** david-lyle_ has joined #openstack-ironic | 13:48 | |
| *** mtanino has quit IRC | 13:49 | |
| *** milan has quit IRC | 13:49 | |
| *** david-lyle has quit IRC | 13:50 | |
| *** xhku has joined #openstack-ironic | 13:51 | |
| *** ametts has joined #openstack-ironic | 13:53 | |
| dtantsur | jroll, an RFE for our attention: https://bugs.launchpad.net/ironic/+bug/1575935 | 13:54 |
| openstack | Launchpad bug 1575935 in Ironic "[RFE] Rebuild should also accept a configdrive" [Wishlist,Confirmed] - Assigned to Stephane Miller (stephaneeee) | 13:54 |
| jroll | dtantsur: my opinion on that, without clicking, is "we should do whatever VMs do" | 13:56 |
| * dtantsur has no idea what VM's do | 13:56 | |
| sambetts | +1 for VM convergence | 13:57 |
| mgould | NobodyCam, monring | 13:57 |
| mgould | *morning | 13:57 |
| sambetts | o/ NobodyCam | 13:57 |
| dtantsur | oh, morning NobodyCam, jlvillal | 13:57 |
| dtantsur | and morning jroll (in case I didn't say it today) | 13:57 |
| jroll | dtantsur: going by api docs it looks like nova supports this, so sure! | 13:57 |
| jroll | heh | 13:57 |
| jroll | morning :) | 13:57 |
| jlvillal | Good morning dtantsur and everyone else! :) | 13:58 |
| NobodyCam | morning mgould sambetts dtantsur jroll | 13:58 |
| NobodyCam | jlvillal: too | 13:58 |
| *** Goneri has joined #openstack-ironic | 13:59 | |
| dtantsur | jroll, sambetts, this sounds like a lot of fun: https://bugs.launchpad.net/ironic/+bug/1576098 | 13:59 |
| openstack | Launchpad bug 1576098 in Ironic "[RFE] Support assign different types of network cards to different networks" [Undecided,New] - Assigned to Zhenguo Niu (niu-zglinux) | 13:59 |
| jroll | O_o | 13:59 |
| sambetts | dtantsur: yup, thats the port mapping issue that we've got a follow up spec for | 14:00 |
| jroll | oh, this one is so played out | 14:00 |
| dtantsur | I mean, I see point. But it's gonna be tricky | 14:00 |
| jroll | cmon LP looooad | 14:00 |
| jroll | dtantsur: there's a dupe for it | 14:00 |
| dtantsur | good to know. a link handy? | 14:00 |
| sambetts | the only way is supporting a cababilitys like thing for portsa | 14:00 |
| sambetts | man I can't type todaty | 14:01 |
| sambetts | -,- | 14:01 |
| dtantsur | :D | 14:01 |
| jroll | dtantsur: I think https://bugs.launchpad.net/ironic/+bug/1544169 | 14:01 |
| openstack | Launchpad bug 1544169 in Ironic "[RFE] Advanced network configuration in ironic" [Wishlist,Confirmed] - Assigned to Vasyl Saienko (vsaienko) | 14:01 |
| dtantsur | that's monday | 14:01 |
| *** catintheroof has joined #openstack-ironic | 14:03 | |
| dtantsur | jroll, do you think https://bugs.launchpad.net/ironic/+bug/1575249 should be a clean step? do we need a spec here? | 14:04 |
| openstack | Launchpad bug 1575249 in Ironic "Add image caching support to agent driver" [Wishlist,In progress] - Assigned to Clif Houck (f-te-u) | 14:04 |
| *** piet has joined #openstack-ironic | 14:04 | |
| dtantsur | jroll, also tell me to shut up at any moment if I distract you :) just going through bug notifications | 14:04 |
| jroll | dtantsur: I don't, because going manageable->available will wipe the image right off the disk :) | 14:07 |
| *** d0ugal has quit IRC | 14:07 | |
| dtantsur | jroll, so it should be a clean step with lower priority, right? | 14:07 |
| jroll | for automated cleaning... maybe, but I like the given model | 14:07 |
| jroll | for example, it allows some service to do statistics to cache the right amount of different images | 14:08 |
| jroll | (or a person) | 14:08 |
| *** links has joined #openstack-ironic | 14:08 | |
| dtantsur | so when and how should it be called? | 14:08 |
| rloo | hello and a good monday morning to everyone, including YOU (jroll, dtantsur, sambetts, NobodyCam, jlvillal, mgould, lucasagomes | 14:09 |
| jroll | dtantsur: while a node is in available, via the API | 14:09 |
| lucasagomes | rloo, hello there, good morning :-) | 14:09 |
| sambetts | hey rloo | 14:09 |
| dtantsur | morning rloo :) | 14:09 |
| jroll | morning rloo :) | 14:09 |
| NobodyCam | hey hey morning rloo :) | 14:09 |
| dtantsur | jroll, directly via IPA API? or via a new Ironic API? I guess this RFE needs a spec :) | 14:09 |
| jroll | dtantsur: it has one, no? | 14:09 |
| jroll | maybe not in the bug itself | 14:10 |
| jroll | dtantsur: https://review.openstack.org/#/c/310594/ | 14:10 |
| dtantsur | ouch, yeah :( | 14:10 |
| * dtantsur is probably too sleepy | 14:10 | |
| jroll | dtantsur: disclaimer with my comments here: we do this in production today | 14:11 |
| jroll | 2 minute 'nova boot' ftw | 14:11 |
| dtantsur | I see | 14:11 |
| *** links has quit IRC | 14:11 | |
| mgould | morning rloo | 14:12 |
| rloo | dtantsur: wrt https://bugs.launchpad.net/python-ironicclient/+bug/1568645 | 14:12 |
| openstack | Launchpad bug 1568645 in python-ironicclient "Improve baremetal unittest framework of the OSC " [Undecided,Invalid] - Assigned to xiexs (xiexs) | 14:12 |
| dtantsur | mmm? | 14:12 |
| *** fragatina has joined #openstack-ironic | 14:13 | |
| rloo | dtantsur: is it true that we don't track refactorings via our bug tracking system? as an rfe/wishlist maybe? | 14:13 |
| rloo | dtantsur: am asking because i know we let some? refactorings get approved w/o a bug i think. | 14:13 |
| rloo | dtantsur: but i also think it is useful to have something track the refactoring. depends on how much of a refactoring too. | 14:13 |
| rloo | dtantsur: cuz what if we don't want the proposed refactoring. | 14:14 |
| dtantsur | rloo, that's my take on it. if it's a huge refactoring (e.g. futurist switch), it's worth tracking. if it's small tests refactoring (like here), it's not IMO | 14:14 |
| dtantsur | rloo, if we don't want the proposed refactoring, we should -2 the patch. closing the bug wont' help :) | 14:14 |
| rloo | dtantsur: 'huge' refactoring is ... questionable. that bug mentions two steps; it seems useful to have the bug that describes the overall intent. anyway, just my opinion. | 14:16 |
| rloo | dtantsur: I mean 'huge' is kinda hand waving. not sure how big 'huge' is. | 14:17 |
| dtantsur | rloo, imagine you saying "omg it's huge" :D my problem is that I don't want every small change posted to the bug tracker. I'll go mad figuring them out | 14:17 |
