*** dmellado has quit IRC | 00:03 | |
*** dmellado has joined #openstack-ironic | 00:04 | |
*** hwoarang has quit IRC | 00:05 | |
*** hwoarang has joined #openstack-ironic | 00:09 | |
openstackgerrit | Eric Fried proposed openstack/python-ironicclient master: WIP: AdapterClient: accept an already-constructed ksa adapter https://review.openstack.org/641169 | 00:15 |
---|---|---|
*** hoonetorg has quit IRC | 00:24 | |
*** hoonetorg has joined #openstack-ironic | 00:26 | |
openstackgerrit | guang-yee proposed openstack/python-ironicclient master: pass endpoint interface to http client https://review.openstack.org/640491 | 00:43 |
*** rloo has quit IRC | 01:02 | |
*** sthussey has quit IRC | 01:04 | |
*** whoami-rajat has joined #openstack-ironic | 01:09 | |
*** oanson has quit IRC | 01:13 | |
*** mmethot has joined #openstack-ironic | 01:15 | |
*** gyee has quit IRC | 01:21 | |
*** sdake has quit IRC | 01:29 | |
openstackgerrit | Lars Kellogg-Stedman proposed openstack/virtualbmc master: [WIP] Add Serial-over-LAN (SOL) support https://review.openstack.org/640888 | 01:41 |
*** sdake has joined #openstack-ironic | 01:46 | |
*** jhesketh has quit IRC | 01:57 | |
*** jhesketh has joined #openstack-ironic | 01:58 | |
*** sdake has quit IRC | 02:05 | |
*** hwoarang has quit IRC | 02:14 | |
*** hwoarang has joined #openstack-ironic | 02:16 | |
openstackgerrit | Merged openstack/networking-baremetal master: Supporting all py3 environments with tox https://review.openstack.org/640794 | 02:18 |
*** dsneddon has quit IRC | 02:21 | |
*** sdake has joined #openstack-ironic | 02:28 | |
*** Qianbiao has joined #openstack-ironic | 02:33 | |
Qianbiao | Morning | 02:33 |
Qianbiao | I try to deploy a node in standalone mode now. | 02:33 |
Qianbiao | It failed, and error is: "Failed to start the iSCSI target to deploy the node 2e05538a-d07c-40b5-8714-28278adb6a2d. Error: {u'message': u'Error finding the disk or partition device to deploy the image onto: No suitable device was found for deployment - root device hints were not provided and all found block devices are smaller than 4294967296B.', u'code': 404, u'type': u'DeviceNotFound', u'details': u'No suitable device was found for | 02:34 |
Qianbiao | deployment - root device hints were not provided and all found block devices are smaller than 4294967296B.'}" | 02:34 |
Qianbiao | any hints? | 02:34 |
Qianbiao | The disk looks all right. But in coreos cmd line, fdisk -l doesnot show those disks. | 02:35 |
*** sdake has quit IRC | 02:42 | |
*** dsneddon has joined #openstack-ironic | 02:45 | |
*** dsneddon has quit IRC | 02:57 | |
*** dsneddon has joined #openstack-ironic | 03:07 | |
*** hwoarang has quit IRC | 03:12 | |
*** dsneddon has quit IRC | 03:13 | |
*** hwoarang has joined #openstack-ironic | 03:16 | |
*** sdake has joined #openstack-ironic | 03:20 | |
*** sdake has quit IRC | 03:23 | |
*** sdake_ has joined #openstack-ironic | 03:24 | |
*** dsneddon has joined #openstack-ironic | 03:25 | |
*** dsneddon has quit IRC | 03:32 | |
*** dsneddon has joined #openstack-ironic | 03:44 | |
*** dsneddon has quit IRC | 03:53 | |
openstackgerrit | Lin Yang proposed openstack/sushy master: Fix wrong default JsonDataReader() argument https://review.openstack.org/641146 | 03:55 |
*** sdake_ has quit IRC | 03:58 | |
*** dsneddon has joined #openstack-ironic | 04:00 | |
*** dsneddon has quit IRC | 04:05 | |
*** hwoarang has quit IRC | 04:12 | |
*** hwoarang has joined #openstack-ironic | 04:18 | |
*** dsneddon has joined #openstack-ironic | 04:35 | |
*** dsneddon has quit IRC | 04:40 | |
*** sdake has joined #openstack-ironic | 04:49 | |
*** sdake has quit IRC | 04:49 | |
*** sdake has joined #openstack-ironic | 04:54 | |
*** sdake has quit IRC | 04:54 | |
*** dsneddon has joined #openstack-ironic | 04:54 | |
*** sdake has joined #openstack-ironic | 04:58 | |
*** sdake has quit IRC | 04:58 | |
*** dsneddon has quit IRC | 04:59 | |
*** sdake has joined #openstack-ironic | 05:10 | |
*** sdake has quit IRC | 05:10 | |
openstackgerrit | Kaifeng Wang proposed openstack/ironic-inspector master: Revise api-ref docs https://review.openstack.org/613515 | 05:11 |
openstackgerrit | Kaifeng Wang proposed openstack/ironic-inspector master: Revise api-ref docs https://review.openstack.org/613515 | 05:13 |
*** ianychoi_ has joined #openstack-ironic | 05:24 | |
*** dsneddon has joined #openstack-ironic | 05:27 | |
*** ianychoi has quit IRC | 05:28 | |
*** dsneddon has quit IRC | 05:32 | |
*** hoonetorg has quit IRC | 05:33 | |
*** sdake has joined #openstack-ironic | 05:38 | |
*** johnthetubaguy has quit IRC | 05:49 | |
*** hoonetorg has joined #openstack-ironic | 05:50 | |
*** dsneddon has joined #openstack-ironic | 06:04 | |
*** dsneddon has quit IRC | 06:09 | |
*** gkadam has quit IRC | 06:13 | |
*** dsneddon has joined #openstack-ironic | 06:16 | |
*** dsneddon has quit IRC | 06:21 | |
*** hwoarang has quit IRC | 06:26 | |
*** hwoarang has joined #openstack-ironic | 06:28 | |
*** e0ne has joined #openstack-ironic | 06:31 | |
*** v12aml has quit IRC | 06:32 | |
*** v12aml has joined #openstack-ironic | 06:39 | |
*** hjensas has quit IRC | 06:50 | |
*** dsneddon has joined #openstack-ironic | 07:04 | |
*** dsneddon has quit IRC | 07:08 | |
*** dsneddon has joined #openstack-ironic | 07:20 | |
*** dsneddon has quit IRC | 07:25 | |
arne_wiebalck | good morning, ironic | 07:33 |
arne_wiebalck | Qianbiao: I think the IPA uses ‘lsblk’ (not fdisk). | 07:36 |
*** hjensas has joined #openstack-ironic | 07:40 | |
*** dsneddon has joined #openstack-ironic | 07:44 | |
*** e0ne has quit IRC | 07:47 | |
*** dsneddon has quit IRC | 07:48 | |
*** dsneddon has joined #openstack-ironic | 07:50 | |
Qianbiao | arne_wiebalck thanks will try | 07:52 |
arne_wiebalck | Qianbiao: coreos is the OS of the deploy image, right? | 07:53 |
*** e0ne has joined #openstack-ironic | 07:54 | |
*** rpittau|afk is now known as rpittau | 08:02 | |
rpittau | good morning ironic! o/ | 08:02 |
*** dtantsur|afk is now known as dtantsur | 08:03 | |
dtantsur | morning ironic | 08:03 |
*** dtantsur is now known as dtantsur|mtg | 08:03 | |
dtantsur|mtg | folks, it looks like the standalone job is broken, could someone take a look? | 08:05 |
rpittau | dtantsur|mtg, I'm checking the job logs for https://review.openstack.org/639050 | 08:07 |
patchbot | patch 639050 - ironic - Allow building configdrive from JSON in the API - 10 patch sets | 08:07 |
dtantsur|mtg | thanks! | 08:08 |
dtantsur|mtg | it may be related to the newly built IPA images | 08:08 |
dtantsur|mtg | probably we just need more RAM? | 08:08 |
rpittau | maybe, it happened the same with tinyipa | 08:08 |
*** dsneddon has quit IRC | 08:14 | |
*** gkadam has joined #openstack-ironic | 08:14 | |
*** e0ne has quit IRC | 08:16 | |
*** amoralej|off is now known as amoralej | 08:19 | |
openstackgerrit | Mark Goddard proposed openstack/ironic-tempest-plugin master: Deploy templates: API test nits https://review.openstack.org/641278 | 08:21 |
*** dsneddon has joined #openstack-ironic | 08:22 | |
*** etingof has quit IRC | 08:24 | |
*** sburette has joined #openstack-ironic | 08:25 | |
*** dsneddon has quit IRC | 08:26 | |
*** pcaruana has joined #openstack-ironic | 08:29 | |
*** tssurya has joined #openstack-ironic | 08:33 | |
openstackgerrit | Riccardo Pittau proposed openstack/ironic master: [DNM] testing ironic-standalone job https://review.openstack.org/641281 | 08:48 |
Qianbiao | arne_wiebalck yes | 08:52 |
*** dsneddon has joined #openstack-ironic | 08:59 | |
iurygregory | morning everyone o/ | 09:07 |
*** andrein has joined #openstack-ironic | 09:17 | |
*** S4ren has joined #openstack-ironic | 09:21 | |
mbuil | dtantsur: I need some help again from an expert like you :). It takes more than an hour for the image to be dumped in the hard drive. However, I only saw HTTP traffic two minutes before it finished. It seems to me that IPA is stuck somewhere for a long time. Any idea of what might be happening or where to look at to debug the issue? thx! | 09:28 |
mbuil | dtantsur: I checked the network and it is fine and also the hard drive seems fine (I tried in 4 different servers with the same result :( ). And again, the HTTP traffic only appears after more than an hour | 09:29 |
*** lekhikadugtal has joined #openstack-ironic | 09:34 | |
lekhikadugtal | etingof: Is it possible to be suggested with some stories ? I went through all the stories and I'm pretty confused regarding where to begin with. | 09:36 |
*** derekh has joined #openstack-ironic | 09:37 | |
*** lekhikadugtal has quit IRC | 09:42 | |
*** moshele has joined #openstack-ironic | 09:50 | |
*** PrinzElvis has joined #openstack-ironic | 09:56 | |
dtantsur|mtg | mbuil: it may be in-memory conversion of the image? | 09:58 |
dtantsur|mtg | I have to apologize, the internet connection is quite bad where I am now | 10:09 |
dtantsur|mtg | rpittau: I've updated the whiteboard, seems the job is really broken | 10:17 |
rpittau | dtantsur|mtg, yeah, I increased the ram but no luck | 10:17 |
rpittau | dtantsur|mtg, something on network | 10:17 |
mbuil | dtantsur|mtg: the in-memory conversion is happening after the image is transferred to the nodes, right? | 10:17 |
dtantsur|mtg | mbuil: yes | 10:18 |
rpittau | dtantsur|mtg, http://paste.openstack.org/show/747329/ | 10:19 |
dtantsur|mtg | rpittau: ansible failure in http://logs.openstack.org/50/639050/10/check/ironic-standalone/827f381/controller/logs/screen-ir-cond.txt.gz?level=WARNING | 10:20 |
rpittau | I see red everywhere :/ | 10:21 |
dtantsur|mtg | your paste is even more weird.. | 10:21 |
dtantsur|mtg | rpittau: I wonder if we have to make the job non-voting for now.. | 10:22 |
rpittau | yeah, but I remember a json change yesterday somewhere ? maybe I'm paranoid | 10:22 |
mbuil | dtantsur|mtg: what I experienced is ironic-conductor reporting "prepare_image: result "None", error "None"" but in reality there was no HTTP traffic going-on. Can you point me to the code which prepares the image please? Maybe I can add some extra logs and check | 10:22 |
dtantsur|mtg | mbuil: it's in ironic-python-agent, look for 'def prepare_image' (sorry, on REALLY bad internet right now) | 10:23 |
mbuil | dtantsur|mtg: no problem, thanks! | 10:30 |
*** e0ne has joined #openstack-ironic | 10:35 | |
dtantsur|mtg | rpittau: I can assume the ansible deploy cannot ssh into the ramdisk for some reason | 10:36 |
dtantsur|mtg | rpittau: the most likely cause is the move to tinycore 8.x? | 10:36 |
rpittau | dtantsur|mtg, I see a connection established in the ansible logs | 10:38 |
dtantsur|mtg | and then broken pipe happens, apparently | 10:39 |
rpittau | dtantsur|mtg, this is more readable http://paste.openstack.org/show/747334/ | 10:39 |
*** moshele has quit IRC | 10:40 | |
dtantsur|mtg | rpittau: one guess: do you know of ansible can detect and use python 3 automatically? | 10:41 |
* dtantsur|mtg checks https://docs.ansible.com/ansible/latest/reference_appendices/python_3_support.html | 10:41 | |