| rloo | dtantsur: the other use for the bug tracking is if there is more than one person that intends to do the same thing; at least this is an indication to others that someone is planning on doing it. | 14:18 |
| dtantsur | i.e. I'll have to assess validness of every one of them. I'd prefer it to be done during reviews instead | 14:18 |
| dtantsur | rloo, I'd prefer bug tracking to stay bug tracking (+ rfe) for reasons stated above :) if we can make the whole core team triage the bugs regularly (just as reviews), I can change my mind | 14:19 |
| rloo | dtantsur: oh. i'd rather assess validness earlier in the process; seems like that would save people time. instead of someone doing all the code changes etc and then finding out that we don't want it. | 14:19 |
| dtantsur | rloo, that would save their time, but waste mine | 14:19 |
| dtantsur | also it will make the future of their proposal depend on one person triaging their bug | 14:19 |
| *** Nisha_away has quit IRC | 14:19 | |
| dtantsur | which is contrary to our 2x +2 rule in gerrit | 14:20 |
| rloo | dtantsur: ah. I see. True. waste your/someone's time now, or waste someone's time (and maybe more than one person's time) later. | 14:20 |
| rloo | dtantsur: the problem is that you are the only one that is doing 99.99% of the bug triaging. | 14:21 |
| rloo | dtantsur: ok, until we can solve that, i guess you can do what works for you :) | 14:22 |
| dtantsur | :) | 14:22 |
| dtantsur | you can bring it to the meeting, if you feel it's important | 14:23 |
| rloo | dtantsur: well, i think it is important, but note that i didn't volunteer to help you with bug triaging. | 14:23 |
| dtantsur | LOL | 14:23 |
| dtantsur | that's how it ends up ;) | 14:23 |
| *** ChrisAusten has joined #openstack-ironic | 14:23 | |
| *** keedya has quit IRC | 14:25 | |
| rloo | dtantsur: i was also asking, cuz when I see patches, i sometimes ask for a bug to be associated with it. so would like to be consistent wrt what we expect bugs/rfes for. | 14:25 |
| dtantsur | yeah, good point | 14:25 |
| dtantsur | I still treat bugs as bugs (in a sense of something broken) with an explicit exception for RFE's (which can be treated as missing functionality) | 14:26 |
| *** itamarl_ has joined #openstack-ironic | 14:27 | |
| *** itamarl has quit IRC | 14:29 | |
| *** itamarl_ is now known as itamarl | 14:29 | |
| *** keedya has joined #openstack-ironic | 14:30 | |
| rloo | hey lucasagomes, when you get a chance, i had some questions on https://review.openstack.org/#/c/308965/. wrt how to deal with it in the future. | 14:32 |
| lucasagomes | rloo, hello there, lemme take a look | 14:33 |
| *** amotoki has quit IRC | 14:34 | |
| *** hshiina has quit IRC | 14:34 | |
| lucasagomes | rloo, ++ for the optimization by generating the list only once + the two other nits. Re: moving from pecan/wsme | 14:35 |
| sambetts | rloo, lucasagomes: didn't we talk about making the API more RESTful, and the URLs safer e.g. /nodes?detail=true | 14:36 |
| *** keedya has quit IRC | 14:36 | |
| sambetts | instead of having "fake" resources | 14:36 |
| lucasagomes | rloo, I believe that once that happens we can make the list static again as part of that work? | 14:36 |
| lucasagomes | sambetts, right, but we can't change it now | 14:37 |
| *** dprince has quit IRC | 14:37 | |
| rloo | sambetts, so you're saying that whatever we move to, we won't have the sam problem we have with wsme/pecan? | 14:37 |
| sambetts | but for new APIs going forward we should do it that way | 14:37 |
| sambetts | no more fake resources on new things | 14:37 |
| lucasagomes | sambetts, ++ agreed, yeah we should rethink the design of the API once v2 start being planned | 14:37 |
| lucasagomes | fwiw, nova also has the /detail and such endpoints | 14:38 |
| lucasagomes | we copied it over | 14:38 |
| sambetts | yeah, I hate it :-P | 14:38 |
| rloo | sambetts: AH, ok, got it. So that list of invalid reserved words really is static and should not change. | 14:38 |
| rloo | sambetts, lucasagomes: oh wait. you mean only when we go to v2? | 14:38 |
| sambetts | I'd hope we shouldn't have to add to it | 14:38 |
| lucasagomes | rloo, yeah, we have to keep the endopoints for v1, because removing it might potentially break people | 14:39 |
| *** mtanino has joined #openstack-ironic | 14:39 | |
| rloo | lucasagomes: we have to keep the existing endpoints for v1. but if we add new endpoints in v1, they will also have to be reserved, or we can do what sam suggested and not treat them as fake resources? | 14:40 |
| lucasagomes | rloo, or if you mean the static list, yeah I believe we will keep pecan/wsme for v1 no? I first thought about translating it using something else | 14:40 |
| lucasagomes | but that's going to be a lot of work :-/ | 14:40 |
| *** stendulker has joined #openstack-ironic | 14:40 | |
| sambetts | tbh I think we should add nodes?detail=true endpoints in parallel with the existing ones, and document them but without remove the old ones, then any new ones should be that styler | 14:40 |
| lucasagomes | rloo, oh for new ones we probably could do as sambetts said yes | 14:40 |
| rloo | sambetts, lucasagomes: OK, got it. so for v1, that reserved list IS static. and for v2, there won't be a reserved list. | 14:41 |
| lucasagomes | tho, the api will be a tad more inconsistent heh | 14:41 |
| lucasagomes | rloo, hope we won't need such list for v2 | 14:41 |
| sambetts | ++ | 14:41 |
| lucasagomes | cause that's a design flaw | 14:41 |
| rloo | lucasagomes: in that case, we should just make the list static instead of generating it. that way, if the unit test fails, we know that someone added a new endpoint. | 14:41 |
| sambetts | lucasagomes: we could fix the inconsistancy by supporting both styles for the old endpoints | 14:41 |
| lucasagomes | rloo, yeah, even tho I generate it inspecting the class I've added a unittest for that (see test_utils.py) | 14:42 |
| lucasagomes | sambetts, right, yeah, honestly I would love to start focusing on v2 if we are going to do that | 14:42 |
| *** fragatina has quit IRC | 14:42 | |
| lucasagomes | v1 already looks like a pig with lipstick heh | 14:42 |
| rloo | lucasagomes: the problem with that code is future direction is lacking. someone looking at it may think, i'll just add a new endpoint and things will work properly. if we don't want to add new endpoints similar to what we have, we need to reflect that somehow outside of irc :) | 14:43 |