dtantsur|mtg | rpittau: /usr/bin/python: not found | 10:42 |
dtantsur|mtg | well, indeed :) | 10:42 |
rpittau | oh well | 10:43 |
dtantsur|mtg | rpittau: we need to make the ansible deploy python 3 aware | 10:43 |
dtantsur|mtg | rpittau: would you be able to look into it? | 10:44 |
rpittau | dtantsur|mtg, I'll start looking into it yeah | 10:44 |
dtantsur|mtg | rpittau: thanks! the easiest way, probably, is to make ansible_python_interpreter a configuration+driver_info option | 10:45 |
rpittau | dtantsur|mtg, ack, I'll see what we have available but from the support page looks like the recommended choice | 10:46 |
*** dougsz has joined #openstack-ironic | 10:52 | |
*** iurygregory is now known as iurygregory|lunc | 10:54 | |
*** FlorianFa has quit IRC | 11:01 | |
*** FlorianFa has joined #openstack-ironic | 11:02 | |
*** pcaruana has quit IRC | 11:04 | |
*** moshele has joined #openstack-ironic | 11:07 | |
*** ianychoi_ is now known as ianychoi | 11:20 | |
*** pcaruana has joined #openstack-ironic | 11:32 | |
*** dsneddon has quit IRC | 11:35 | |
*** etingof has joined #openstack-ironic | 11:37 | |
*** iurygregory|lunc is now known as iurygregory | 12:02 | |
*** dsneddon has joined #openstack-ironic | 12:06 | |
*** S4ren has quit IRC | 12:07 | |
*** hritvi has joined #openstack-ironic | 12:09 | |
*** dsneddon has quit IRC | 12:11 | |
*** bfournie has quit IRC | 12:23 | |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector master: [WIP] Pass reset_interfaces when updating a driver from the rules https://review.openstack.org/641340 | 12:23 |
*** andrein has quit IRC | 12:37 | |
*** andrein has joined #openstack-ironic | 12:37 | |
*** dsneddon has joined #openstack-ironic | 12:40 | |
*** dsneddon has quit IRC | 12:44 | |
*** lekhikadugtal has joined #openstack-ironic | 12:45 | |
*** lekhikadugtal has quit IRC | 12:50 | |
*** arne_wiebalck_ has joined #openstack-ironic | 12:50 | |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector master: Pass reset_interfaces when updating a driver from the rules https://review.openstack.org/641340 | 12:55 |
*** Skyraven has joined #openstack-ironic | 12:56 | |
Skyraven | hi all:) I'm trying to configure for the first time Baremetal Ironic in the overcloud, read some of the Redhat docs, read also some tutorials I found with google but I'm still having a massive confusion. Can I have a custom nic config for tenant network? I mean to say: I clean it and provision it on the port defined in the json I import..let's say interface: eno1 with mac address 1 | 12:57 |
Skyraven | can I configure that when I do openstack server create...I get the "tenant" network on another interface? or a lag ? | 12:57 |
Skyraven | server has eno1, ens3f0, ens3f1...I'm thinking if it's possible to have cleaning + prov on eno1 (this I'm sure it is and know how to do it) but then when booting it up finally to have ens3f0+f1 = LAG = access or trunk in VLAN xzy | 12:58 |
Skyraven | so far nothing seems to point in the direction that this would be possible and is more hinting that tenant network would have to also be on eno1 | 12:58 |
Skyraven | or I need to fool it with a lot of --nic parameters == fake/null networks until I reach for example the number of nic corresponding to ens3f0 ...and then that woudl get the tenant net | 12:59 |
Skyraven | am I too confused or thinking wrong? I'm sure I did not express myself very straighforward but reached a dead-end:( | 12:59 |
dtantsur|mtg | Skyraven: I *think* nova doesn't allow per-NIC configuration. But you can set pxe_enabled=False on ens3f0 to avoid it used for provisioning, then request two NICs from nova: one for eno1, the other for ens3f0 | 12:59 |
dtantsur|mtg | just a raw though, other folks may have better ideas (and I'm not HORRIBLE internet today) | 12:59 |
dtantsur|mtg | but in the end, Nova is not aware of physical NICs, so you may need some kind of hackary | 13:00 |
Skyraven | dtantsur|mtg: think I was getting the idea right then :) I suspect I should introspect the node first (in Undercloud/Director), then delete it from there (but I see on which nic number / index ens3f0 falls), then when I boot it up | 13:00 |
*** andy_ has joined #openstack-ironic | 13:00 | |
Skyraven | I do openstack server create --nic net-id= ... -- nic net-id= ... --nic net-id = until I fall on the index of ens3f0 | 13:00 |
Skyraven | am I wrong ? | 13:00 |
dtantsur|mtg | Skyraven: are all NICs connected to one or different physical networks (from Neutron's point of view)? | 13:01 |
*** andrein has quit IRC | 13:02 | |
* dtantsur|mtg summons hjensas | 13:03 | |
*** bfournie has joined #openstack-ironic | 13:03 | |
hjensas | dtantsur|mtg: Hi, reading. | 13:03 |
dtantsur|mtg | if the physical networks are different, ports in ironic can be marked with a relevant physical_network | 13:04 |
* TheJulia_sick starts to wake up | 13:05 | |
*** TheJulia_sick is now known as TheJulia | 13:05 | |
*** dsneddon has joined #openstack-ironic | 13:05 | |
dtantsur|mtg | morning TheJulia | 13:06 |
dtantsur|mtg | how are you today? | 13:06 |
hjensas | morning TheJulia hope you are feeling better. | 13:06 |
TheJulia | feeling human-ish. Summer has it now though. :( | 13:06 |
jroll | morning dtantsur|mtg, TheJulia, hjensas, anyone else lurking :) | 13:07 |
dtantsur|mtg | sigh | 13:07 |
dtantsur|mtg | TheJulia: send her our best wishes | 13:07 |
dtantsur|mtg | morning jroll | 13:07 |
TheJulia | will do | 13:07 |
jroll | glad you're feeling better julia, hope she recovers as quickly as you did! | 13:07 |