| sambetts | hahahaha great way to describe it | 14:43 |
| NobodyCam | no one liked the V2 idea when I brought it up :( | 14:43 |
| lucasagomes | NobodyCam, I like it | 14:43 |
| dtantsur | I did | 14:43 |
| rloo | lucasagomes, sambetts: hey, don't make fun of ms piggy. | 14:43 |
| NobodyCam | :) | 14:43 |
| NobodyCam | rloo: lol | 14:43 |
| lucasagomes | rloo, right... I can make it static again, I mean, I don't mind because both for now works. I see your point and also deva's point | 14:44 |
| rloo | lucasagomes, sambetts: sorry, it is Miss Piggy: https://en.wikipedia.org/wiki/Miss_Piggy | 14:44 |
| lucasagomes | but I hope that bits of the api won't change until EOL of v1 | 14:44 |
| lucasagomes | rloo, you want me to put a patch making it static again and then we can keep the discussion there? | 14:44 |
| rloo | lucasagomes: I think we should add comments in the code (assuming we all agree with doing as sam said and the inconsistency in the API) | 14:45 |
| rloo | lucasagomes: we don't know when we will end v1 and i'd rather we document things. too many things to remember now as it is. i'm not going to remember. | 14:45 |
| lucasagomes | rloo, agree, just wondering if we need to do it only on that class | 14:46 |
| lucasagomes | or all classes with subcontrollers | 14:46 |
| rloo | lucasagomes: adding a follow up patch should work. | 14:46 |
| lucasagomes | I believe that class only since the "ident" is some word | 14:46 |
| rloo | lucasagomes: it is only for names right? | 14:46 |
| lucasagomes | yeah | 14:46 |
| lucasagomes | rloo, will do that, and fix the opmization thingy you pointed out too | 14:46 |
| rloo | lucasagomes: so only for nodes for now. but maybe we should make that change that sam suggested anyway. | 14:46 |
| lucasagomes | (and the nits) | 14:46 |
| lucasagomes | ack, 1 sec will cook a patch up for it | 14:47 |
| lucasagomes | rloo, btw, thanks for pointing it out | 14:47 |
| rloo | lucasagomes: yw. i debated about approving/not approving it. | 14:47 |
| rloo | lucasagomes: i figured we still needed to fix it so i approved it. | 14:48 |
| lucasagomes | fair enough | 14:48 |
| rloo | lucasagomes: i am going to try to focus more on documenting any decisions/thoughts we have, to reduce technical debt (or at least make it easier to remember later why we did whatever) | 14:48 |
| sambetts | rloo: +1 too many conversations are lost to the blackhole of IRC | 14:49 |
| lucasagomes | +1 | 14:49 |
| lucasagomes | some things may fall into our FAQ (which currently is quite poor) | 14:49 |
| rloo | sambetts: exactly. and i realize, even emails aren't sufficient. who is going to scroll back and look. | 14:49 |
| lucasagomes | ref: http://docs.openstack.org/developer/ironic/dev/faq.html | 14:50 |
| rloo | lucasagomes: yeah, faq is good too. but i think if it is something that is localized, can be documented in code, that would be easier for developers. | 14:50 |
| sambetts | rloo: I'd love to see something like the #info bot thingy added to our IRC which could be used to indicate decisions/conversations made in IRC | 14:51 |
| sambetts | so there could be an automated minutes pushed out on the mailer at the end of each day for example | 14:51 |
| lucasagomes | sambetts, PixieBot :-) | 14:51 |
| rloo | sambetts: i think that'd be useful, but i'd rather that info be elsewhere. code or faq seems like the best places where newbies would think to look. | 14:51 |
| lucasagomes | NobodyCam, you sitll working on that bot? | 14:51 |
| rloo | sambetts: not just newbies. folks like me with short memories too :) | 14:52 |
| * lucasagomes has short memory | 14:52 | |
| NobodyCam | oh poop I need to get back on that | 14:52 |
| rloo | sambetts: make it easier to review code if it is right there in your face :) | 14:52 |
| sambetts | Maybe it could automatically build a wiki page or something? With that days highlights :/ and a link to that point in the openstack evesdrop for more context | 14:53 |
| rloo | dtantsur: did you see https://etherpad.openstack.org/p/requirements-cruft, line 50 -- ironic-discoverd | 14:54 |
| *** d0ugal has joined #openstack-ironic | 14:55 | |
| dtantsur | rloo, yeah, my patch is on line 52 :) | 14:55 |
| dtantsur | but thanks for raising it! | 14:56 |
| *** Goneri has quit IRC | 14:56 | |
| rloo | dtantsur: ah, ok. let me update. cuz line 51 said 'i would check w/ironic folks ... | 14:56 |
| dtantsur | sure | 14:58 |
| *** sacharya has joined #openstack-ironic | 14:58 | |
| rloo | dtantsur: so you had a patch to remove it, but it hasn't been approved yet. and then https://review.openstack.org/#/c/313826/ was proposed? | 14:58 |
| rloo | dtantsur: shouldn't we abandon one of them? | 14:58 |
| dtantsur | correct | 14:58 |
| *** ijw_ has quit IRC | 14:58 | |
| rloo | dtantsur: ugh. ok, someone -1'd the second one cuz it is a dup of yours. | 14:58 |
| *** Guest30018 is now known as mfisch | 14:59 | |
| rloo | dtantsur: is it a no-no to edit stuff other people put in etherpads? | 14:59 |
| *** mfisch is now known as Guest56459 | 14:59 | |
| dtantsur | rloo, I thought that's what etherpads are for | 14:59 |
| dtantsur | they would use email otherwise | 15:00 |
| rloo | dtantsur: ok. am going to update that etherpad then. | 15:00 |
| rloo | dtantsur:updated and now i know that you've dealt with it :) | 15:02 |
| dtantsur | :) | 15:02 |
| mjturek1 | does anyone know if there's a ironic-neutron meeting today? the wiki says it's now in meeting-4 but it's a different team there | 15:04 |
| *** clenimar has joined #openstack-ironic | 15:04 | |
| sambetts | mjturek1: the time is in UTC so its in an hour | 15:05 |
| mjturek1 | sambetts: ahhh thanks. Must've been looking at the wrong timezone by accident | 15:06 |
| sambetts | :) | 15:06 |
| mjturek1 | :) | 15:06 |
| *** d0ugal has quit IRC | 15:08 | |
| *** yolanda has quit IRC | 15:09 | |
| *** Goneri has joined #openstack-ironic | 15:11 | |
| openstackgerrit | Imre Farkas proposed openstack/ironic: [WIP] DRAC RAID configuration https://review.openstack.org/176319 | 15:14 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Follow-up patch of 8e5e69869df476788b3ccf7e5ba6c2210a98fc8a https://review.openstack.org/314149 | 15:16 |