Skyraven | dtantsur|mtg: think I have a massive confusion in my head from all the recent reading and overlapping things :(. So far I only saw NeutronBridgeMappings that I used for creating the custom baremetal network on which pxe boot/cleaning will happen. Not sure how/where to define for something else. For normal hypervisors (not baremetal) that info at least is coupled with the nic-config per Compute Node. Here I'm massively con | 13:07 |
TheJulia | At some point later today I'm taking her to the doctor and hopefully getting her antivirals | 13:07 |
dtantsur|mtg | TheJulia, jroll, https://etherpad.openstack.org/p/BER19-OPS-BAREMETAL has notes | 13:07 |
TheJulia | dtantsur|mtg: good notes, bad notes? notes that will make me cheer in joy or cry in sorrow? | 13:08 |
jroll | dtantsur|mtg: cool, is the session over now? | 13:08 |
dtantsur|mtg | Skyraven: hmm, okay, so I guess you use flat networking.. | 13:08 |
Skyraven | umm not really | 13:09 |
jroll | I love that we constantly get feedback that our docs are great, and at the same time that we need better docs :) | 13:09 |
Skyraven | I went the Redhat way with composable network, no l3 communication between undercloud pxe net and overcloud bm net. | 13:09 |
dtantsur|mtg | TheJulia, jroll, over. some good stuff, some usual complaints about docs. nothing breathtaking either way | 13:09 |
Skyraven | on the Controller it is a vlan inside br-ex which is Neutron Bridge datacentre | 13:09 |
TheJulia | jroll: everyone wants documentation tailored to their deployment scenario | 13:09 |
jroll | yep :) | 13:09 |
TheJulia | with diagrams and wget -o https://script.url | bash commands.... | 13:10 |
Skyraven | then I saw I have to just make in the overcloud the cleaning network (Default name provisioning) to sit on the same vlan + different allocation range | 13:10 |
Skyraven | I have Ironic listening on it also on the Controller and here my understanding fully stopped | 13:10 |
* TheJulia might have the sarcasm knob turned to 13 this morning | 13:10 | |
jroll | heh | 13:11 |
*** dsneddon has quit IRC | 13:11 | |
jroll | TheJulia: https://www.staples-3p.com/s7/is/image/Staples/s0105150_sc7?wid=512&hei=512 | 13:11 |
TheJulia | I've been thinking about "12.1.0... We re-calibrated 11!" | 13:12 |
iurygregory | morning TheJulia and jroll o/ | 13:12 |
jroll | hey iurygregory \o | 13:12 |
dtantsur|mtg | Skyraven: but it's still without full VLAN tenant network integration, right? no networking-ansible, etc? | 13:14 |
dtantsur|mtg | Skyraven: I think the way you should plan it then is to leave one NIC for ironic (en0?), and enroll only a port for it. then configure other NICs any way. | 13:15 |
dtantsur|mtg | if you want fully managed networks, it's a bit more difficult | 13:15 |
Skyraven | dtantsur|mtg: that's where I reached a stumble point :) no networking-ansible for now | 13:16 |
hjensas | Skyraven: If you do openstack baremetal node show node-0 -c network_interface - What interface did you set? | 13:16 |
* dtantsur|mtg feels like writing an "ironic networking quickstart" today | 13:16 | |
hjensas | Skyraven: replace node-0 with the name or uuid of your baremetal node ... | 13:16 |
TheJulia | dtantsur|mtg: As a matrix right? | 13:19 |
TheJulia | or Choose your own adventure? | 13:19 |
dtantsur|mtg | TheJulia: the guide? even simpler, something starting with "Here is how networking can happen in ironic..." | 13:19 |
TheJulia | I am all for the can | 13:20 |
TheJulia | I have resisted authoritative "you will do" documentation desires in the past. | 13:20 |
Skyraven | I am just redoing the setup ... tried 1h ago with a flat net+l3 routed to provisioning net..but now have this other one with isolated /composable network proceeding with little steps..I will have it in 10 min ready (the output of that command and new setup). Confusion is on the flow/architecture level...think I'm missing a few points...as to how I bind a different tenant network to it on deployment (especially since tenan | 13:21 |
Skyraven | which needs mapping somewhere..usually for hypervisors it was easy...physical segment --> ovs-bridge--> ovs bridge in nic-configs of a node..but in baremetal case..I need to clear up my thoughts..I'm mega confused) | 13:22 |
Skyraven | proceeding now with the composable network and I get back :) | 13:22 |
*** Qianbiao has quit IRC | 13:22 | |
*** amoralej is now known as amoralej|lunch | 13:26 | |
hjensas | Skyraven: I have to go pick up kids at school, but I would recommend reading this docstring: https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/network/common.py#L144-L169 | 13:34 |
hjensas | Skyraven: If you want provisioning,cleaning,rescue on eno1 - set the corresponding ironic baremetal port to pxe_enabled. | 13:34 |
hjensas | Skyraven: then if you want to use a LAG on ens3f0, ens3f1 you want to create a portgroup in ironic. | 13:35 |
Skyraven | eno1 set already:) did an experiment before. Understood for second part. Thanks. I get unto reading. | 13:35 |
hjensas | Skyraven: you probably also want to set the 'physical' network property on ironic ports to match the physical network property of the network in neutron. | 13:35 |
*** dsneddon has joined #openstack-ironic | 13:36 | |
Skyraven | ah so that's the connecting element I was missing when I was thinking about the other case with Nova Computes | 13:36 |
Skyraven | thank you | 13:36 |
*** mjturek has joined #openstack-ironic | 13:37 | |
hjensas | Skyraven: ok, good luck. I will be back in a couple of hours. | 13:38 |
*** hjensas is now known as hjensas|afk | 13:40 | |
*** mjturek has quit IRC | 13:40 | |
*** arne_wiebalck has quit IRC | 13:41 | |
*** arne_wiebalck_ is now known as arne_wiebalck | 13:41 | |
*** dsneddon has quit IRC | 13:42 | |
Skyraven | Skyraven: am still at work but this is making me curious. I have to do some tshoot for other projects but will get back to this and if not, later in the evening:) I want to see it work:) I go on vacation Thursday evening and this bugs me..the usual techie feeling when curious | 13:42 |
*** arne_wiebalck_ has joined #openstack-ironic | 13:42 | |
*** hjensas|afk has quit IRC | 13:44 | |
*** dsneddon has joined #openstack-ironic | 13:46 | |
openstackgerrit | Nikolay Fedotov proposed openstack/ironic-python-agent master: Discover IPv6 BMC address https://review.openstack.org/640795 | 13:47 |
openstackgerrit | Ilya Etingof proposed openstack/ironic master: Set boot_mode in node properties during Redfish introspection https://review.openstack.org/641357 | 13:51 |
*** pcaruana has quit IRC | 13:51 | |
*** hritvi has quit IRC | 13:52 | |
*** dsneddon has quit IRC | 13:54 | |
*** sdake has quit IRC | 13:54 | |
*** sdake has joined #openstack-ironic | 14:00 | |
*** pcaruana has joined #openstack-ironic | 14:01 | |
arne_wiebalck_ | etingof: To summarise what we found yesterday concerning the parallel power sync I’d say: a) increasing the workers reduces the sync runtime linearly (in our specific setup, this scaling works for 1…~4 workers, then we hit a plateau), so the patch works! b) reducing the ipmitool command duration has a postive impact, but does not bring scaling back when using<4 workers c) conductor CPU utilisation goes to 80% during the | 14:07 |
arne_wiebalck_ | sync, so the best guess so far for the plateau is that we hit what a single process can achieve. | 14:07 |
*** arne_wiebalck has left #openstack-ironic | 14:07 | |
*** arne_wiebalck_ has left #openstack-ironic | 14:08 | |
*** arne_wiebalck_ has joined #openstack-ironic | 14:08 | |
*** FlorianFa has quit IRC | 14:09 | |
*** rloo has joined #openstack-ironic | 14:12 | |
openstackgerrit | Riccardo Pittau proposed openstack/ironic master: [WIP] Add option to set python interpreter for ansible https://review.openstack.org/641376 | 14:14 |
*** jaypipes has quit IRC | 14:14 | |
*** jaypipes has joined #openstack-ironic | 14:15 | |
*** baha has joined #openstack-ironic | 14:16 | |
*** gkadam_ has joined #openstack-ironic | 14:16 | |
*** sdake has quit IRC | 14:17 | |
*** gkadam has quit IRC | 14:19 | |
etingof | arne_wiebalck_, yeah, interesting... | 14:20 |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: [WIP] Reference architecture: expand on networking https://review.openstack.org/641383 | 14:20 |
dtantsur|mtg | TheJulia: just some structure and a few thoughts ^^^ | 14:20 |
TheJulia | dtantsur|mtg: awesome | 14:21 |
*** sdake has joined #openstack-ironic | 14:22 | |
*** dsneddon has joined #openstack-ironic | 14:24 | |
*** sdake has quit IRC | 14:24 | |
*** jcoufal has joined #openstack-ironic | 14:27 | |
*** amoralej|lunch is now known as amoralej | 14:29 | |
*** dsneddon has quit IRC | 14:29 | |
*** mjturek has joined #openstack-ironic | 14:30 | |
openstackgerrit | Lars Kellogg-Stedman proposed openstack/ironic master: [WIP] honor ipmi_port in serial console drivers https://review.openstack.org/640930 | 14:33 |
openstackgerrit | Lars Kellogg-Stedman proposed openstack/ironic master: honor ipmi_port in serial console drivers https://review.openstack.org/640930 | 14:34 |
*** arne_wiebalck has joined #openstack-ironic | 14:34 | |
iurygregory | TheJulia, quick question do we really need generate_testr_results function ? https://github.com/openstack/python-ironicclient/blob/master/ironicclient/tests/functional/hooks/post_test_hook.sh#L17 | 14:34 |
*** arne_wiebalck has quit IRC | 14:35 | |
*** arne_wiebalck_ is now known as arne_wiebalck | 14:35 | |
*** arne_wiebalck has quit IRC | 14:35 | |
iurygregory | i was looking how mistral people did to move their functional job for python-mistralclient, and zuulv3 we dont have some legacy variables and we will need some workaround | 14:37 |
*** arne_wiebalck_ has joined #openstack-ironic | 14:37 | |
*** moshele has quit IRC | 14:37 | |
*** sdake has joined #openstack-ironic | 14:38 | |
TheJulia | iurygregory: I don't know. | 14:38 |
TheJulia | i suspect it is handy to have because of the resulting output | 14:38 |
iurygregory | testrepository.subunit we can open in the browser XD | 14:39 |
TheJulia | is it actually rendered as html? | 14:40 |
iurygregory | and we have the functional results in the job log | 14:40 |
iurygregory | if you use subunit2html you can get an html | 14:40 |
* iurygregory going to grab coffee brb | 14:41 | |
arne_wiebalck_ | etingof: For now, I’d go with dtantsur|mtg’s suspicion that we more or less max out the cpu … also, I’m doing these tests on our prod system (which is required since we need some scale, but it remains a prod system :) ). So, unless you have other suggestions, I suggest we record our findings somewhere for now and come back to this when we see a need. What do you thikn? | 14:41 |
*** dsneddon has joined #openstack-ironic | 14:42 | |
etingof | arne_wiebalck_, let me think it over for some more - may be we could devise more insightful experiments on your production system... | 14:45 |
arne_wiebalck_ | etingof: sure, np. For now I have the code on one of the controllers and can leave ith there for the time being. | 14:46 |