| lucasagomes | rloo, sambetts ^ lemme know if that's good enough in the comments | 15:16 |
| rloo | lucasagomes: will look in a few... | 15:17 |
| lucasagomes | cool, no rush | 15:18 |
| *** itamarl has quit IRC | 15:18 | |
| *** Nisha has joined #openstack-ironic | 15:19 | |
| *** yolanda has joined #openstack-ironic | 15:19 | |
| *** davideagnello has joined #openstack-ironic | 15:24 | |
| *** d0ugal has joined #openstack-ironic | 15:25 | |
| *** jistr has quit IRC | 15:33 | |
| * dtantsur relocates for the meeting, brb | 15:37 | |
| *** dtantsur is now known as dtantsur|bbl | 15:37 | |
| NobodyCam | rloo: I don't get the comment on the commit message for 314149 | 15:40 |
| rloo | NobodyCam: sec | 15:40 |
| NobodyCam | oh never mind I see it now | 15:41 |
| rloo | NobodyCam: ok :) | 15:41 |
| *** daemontool__ has quit IRC | 15:41 | |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: TinyIPA: Shave off some file size from tinyipa ramdisk https://review.openstack.org/313675 | 15:41 |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: TinyIPA: Precompile python code for faster load https://review.openstack.org/313103 | 15:41 |
| *** harlowja_at_home has joined #openstack-ironic | 15:43 | |
| *** rama_y has joined #openstack-ironic | 15:45 | |
| openstackgerrit | Andre keedy proposed openstack/ironic-inspector: Replace node data via uuid/data/unprocessed API https://review.openstack.org/308379 | 15:46 |
| *** ohamada has quit IRC | 15:46 | |
| *** ohamada has joined #openstack-ironic | 15:46 | |
| *** ohamada has quit IRC | 15:47 | |
| *** thrash is now known as thrash|lunch | 15:47 | |
| *** Sukhdev has joined #openstack-ironic | 15:47 | |
| *** fragatina has joined #openstack-ironic | 15:50 | |
| *** d0ugal has quit IRC | 15:51 | |
| *** wajdi has joined #openstack-ironic | 15:51 | |
| *** fragatina has quit IRC | 15:53 | |
| *** fragatin_ has joined #openstack-ironic | 15:53 | |
| *** jistr has joined #openstack-ironic | 15:54 | |
| stendulker | jroll: Hi | 15:57 |
| *** davideagnello has quit IRC | 15:58 | |
| *** hshiina has joined #openstack-ironic | 15:58 | |
| *** dprince has joined #openstack-ironic | 15:58 | |
| jroll | stendulker: hey, what's up? | 15:59 |
| stendulker | jroll: this is regarding adding proliantuils to ironic governance https://review.openstack.org/#/c/313667/ | 16:00 |
| stendulker | jroll: This is wrt CI discussion we had at summit, that CI information is fetched from stackalytics | 16:00 |
| jroll | stendulker: yes, what about it? | 16:00 |
| jroll | rather, what's your question? | 16:01 |
| stendulker | jroll: Was following dracclient for the same | 16:01 |
| *** tesseract has quit IRC | 16:01 | |
| stendulker | and have raised thsi review... Had raised another patch https://review.openstack.org/#/c/311277/ in driverlog that is merged | 16:01 |
| *** fragatin_ has quit IRC | 16:02 | |
| jroll | stendulker: sure, I've seen it, I put it up for discussion in the meeting today? | 16:02 |
| jroll | stendulker: if this is just for stackalytics CI things, not sure it's needed | 16:02 |
| stendulker | jroll: not sure on the impact of the same | 16:02 |
| stendulker | jroll: ok... | 16:02 |
| stendulker | jroll: if its incorrect, i can abandon the same.. | 16:03 |
| jroll | stendulker: well, it's a discussion I want to have with the community anyway, so I'll leave it until the meeting | 16:04 |
| stendulker | jroll: ok :) | 16:04 |
| *** milan has joined #openstack-ironic | 16:05 | |
| NobodyCam | jroll: as I understand release management would then become (or is going to become) the responsibility of the main projects PTL??? | 16:05 |
| *** rcernin has quit IRC | 16:06 | |
| *** afaranha has joined #openstack-ironic | 16:06 | |
| stendulker | jroll: thank you | 16:06 |
| *** d0ugal has joined #openstack-ironic | 16:06 | |
| *** keedya has joined #openstack-ironic | 16:06 | |
| jroll | NobodyCam: everything in the ironic governance will become managed by the release team. approval of those releases is up to the PTL *or* a *single* release liaison | 16:06 |
| jroll | NobodyCam: the release team wants one point of contact per "project" (project meaning e.g. everything under the ironic governance) | 16:07 |
| jroll | NobodyCam: that said, anyone can still propose a release, I just need to +1 it | 16:07 |
| *** fragatina has joined #openstack-ironic | 16:07 | |
| NobodyCam | ahh okay, Thank you that's helps my understanding | 16:08 |
| jroll | np | 16:08 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/virtualbmc: Add unittests for the utils.py module https://review.openstack.org/314197 | 16:11 |
| *** piet has quit IRC | 16:13 | |
| *** jjohnson2_ has quit IRC | 16:14 | |
| *** jjohnson2_ has joined #openstack-ironic | 16:14 | |
| openstackgerrit | Merged openstack/ironic-specs: Place priorities list in reverse chronological order https://review.openstack.org/312794 | 16:15 |
| rloo | folks, I forgot to send email out last week about it. If you are a primary contact for any of the newton priority items, please update the subteam reports in whiteboard. And others that are on subteams already -- you don't need this reminder :) | 16:15 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Follow-up patch of 8e5e69869df476788b3ccf7e5ba6c2210a98fc8a https://review.openstack.org/314149 | 16:20 |
| lucasagomes | rloo, NobodyCam ^ updated the commit message and the added the missing period | 16:20 |
| rloo | thx lucasagomes | 16:21 |
| NobodyCam | lol awesome lucasagomes Thank you | 16:22 |
| lucasagomes | :-) | 16:22 |
| * jlvillal goes to update the whiteboard... | 16:22 | |
| lucasagomes | ty for the review | 16:22 |
| jlvillal | dtantsur|bbl: I change 09.05.2016 to 09-May-2016. Since Europe and USA use different formatting :) To USA it looks like September 5, 2016 | 16:24 |
| jlvillal | For the whiteboard, gate broken status | 16:24 |
| NobodyCam | lucasagomes: +2 will hit +a as soon as we get test results | 16:24 |
| lucasagomes | cool ty | 16:24 |
| NobodyCam | jlvillal: did 314024 land? | 16:25 |
| jlvillal | NobodyCam: I don't know... | 16:25 |
| jlvillal | NobodyCam: I was only changing the date. | 16:25 |
| jlvillal | NobodyCam: I've been heads down this morning doing my (late) status report. And then Grenade stuff. So I haven't been paying attention :( | 16:26 |