*** dsneddon has quit IRC | 14:47 | |
arne_wiebalck_ | etingof: the patch doesn’t seem to reference a story … is there one? I was looking for a place to document what we did and found … | 14:50 |
openstackgerrit | Riccardo Pittau proposed openstack/ironic master: [WIP] Add option to set python interpreter for ansible https://review.openstack.org/641376 | 14:57 |
etingof | arne_wiebalck_, ah, I think we have it scattered over ptg discussion -- https://etherpad.openstack.org/p/ironic-stein-ptg | 14:59 |
*** gkadam__ has joined #openstack-ironic | 15:00 | |
etingof | arne_wiebalck_, I can open up a story or you can if you in the writing mood | 15:00 |
*** sdake has quit IRC | 15:00 | |
*** sthussey has joined #openstack-ironic | 15:01 | |
TheJulia | ironic-cores: you may want to check openstack-discuss when you have a chance | 15:03 |
*** gkadam_ has quit IRC | 15:03 | |
* iurygregory back o/ | 15:05 | |
*** sdake has joined #openstack-ironic | 15:05 | |
iurygregory | TheJulia, well i think we dont get the html result, at least I didn't found in the logs folder http://logs.openstack.org/91/640491/5/check/ironicclient-dsvm-functional/0beb21d/logs/ | 15:06 |
*** moshele has joined #openstack-ironic | 15:07 | |
*** dsneddon has joined #openstack-ironic | 15:13 | |
openstackgerrit | Ilya Etingof proposed openstack/ironic master: Set boot_mode in node properties during Redfish introspection https://review.openstack.org/641357 | 15:17 |
*** dsneddon has quit IRC | 15:18 | |
TheJulia | dtantsur|mtg: do you know if any of the topics on the topics on the brainstorming etherpad were submitted as forum topics? | 15:21 |
dtantsur|mtg | TheJulia: not by me (and I won't be able until Friday) | 15:22 |
*** efried has joined #openstack-ironic | 15:22 | |
efried | o/ | 15:22 |
TheJulia | dtantsur|mtg: entry closes ?tomorrow? I believe | 15:22 |
dtantsur|mtg | oh | 15:23 |
jroll | uh oh, efried is gonna make us do stuff | 15:23 |
TheJulia | ironic-cores, please skim over https://etherpad.openstack.org/p/DEN-train-ironic-brainstorming and add ++ to anything that seems topic worthy from your poitn of view | 15:23 |
dtantsur|mtg | hi efried | 15:23 |
TheJulia | rutro | 15:23 |
TheJulia | efried will surely ask politely | 15:23 |
efried | haha, no, I just wanted to ask about https://review.openstack.org/#/c/640879/ | 15:23 |
patchbot | patch 640879 - nova - pass endpoint interface to Ironic client - 3 patch sets | 15:23 |
TheJulia | :) | 15:23 |
TheJulia | And possibly enjoy a beer with us | 15:23 |
efried | can wait until your discussion is done | 15:24 |
efried | yeah, that for sure. | 15:24 |
dtantsur|mtg | TheJulia: the last time I thought about it, I thought about 1. A feedback session, 2. A standalone roadmap session, 3. An airship session. Now ops folks made me think about something about docs.. | 15:24 |
jroll | efried: go ahead and ask, we're pretty good at concurrency here | 15:24 |
TheJulia | We did a sesison with airship in denver, we never got clarity into their setup reqs, so I feel like working sessions would be better for airship :\ | 15:24 |
TheJulia | efried: we simply haven't managed to get a cloning machine online yet, so there may be some parallelism limits | 15:25 |
efried | Okay, so reading that patch, it looks like it's something you'll want to backport. Is that true? | 15:25 |
arne_wiebalck_ | etingof: ok, will do | 15:25 |
dtantsur|mtg | TheJulia: I don't insist on an airship session :) | 15:25 |
jroll | efried: I would agree it feels backport-worthy | 15:26 |
efried | That's going to be a bit twiddly, so I wanted to make sure the plan is right. | 15:26 |
etingof | arne_wiebalck_, ping me once the story is up, I could possibly add some scariness | 15:26 |
TheJulia | efried: I woulc concur | 15:26 |
TheJulia | would concur | 15:26 |
efried | IIUC, that patch is fully dependent on the ironicclient side | 15:26 |
efried | i.e. won't work without it | 15:26 |
efried | which means as it stands it can't be backported. | 15:26 |
TheJulia | it uses the keystone auth adapter | 15:27 |
efried | So we need the PS2 version reinstated, because it'll "work" (not blow up, though still be subject to the reported bug) without a new version of ironicclient. | 15:27 |
dtantsur|mtg | it's another sign for us to just migrate to openstacksdk for x-project interactions... | 15:27 |
TheJulia | I think.... | 15:27 |
efried | dtantsur|mtg: ++ - more on that later. | 15:27 |
jroll | is there any reason we couldn't backport the ironicclient patch? | 15:27 |
dtantsur|mtg | I want to try it with ironic-inspector soon(ish) | 15:27 |
dtantsur|mtg | jroll: it's kind of a feature IIUC | 15:28 |
efried | jroll: the ironicclient patch can and should be backported, yes. | 15:28 |
* TheJulia doesn't remember a client patch | 15:28 | |
dtantsur|mtg | and anyway we shouldn't update nova's dependency on ironicclient on stable branches | 15:28 |
efried | sorry, the client patch is here: https://review.openstack.org/#/c/640491 | 15:28 |
patchbot | patch 640491 - python-ironicclient - pass endpoint interface to http client - 5 patch sets | 15:28 |
efried | dtantsur|mtg: bingo. | 15:28 |
TheJulia | ugh there is a client patch | 15:28 |
TheJulia | Yeah, thats a no-go | 15:28 |
efried | We can backport the ironicclient fix and bump upper-constraints in stable, but we can't bump nova's lower-constraints in stable. | 15:28 |
TheJulia | This feels very deja-vu to me... | 15:29 |
jroll | blurgh, stable branches are too hard | 15:29 |
efried | So in order to pick up the bug fix, you would need to pick up both the new ironicclient release with the backported fix; and the nova backport. | 15:29 |
*** moshele has quit IRC | 15:29 | |
efried | but people who don't care or are not affected can still pick up the nova side without the ironicclient fix, without it blowing up. | 15:30 |
*** moshele has joined #openstack-ironic | 15:30 | |
efried | and then we can put the PS3 version of this nova patch into master, along with a l-c bump to require the new ironicclient. | 15:30 |
TheJulia | yup | 15:30 |
jroll | yeah, backporting PS2 is fine with me | 15:30 |
efried | I tried to explain that in PS2, but I guess I wasn't clear. So I wanted to make sure I was at least right before I made the developer respin and split and... | 15:31 |
efried | so, on a related note, I wanted to show y'all something I started working on as a result of seeing yet again how this adapter => ironicclient kwargs => different adapter horribleness can cause us headaches. | 15:32 |
efried | It's at least a step toward openstacksdk-ness, but in terms I understand. | 15:33 |
TheJulia | efried: I will buy you a beer if it is pure restification of the api interaction | 15:33 |
efried | The short version is: make ironicclient accept and use the ksa adapter from the caller. | 15:33 |
TheJulia | efried: that works too | 15:33 |
* TheJulia adds beer for efried to database | 15:33 | |
*** hjensas has joined #openstack-ironic | 15:33 | |
efried | The ironicclient side: https://review.openstack.org/#/c/641169/ | 15:34 |
patchbot | patch 641169 - python-ironicclient - WIP: AdapterClient: accept an already-constructed ... - 1 patch set | 15:34 |
efried | and then nova is like a one-liner: https://review.openstack.org/#/c/641170/1/nova/virt/ironic/client_wrapper.py | 15:34 |
patchbot | patch 641170 - nova - WIP: use ironicclient AdapterClient - 1 patch set | 15:34 |
efried | I ran out of time working this up last night, and apparently the CI job ironic-tempest-ipa-wholedisk-bios-agent_ipmitool-tinyipa isn't set up to pull ironicclient from source, so I actually have no idea whether it works yet. | 15:35 |
efried | but that's the idea. | 15:35 |
openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/python-ironicclient master: Move to zuulv3 https://review.openstack.org/633010 | 15:35 |
TheJulia | efried: I'm +1 for concept level | 15:36 |
*** moshele has quit IRC | 15:39 | |
*** ijpascual has quit IRC | 15:40 | |
*** ijpascual has joined #openstack-ironic | 15:41 | |
efried | okay, I'll see how far I can get. | 15:42 |
efried | Thanks y'all. | 15:42 |
rloo | TheJulia, jroll: wrt the client PR to fix 'interface'. Want to make sure i understand the stuff about stable branches above.: https://review.openstack.org/#/c/640491 | 15:44 |
patchbot | patch 640491 - python-ironicclient - pass endpoint interface to http client - 5 patch sets | 15:44 |
rloo | TheJulia, jroll: can we backport that? | 15:45 |
TheJulia | if as a kwarg, yes | 15:45 |
TheJulia | i.e. non breaking | 15:45 |
TheJulia | since it is a bug fix as well since we apparently had the basic capability back in queens | 15:46 |
rloo | TheJulia: ok, it is to fix a bug. (the 'interface' part, the 'valid_interfaces' is questionable) | 15:46 |
rloo | TheJulia: i think it broken in queens. i think it works in pike. according to the reporter. | 15:47 |
rloo | or maybe it broke in pike, forgot now. | 15:47 |
jroll | rloo: yeah, it feels backportable to me as it's a regression | 15:47 |
*** dsneddon has joined #openstack-ironic | 15:47 | |
rloo | reporter/story said it used to work in pike. | 15:48 |
jroll | man TheJulia's email points out my abysmal review stats | 15:48 |
* jroll hides in shame | 15:48 | |
rloo | the 'valid_interfaces' part is new, we never supported that. | 15:48 |
TheJulia | jroll: Hey now, I'm not yelling shame ;) | 15:48 |
jroll | it was coincidental :) | 15:49 |
* TheJulia is trying to be lower-key and understanding | 15:49 | |
* rloo is not that much better than jroll. but hopefully should be able to start participating more. | 15:50 | |
dtantsur|mtg | folks, I'll stop pretending IRC works for me. See you when I get back to normal internet.. | 15:50 |
*** dtantsur|mtg is now known as dtantsur|afk | 15:50 | |
rloo | bye dtantsur|mtg, enjoy your time away from us! | 15:50 |
dtantsur|afk | I'd prefer time with you :) | 15:50 |
rloo | dtantsur|afk made my day :D | 15:51 |
rpittau | dtantsur|afk, I'm surprised you resisted for so long :D | 15:51 |
TheJulia | :) | 15:51 |
rpittau | wondering if we should make ironic-standalone non voting until the issue with ansible interface is fixed | 15:53 |
*** pcaruana has quit IRC | 15:53 | |
TheJulia | rpittau: more context please | 15:53 |
rpittau | yup, right, ironic-standalone is failing a test with ansible interface because the new IPA image has only python3 and by default ansible interpreter is python | 15:54 |
TheJulia | bahahahhahahaha | 15:54 |
rpittau | that's also the reason of this: https://review.openstack.org/641376 | 15:54 |
patchbot | patch 641376 - ironic - [WIP] Add option to set python interpreter for ans... - 2 patch sets | 15:54 |
rpittau | yeah..... | 15:54 |
TheJulia | rpittau: yeah, please put up a patch to mark it non-voting for now so we can continue to merge other work this week | 15:55 |
rpittau | TheJulia, ack | 15:55 |
openstackgerrit | Riccardo Pittau proposed openstack/ironic master: [WIP] Add option to set python interpreter for ansible https://review.openstack.org/641376 | 15:56 |