| NobodyCam | jlvillal: I thought we were waiting for https://review.openstack.org/#/c/314024 to land | 16:27 |
| jroll | yes we are waiting for that to land | 16:31 |
| jlvillal | jroll: What would you say is the status of tempest smoke? | 16:34 |
| jroll | jlvillal: I haven't touched it since posting the one patch | 16:35 |
| jlvillal | thanks | 16:35 |
| jroll | jlvillal: I did have a question for you about that, I see smoke is ran before resource create | 16:35 |
| jroll | so is smoke passing for you, if you're getting that far? | 16:35 |
| jroll | (on the "old" version) | 16:35 |
| jlvillal | So in my devstack-gate-test I am having it do 'baremetal' for the tempest part. | 16:35 |
| jlvillal | So that can get farther along. Of course we need tempest smoke working at some point. | 16:36 |
| jroll | oh | 16:36 |
| jroll | which runs nothing | 16:36 |
| *** wajdi has quit IRC | 16:36 | |
| jlvillal | It runs about 70 tests or so. | 16:36 |
| jlvillal | I think | 16:36 |
| jlvillal | The same as we run in our normal Ironic gate. | 16:37 |
| jroll | oh, baremetal instead of smoke, not baremetal combined with smoke (which have no overlap) | 16:37 |
| jroll | right | 16:37 |
| jroll | okay, I didn't realize that wasn't representative of the typical grenade job | 16:37 |
| * jroll will keep hacking on it then | 16:37 | |
| *** rcernin has joined #openstack-ironic | 16:37 | |
| jlvillal | Awesome :) | 16:37 |
| jroll | jlvillal: fyi, I did deploy this so we can start making progress on the gate side https://review.openstack.org/#/c/314066/ | 16:39 |
| * jlvillal looks | 16:39 | |
| jlvillal | jroll: Awesome! :) | 16:41 |
| jlvillal | sambetts: We all love tinyipa :) | 16:41 |
| sambetts | jlvillal: there is a couple more optimisation patches up, I've got tinyIPA down to 49mb | 16:42 |
| jlvillal | jroll: I will update the devstack-gate-test to try 7 tinyipa vms | 16:42 |
| jlvillal | after meeting | 16:42 |
| jroll | jlvillal: well, these are needed for smoke tests, resource create isn't really affected by it | 16:43 |
| *** athomas has quit IRC | 16:43 | |
| jroll | sambetts: have you tried running with 128 or 256 MB of ram yet? :) | 16:43 |
| * jlvillal is working on getting it running on his new home system. To avoid proxy hell... | 16:43 | |
| sambetts | jroll: no, but I really need to because getting rid of the squashfs mounting in the boot process might have helps to get the ram requirement down even further | 16:44 |
| *** moshele has quit IRC | 16:45 | |
| *** david-lyle_ is now known as david-lyle | 16:45 | |
| *** nathan-h has quit IRC | 16:46 | |
| *** wajdi has joined #openstack-ironic | 16:47 | |
| jroll | sambetts: yeah, would love to see that :) | 16:47 |
| sambetts | jroll: yeah that would be a huge improvement | 16:48 |
| *** ijw has joined #openstack-ironic | 16:49 | |
| *** derekh has quit IRC | 16:50 | |
| *** ijw has quit IRC | 16:56 | |
| *** dtantsur|bbl is now known as dtantsur | 16:57 | |
| *** penick has joined #openstack-ironic | 16:57 | |
| dtantsur | jlvillal, heh, thanks :) | 16:57 |
| dtantsur | sambetts, 49mb wow. man you rock | 16:57 |
| *** xhku has quit IRC | 16:59 | |
| dtantsur | sambetts, speaking of tinyipa. we can partially work around missing logs by setting up a python logging handler to dump IPA logs into a file and send it to inspector as well. wdyt? | 16:59 |
| dtantsur | (this is for case when journald and/or syslog are not present) | 16:59 |
| sambetts | dtantsur: teeing IPA logs to a file makes sense to me | 17:00 |
| *** keedya has quit IRC | 17:00 | |
| sambetts | wouldn't get the boot logs like you would from journald or syslog though :/ | 17:02 |
| *** jistr has quit IRC | 17:02 | |
| dtantsur | sambetts, yes, yes... better than nothing, I guess | 17:03 |
| *** harlowja_at_home has quit IRC | 17:04 | |
| *** ijw has joined #openstack-ironic | 17:05 | |
| *** thrash|lunch is now known as thrash | 17:05 | |
| *** yolanda has quit IRC | 17:06 | |
| dtantsur | a stable backport if someone has some time: https://review.openstack.org/#/c/314147/ | 17:06 |
| *** daemontool__ has joined #openstack-ironic | 17:09 | |
| *** ijw has quit IRC | 17:10 | |
| *** nicodemos has joined #openstack-ironic | 17:10 | |
| mgould | dtantsur, +1 | 17:11 |
| dtantsur | thnx | 17:11 |
| *** yolanda has joined #openstack-ironic | 17:12 | |
| *** cdearborn has joined #openstack-ironic | 17:15 | |
| *** wajdi has quit IRC | 17:32 | |
| *** davideagnello has joined #openstack-ironic | 17:32 | |
| *** ijw has joined #openstack-ironic | 17:32 | |
| *** ijw has quit IRC | 17:33 | |
| mgould | I have an Ironic node which failed to deploy properly | 17:34 |
| mgould | now Nova thinks that instance is holding the node, so it won't let me boot another baremetal node | 17:34 |
| *** ijw has joined #openstack-ironic | 17:34 | |
| mgould | how do I get it to release the lock? | 17:34 |
| dtantsur | mgould, did you try 'nova delete' on the instance? | 17:35 |
| mgould | dtantsur, yes | 17:35 |
| *** ppiela has quit IRC | 17:35 | |
| mgould | and it succeeded | 17:35 |
| lucasagomes | mgould, the ironic node still have the instance_uuid field set? | 17:36 |
| mgould | lucasagomes, yes | 17:38 |
| *** ijw has quit IRC | 17:39 | |
| lucasagomes | mgould, right, so try ironic node-update <node uuid> remove instance_uuid | 17:41 |
| lucasagomes | that should free it up | 17:41 |
| *** davideagnello has quit IRC | 17:43 | |
| mgould | lucasagomes, thanks! | 17:44 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: Update the api-ref documentation for Drivers https://review.openstack.org/313708 | 17:45 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: Update the api-ref documentation for Ports https://review.openstack.org/313187 | 17:45 |
| lucasagomes | yw :-) | 17:45 |
| *** elfosardo has joined #openstack-ironic | 17:45 | |
| *** harlowja has joined #openstack-ironic | 17:48 | |
| *** Sukhdev has quit IRC | 17:50 | |
| *** garthb has joined #openstack-ironic | 17:51 | |
| *** wajdi has joined #openstack-ironic | 17:56 | |
| *** penick has quit IRC | 17:56 | |
| *** Sukhdev has joined #openstack-ironic | 17:57 | |
| *** Nisha_away has joined #openstack-ironic | 17:57 | |
| *** davideagnello has joined #openstack-ironic | 17:58 | |
| *** ChubYann has joined #openstack-ironic | 17:59 | |
| *** penick has joined #openstack-ironic | 18:00 | |
| *** Nisha has quit IRC | 18:00 | |
| * mgould -> home; good night! | 18:02 | |
| *** mgould has quit IRC | 18:02 | |
| NobodyCam | night mgould | 18:02 |
| lucasagomes | I'm also going to call it a day | 18:02 |
| *** penick has quit IRC | 18:02 | |
| *** ppiela has joined #openstack-ironic | 18:02 | |
| lucasagomes | have a great evening all, see you tomorrow :-) | 18:02 |
| dtantsur | g'night folks | 18:02 |
| dtantsur | g'night lucasagomes | 18:02 |
| NobodyCam | night dtantsur | 18:02 |
| NobodyCam | night lucasagomes | 18:03 |
| *** dtantsur is now known as dtantsur|afk | 18:03 | |
| *** lucasagomes is now known as lucas-dinner | 18:04 | |
| jroll | bbiab | 18:04 |
| stendulker | jroll: A query, being in ironic governance would allow making a relase of proliantutils off-cycle of openstack release, say wit defect fixes or so... | 18:07 |
| jroll | stendulker: yes, you'll be able to release at any time (other than during release freezes), via a patch to the openstack/releases repo with a +1 from me | 18:07 |
| *** ifarkas has quit IRC | 18:08 | |
| stendulker | jroll: Are there any additional processes that we may have to follow? I could not get any link, hence these questions. If there is any lik i would go through the same | 18:08 |
| jroll | stendulker: don't believe so | 18:08 |
| *** hshiina has quit IRC | 18:10 | |
| stendulker | jroll: thank you | 18:10 |
| jroll | np | 18:11 |
| *** wajdi_ has joined #openstack-ironic | 18:16 | |
| *** penick has joined #openstack-ironic | 18:17 | |
| openstackgerrit | Merged openstack/ironic-inspector: Updated from global requirements https://review.openstack.org/313730 | 18:18 |
| *** wajdi__ has joined #openstack-ironic | 18:19 | |
| *** wajdi has quit IRC | 18:19 | |
| *** electrofelix has quit IRC | 18:21 | |
| *** Nisha_away has quit IRC | 18:21 | |
| *** wajdi_ has quit IRC | 18:22 | |
| *** ijw has joined #openstack-ironic | 18:29 | |
| *** raildo has joined #openstack-ironic | 18:31 | |
| raildo | jroll: hey, if you have some free time, can you take a look on this patch later? https://review.openstack.org/#/c/313690 :) | 18:32 |
| *** openstackgerrit has quit IRC | 18:33 | |
| *** openstackgerrit has joined #openstack-ironic | 18:33 | |
| *** jcoufal has quit IRC | 18:47 | |
| *** mtanino has quit IRC | 18:47 | |
| *** keedya has joined #openstack-ironic | 18:49 | |
| *** yolanda has quit IRC | 18:51 | |
| *** dprince has quit IRC | 18:53 | |
| *** piet has joined #openstack-ironic | 18:53 | |
| *** irf has joined #openstack-ironic | 18:53 | |
| *** nicodemos has quit IRC | 18:53 | |
| irf | hello, i am back .. | 18:54 |
| *** AJaeger has joined #openstack-ironic | 18:54 | |
| irf | nova hypervisor-list command shows total 5 hypervisors ... | 18:55 |
| *** AJaeger has left #openstack-ironic | 18:55 | |
| irf | but in reality only two nova compute services are runnimg .... | 18:55 |
| irf | how come command is showing 5 hypervisors ... | 18:56 |
| *** baoli has quit IRC | 18:56 | |
| *** cloudnul- has joined #openstack-ironic | 18:56 | |
| *** cloudnul- has quit IRC | 18:57 | |
| *** stendulker has quit IRC | 18:58 | |
| *** ijw has quit IRC | 18:58 | |
| *** ijw has joined #openstack-ironic | 18:59 | |
| *** jtomasek_ has joined #openstack-ironic | 19:00 | |
| *** dprince has joined #openstack-ironic | 19:00 | |
| *** jcoufal has joined #openstack-ironic | 19:01 | |
| openstackgerrit | Andre keedy proposed openstack/ironic-inspector: Replace node data via uuid/data/unprocessed API https://review.openstack.org/308379 | 19:01 |
| sambetts | irf: every ironic node appears as a hypervisor that can support exactly 1 VM | 19:06 |
| sambetts | so 4 Ironic nodes == 4 hypervisors + your virtual machine one == 5 totla | 19:07 |
| sambetts | total | 19:07 |
| *** garthb has quit IRC | 19:07 | |
| irf | ok | 19:08 |
| *** spandhe has joined #openstack-ironic | 19:08 | |
| *** Sukhdev has quit IRC | 19:08 | |
| irf | ok.. now as i have configured the nova compute service for baremetal node ... it seems there is still an issue ... | 19:09 |
| irf | still getting same error : no valid hosst was found... | 19:10 |
| irf | again some link is missing i guess. | 19:11 |
| irf | here is the output for hypervisor-stats command | 19:12 |
| irf | http://pastebin.com/kSGyMAZQ | 19:12 |
| sambetts | do you know if those stats have increased since the Ironic nodes have appeared as hypervisors? | 19:14 |
| *** Sukhdev has joined #openstack-ironic | 19:15 | |
| irf | yes .. it has increased | 19:15 |
| irf | but one thing is there ,, only one hypervisor is valid among all 4 ironic hypervisos ... | 19:16 |
| irf | shud we delete the rest of the hypervisors ? | 19:16 |
| sambetts | no, having 5 hypervisors is right if you have 1 VM hypervisor and 4 ironic nodes | 19:18 |
| irf | ironic node means bare metal machine rite ? | 19:20 |
| sambetts | yes | 19:20 |
| irf | ok so i have just one ironic node | 19:21 |
| sambetts | when you pasted the output of ironic node-list the other day I saw more than one | 19:22 |
| irf | here is teh output of ironic node-list : http://pastebin.com/eYKjjKPy | 19:24 |
| irf | and only last entry is valid | 19:24 |
| *** jjohnson2_ has quit IRC | 19:25 | |
| *** jjohnson2 has joined #openstack-ironic | 19:25 | |
| irf | shud i remove the nodes which are not valid ? | 19:25 |
| sambetts | all ironic nodes will show up in nova once the nova compute is talking to Ironic, but the hypervisor stats will only increase by the number of nodes that are availiable and pass node validate | 19:26 |
| sambetts | or are in mainteance | 19:26 |
| sambetts | aren't in mainteance | 19:26 |
| irf | ok.. so i think i shud remove the nodes which are not valid | 19:26 |
| sambetts | that would help to debug :) | 19:28 |
| devananda | rloo: your email this morning is formatted very strangely | 19:28 |
| rloo | devananda: :-(, in meeting now. trying different mail system. | 19:29 |
| jroll | devananda: rloo: looks fine to me in plain text | 19:36 |
| *** mbound has joined #openstack-ironic | 19:36 | |
| irf | ok.. i have removed the invalid nodes .. shud i restart the services and try again ? | 19:37 |
| rloo | jroll, devananda: looking. it seems ok although there are lots of blank space. I sent that email via Outlook instead of google. | 19:38 |
| jroll | yeah agree | 19:38 |
| *** baoli has joined #openstack-ironic | 19:44 | |
| *** baoli has quit IRC | 19:47 | |
| *** mbound has quit IRC | 19:47 | |
| *** baoli has joined #openstack-ironic | 19:47 | |
| *** Sukhdev has quit IRC | 19:47 | |
| *** [1]cdearborn has joined #openstack-ironic | 19:49 | |
| *** afaranha has quit IRC | 19:50 | |
| sambetts | irf: you shouldn't need to restart the services, they should pick up the changes automatically | 19:50 |
| *** [2]cdearborn has joined #openstack-ironic | 19:54 | |
| *** [1]cdearborn has quit IRC | 19:54 | |
| *** irf has quit IRC | 19:57 | |
| *** irf has joined #openstack-ironic | 19:58 | |
| *** ijw has quit IRC | 20:00 | |
| irf | ok | 20:00 |
| *** ijw has joined #openstack-ironic | 20:00 | |
| *** jcoufal has quit IRC | 20:01 | |
| jroll | TheJulia: cinerama: devananda: NobodyCam: there's a bifrost+kolla thread on the ML, in case you haven't seen it | 20:02 |
| *** fragatina has quit IRC | 20:04 | |
| *** cdearborn has quit IRC | 20:05 | |
| *** [2]cdearborn is now known as cdearborn | 20:05 | |
| *** lucas-dinner has quit IRC | 20:06 | |
| jroll | betherly: krotscheck: https://review.openstack.org/#/c/313270/ needs someone to do the needful | 20:08 |
| NobodyCam | cool Ty jroll :) | 20:09 |
| betherly | jroll: +2'd it :) | 20:10 |
| *** spandhe has quit IRC | 20:11 | |
| jroll | cools, thanks | 20:12 |
| *** baoli has quit IRC | 20:13 | |
| *** baoli has joined #openstack-ironic | 20:14 | |
| krotscheck | jroll: Have pushed the button | 20:14 |
| jroll | ty | 20:14 |
| *** baoli has quit IRC | 20:14 | |
| *** baoli has joined #openstack-ironic | 20:15 | |
| devananda | jroll: yea, replying now | 20:15 |
| *** rcernin has quit IRC | 20:15 | |
| devananda | jroll: do you know sean mooney's irc nic? | 20:15 |
| *** lucasagomes has joined #openstack-ironic | 20:16 | |
| jroll | devananda: not off the top of my head, no | 20:16 |
| *** ChrisAusten has quit IRC | 20:17 | |
| *** lucasagomes is now known as lucas-dinner | 20:17 | |
| *** hurgleburgler has joined #openstack-ironic | 20:20 | |
| *** hurgleburgler has left #openstack-ironic | 20:20 | |
| *** baoli has quit IRC | 20:21 | |
| *** cdearborn has quit IRC | 20:23 | |
| *** fragatina has joined #openstack-ironic | 20:23 | |
| *** cdearborn has joined #openstack-ironic | 20:24 | |
| *** sambetts is now known as sambetts|afk | 20:29 | |
| harlowja | jroll https://review.openstack.org/#/c/313699/ if u get some time | 20:32 |
| harlowja | its the imho much more sane version of that spec matcher stuff | 20:33 |
| harlowja | with an actual well defined grammar and all | 20:33 |
| jroll | harlowja: I don't know that code well at all, so changes to it scare the heck out of me | 20:34 |
| *** dprince has quit IRC | 20:34 | |
| jroll | but tests do pass, well done :) | 20:34 |
| harlowja | that should mean this new change is better, cause u can at least view the grammar :-P | 20:34 |
| jroll | tbh though, that code scares most people already, and imo this looks scarier | 20:34 |
| harlowja | https://review.openstack.org/#/c/313699/8/oslo_utils/specs_matcher.py (See make grammer function) | 20:34 |
| harlowja | hmmmm, a well defined grammar scares people ? | 20:35 |
| jroll | at any rate, I'd have nova folks look | 20:35 |
| jroll | the code is the scary part, not the fact it's well defined | 20:35 |
| harlowja | hmmm, confused, lol | 20:35 |
| jroll | it's much more difficult to reason about by reading | 20:36 |
| jroll | (imo) | 20:36 |
| jroll | unless you've written a grammar before :) | 20:36 |
| harlowja | right, don't create informal DSLs is the story of that one | 20:36 |
| harlowja | always create formal ones | 20:36 |
| jroll | anyway, I would see what nova folks think about it | 20:36 |
| harlowja | cause informal ones u get half-made-up-stuff | 20:37 |
| harlowja | at least this one has a grammar that can be read by folks who can read it :-P | 20:37 |
| harlowja | jroll who in nova knows that stuff :-P | 20:37 |
| *** ChrisAusten has joined #openstack-ironic | 20:38 | |
| jroll | harlowja: great question :) | 20:38 |
| jroll | vish? :P | 20:38 |
| jroll | ooo I know ask termie | 20:38 |
| harlowja | any people that are still around :-P | 20:38 |
| harlowja | and/or not in secret companies | 20:38 |
| harlowja | lol | 20:38 |
| *** baoli has joined #openstack-ironic | 20:38 | |
| jroll | I have no clue | 20:39 |
| jroll | that's part of the reason nobody wants to touch this | 20:39 |
| jroll | it also isn't well tested outside of unit tests | 20:39 |
| *** e0ne has joined #openstack-ironic | 20:39 | |
| jroll | because this is primarily for arbitrary flavor<->compute node matching based on json blobs | 20:39 |
| harlowja | lol | 20:39 |
| harlowja | well at least now there is a real grammar :-P | 20:40 |
| harlowja | with a real evaulator | 20:40 |
| jroll | sure, I just don't see how useful that is | 20:40 |
| *** baoli has quit IRC | 20:40 | |
| jroll | also | 20:40 |
| jroll | how would this work on top of your patch? do I need to pass an entire grammar in? https://review.openstack.org/#/c/308399/2 | 20:40 |
| * jroll assumes so | 20:41 | |
| *** baoli has joined #openstack-ironic | 20:42 | |
| harlowja | shouldn't be to hard to do that | 20:42 |
| harlowja | without whole grammer | 20:42 |
| harlowja | so nothing drastic for that | 20:42 |
| jroll | not really seeing how at a glance | 20:42 |
| harlowja | k, one sec | 20:42 |
| *** ChrisAusten has quit IRC | 20:42 | |
| * harlowja cooks up a review | 20:43 | |
| jroll | harlowja: I mean, you don't need to show me | 20:43 |
| harlowja | nope to late | 20:43 |
| harlowja | i'm already cooking | 20:43 |
| jroll | realistically I think we need nova buy in on completely re-writing that | 20:43 |
| jroll | I'm apathetic as long as zero deployments break :P | 20:44 |
| *** fragatina has quit IRC | 20:44 | |
| harlowja | so that means we need to find vish, lol | 20:44 |
| *** baoli has quit IRC | 20:47 | |
| jroll | heh | 20:48 |
| *** baoli has joined #openstack-ironic | 20:48 | |
| harlowja | https://review.openstack.org/314274 (how to override some things) | 20:48 |
| harlowja | anyways, i'll find vish lol | 20:50 |
| jroll | heh | 20:50 |
| * harlowja heads over to the oracle buildings | 20:50 | |
| jroll | was also going to ask, what about _make_grammar(), that's where my concern was | 20:51 |
| *** jcoufal has joined #openstack-ironic | 20:51 | |
| harlowja | ah, u want to chop pieces out of the grammar | 20:51 |
| harlowja | sounds like a bad idea, lol | 20:51 |
| jroll | I want to be able to use a subset, yes | 20:52 |
| jroll | or superset | 20:52 |
| jroll | or just my own set of methods | 20:52 |
| jroll | I mean, if you asked anyone today, they'd probably say all of this is a bad idea | 20:53 |
| jroll | but this is the hand we were dealt | 20:53 |
| harlowja | x:-P | 20:53 |
| harlowja | :-P i mean | 20:53 |
| harlowja | ya, so at that point i'd wonder if its better to define your own make_grammar and leave the evaluation as is | 20:53 |
| harlowja | and that means figuring out the grammar u want | 20:53 |
| jroll | or just leave the code as-is >.> | 20:54 |
| jroll | like, an informal DSL thing is bad, but it isn't *that* bad | 20:54 |
| harlowja | in a unsupportable/scary to people state :-P | 20:54 |
| harlowja | but ya, i get the viewpoint | 20:55 |
| harlowja | to me having a DSL that is informal is worse than having one thats formal (at least it forces u to be explicit about the actual DSL) | 20:55 |
| harlowja | but ya, point taken | 20:56 |
| *** baoli has quit IRC | 20:59 | |
| *** wajdi__ has quit IRC | 21:00 | |
| *** wajdi has joined #openstack-ironic | 21:01 | |
| *** Sukhdev has joined #openstack-ironic | 21:01 | |
| *** wajdi_ has joined #openstack-ironic | 21:02 | |
| *** ChrisAusten has joined #openstack-ironic | 21:04 | |
| *** rloo has quit IRC | 21:05 | |
| *** rloo has joined #openstack-ironic | 21:05 | |
| *** baoli has joined #openstack-ironic | 21:06 | |
| *** wajdi has quit IRC | 21:06 | |
| *** Goneri has quit IRC | 21:06 | |
| *** ChrisAusten has quit IRC | 21:07 | |
| *** fragatina has joined #openstack-ironic | 21:11 | |
| *** ChrisAusten has joined #openstack-ironic | 21:11 | |
| *** Guest56459 is now known as mfisch | 21:15 | |
| *** mfisch is now known as Guest87085 | 21:15 | |
| *** spandhe has joined #openstack-ironic | 21:17 | |
| *** Guest87085 is now known as mfisch | 21:19 | |
| *** mfisch has quit IRC | 21:20 | |
| *** mfisch has joined #openstack-ironic | 21:20 | |
| *** mjturek1 has quit IRC | 21:22 | |
| *** jjohnson2 has quit IRC | 21:25 | |
| *** d0ugal has quit IRC | 21:26 | |
| persia | I ran into an interesting use case today: wanting to remotely drive network-attached hardware that can neither run linux nor has a BMC, but does have an API to load code from the network. Is this something that is Ironic? If so, which bit should I look at? | 21:33 |
| persia | (more specifically, these are PCI-attached coprocessor cards with high-speed network interfaces, mostly only using PCI for power, rather than being PCI-passthrough candidates) | 21:34 |
| *** irf has quit IRC | 21:34 | |
| *** jtomasek_ has quit IRC | 21:38 | |
| *** jcoufal has quit IRC | 21:41 | |
| devananda | persia: can you describe a use case for this hardware? | 21:42 |
| *** e0ne has quit IRC | 21:42 | |
| mrda | Morning all | 21:43 |
| devananda | persia: I'm curious if our logical modelling maps to it or not. you might look at http://docs.openstack.org/developer/ironic/api/ironic.drivers.base.html#ironic.drivers.base.BaseDriver | 21:44 |
| *** mbound has joined #openstack-ironic | 21:46 | |
| *** thrash is now known as thrash|g0ne | 21:48 | |
| *** mbound has quit IRC | 21:48 | |
| *** mbound has joined #openstack-ironic | 21:48 | |
| *** wajdi has joined #openstack-ironic | 21:50 | |
| mrda | jroll, odyssey4me: So just on keystone v2/v3 thing, we did have trouble getting the right collection of params together working (even after talking to a keystone SME) which ended us settling on what we have today. I can re-test if we think things have changed here. | 21:51 |
| *** wajdi_ has quit IRC | 21:53 | |
| jroll | mrda: yeah, okay. thanks | 21:59 |
| *** wajdi has quit IRC | 21:59 | |
| jroll | fyi, I recap'd the summit http://jroll.ghost.io/newton-summit-recap/ | 21:59 |
| *** rama_y has quit IRC | 22:00 | |
| *** Sukhdev has quit IRC | 22:00 | |
| *** catintheroof has quit IRC | 22:01 | |
| *** amotoki has joined #openstack-ironic | 22:03 | |
| mrda | jroll: excellent! | 22:05 |
| *** cdearborn has quit IRC | 22:08 | |
| persia | devananda: one use case would be to run FPGA kernels to accelerate analytic workloads (the current common practice is to use PCI pass through and Nova to access FPGAs in an OpenStack environment) | 22:08 |
| persia | Indeed, I think one would be able to compose a BaseDriver interface to do that which is needed. I will need to dig more to check, but that seems a sensible place to start. Thanks! | 22:11 |
| *** rbudden has quit IRC | 22:16 | |
| *** ChrisAusten has quit IRC | 22:23 | |
| jroll | sent out a poll about the midcycle | 22:24 |
| jroll | see y'all tomorrow | 22:24 |
| *** piet has quit IRC | 22:30 | |
| *** sacharya has quit IRC | 22:37 | |
| *** sambetts|afk has quit IRC | 22:43 | |
| *** keedya has quit IRC | 22:45 | |
| *** sambetts has joined #openstack-ironic | 22:47 | |
| *** ChrisAusten has joined #openstack-ironic | 22:52 | |
| *** ametts has quit IRC | 22:58 | |
| *** ijw has quit IRC | 23:07 | |
| *** ijw has joined #openstack-ironic | 23:07 | |
| *** yuanying has joined #openstack-ironic | 23:14 | |
| *** mbound has quit IRC | 23:18 | |
| *** rloo has quit IRC | 23:20 | |
| openstackgerrit | Michael Still proposed openstack/ironic: Cleanup unused conf variables https://review.openstack.org/280493 | 23:23 |
| *** mbound has joined #openstack-ironic | 23:25 | |
| *** penick has quit IRC | 23:28 | |
| *** chlong has joined #openstack-ironic | 23:33 | |
| *** ChrisAusten has quit IRC | 23:33 | |
| *** ijw has quit IRC | 23:35 | |
| *** ijw has joined #openstack-ironic | 23:35 | |
| *** sacharya has joined #openstack-ironic | 23:37 | |
| *** spandhe has quit IRC | 23:39 | |
| *** sacharya has quit IRC | 23:42 | |
| *** mbound has quit IRC | 23:46 | |
| *** gabriel-bezerra has quit IRC | 23:47 | |
| *** rloo has joined #openstack-ironic | 23:54 | |
| *** ijw has quit IRC | 23:56 | |
| *** ijw_ has joined #openstack-ironic | 23:56 | |
| *** ChrisAusten has joined #openstack-ironic | 23:56 | |
| *** Sukhdev has joined #openstack-ironic | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!