* etingof notes arne_wiebalck_'s story on power sync | 15:57 | |
arne_wiebalck_ | etingof: I did not add the kill patch (has it’s own story already and the point of this one is now mostly to record experiences with the parallelization), sounds ok? | 15:58 |
efried | [ update: I summarized and downvoted on the patch. If you get a chance, please make sure I represented things accurately. https://review.openstack.org/#/c/640879/ ] | 16:00 |
patchbot | patch 640879 - nova - pass endpoint interface to Ironic client - 3 patch sets | 16:00 |
openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Temporary marking ironic-standalone non-voting https://review.openstack.org/641428 | 16:02 |
*** pcaruana has joined #openstack-ironic | 16:06 | |
*** sdake has quit IRC | 16:09 | |
*** Skyraven has quit IRC | 16:13 | |
*** sdake has joined #openstack-ironic | 16:15 | |
*** tssurya has quit IRC | 16:18 | |
rloo | TheJulia: btw, last night I looked at the client PRs that are open; the only one that we really need to get in by tomorrow is https://review.openstack.org/#/c/640491/ | 16:22 |
patchbot | patch 640491 - python-ironicclient - pass endpoint interface to http client - 5 patch sets | 16:22 |
rloo | TheJulia: do you need to write up some release notey thing for it? | 16:22 |
rloo | TheJulia: it == client release. | 16:22 |
rloo | TheJulia: and the only other thing we should get in but I think we can merge after the release, is the zuul v3 work. | 16:23 |
TheJulia | I'm wrappe dup in another meeting at the moment | 16:24 |
rloo | TheJulia: no worries, that is just FYI stuff :) | 16:24 |
arne_wiebalck_ | etingof: I added our test results as a comment now. | 16:25 |
etingof | arne_wiebalck_, thank you! | 16:26 |
*** arne_wiebalck_ has quit IRC | 16:32 | |
*** arne_wiebalck_ has joined #openstack-ironic | 16:32 | |
*** arne_wiebalck has joined #openstack-ironic | 16:33 | |
openstackgerrit | Riccardo Pittau proposed openstack/ironic master: [WIP] Add option to set python interpreter for ansible https://review.openstack.org/641376 | 16:34 |
*** bfournie has quit IRC | 16:39 | |
efried | rloo, TheJulia: Who's guang-yee? What time zone? If we're running up against client lib freeze, I would be willing to do some reshuffling of that patch if you need. | 16:40 |
openstackgerrit | Riccardo Pittau proposed openstack/ironic master: [WIP] Add option to set python interpreter for ansible https://review.openstack.org/641376 | 16:40 |
rloo | efried: no idea. | 16:40 |
rloo | efried: the client freeze is tomorrow. | 16:40 |
rloo | efried: the nova one doesn't need to go in by tomorrow | 16:40 |
efried | right | 16:41 |
rloo | efried: i don't really think the nova one needs a depends-on (i forgot if it has that or not). if we go with PS2. | 16:41 |
efried | I was talking about the ironicclient patch https://review.openstack.org/#/c/640491/ | 16:41 |
patchbot | patch 640491 - python-ironicclient - pass endpoint interface to http client - 5 patch sets | 16:41 |
efried | rloo: Correct, see the discussion I dropped there | 16:41 |
rloo | efried: we need people to review that client patch. | 16:41 |
rloo | efried: maybe you can answer my question about 'valid_interfaces' ? | 16:42 |
efried | I did | 16:42 |
rloo | efried: maybe i should look first before asking :D | 16:42 |
*** gkadam__ has quit IRC | 16:42 | |
rloo | efried: ok, yes, i think the right thing to do is to separate that client PR into two; one that fixes 'interface' and one that adds support for 'valid_interfaces'. | 16:43 |
efried | If we split ironicclient side into two patches (1: backportable, 'interface' only; 2: master, add 'valid_interfaces') then the nova side patches as split would correspond (backportable PS2 maps to 1, no depends-on; master PS3 maps to 2, still no depends-on, but a l-c bump). | 16:43 |
rloo | efried: but i'm fine bundling them both if others are fine (me lazy) | 16:43 |
openstackgerrit | Lars Kellogg-Stedman proposed openstack/ironic master: honor ipmi_port in serial console drivers https://review.openstack.org/640930 | 16:43 |
rloo | efried: ah, yes, that's the better solution. | 16:43 |
efried | PTL make the call? | 16:43 |
rloo | efried: i think she's in a meeting. i did ask and she seemed fine with backporting it as-is. | 16:44 |
efried | with a 'features' reno add? | 16:44 |
rloo | efried: i'm also good if you want to split up that patch into two. which is really the right thing to do. | 16:45 |
rloo | efried: you nova folks are so hard-nosed/do-the-right-thing! I love it! | 16:45 |
efried | Okay, let me see what I can get done in 15 minutes; else it has to wait until this afternoon, which is short runway for reviews... | 16:45 |
rloo | thx efried. i could do it, but i think it'd be better for me to use my +2 powers. | 16:45 |
efried | agree | 16:46 |
*** sdake has quit IRC | 16:48 | |
*** sdake has joined #openstack-ironic | 16:49 | |
openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/python-ironicclient master: Move to zuulv3 https://review.openstack.org/633010 | 16:53 |
*** pcaruana has quit IRC | 16:55 | |
*** sdake has quit IRC | 16:56 | |
*** arne_wiebalck_ has quit IRC | 16:56 | |
openstackgerrit | Eric Fried proposed openstack/python-ironicclient master: pass endpoint interface to http client https://review.openstack.org/640491 | 16:57 |
efried | rloo: ^ | 16:57 |
efried | rloo: That's the backportable side. | 16:58 |
rloo | efried: thx, will look in a few min. | 16:58 |
*** gyee has joined #openstack-ironic | 17:06 | |
*** sburette has quit IRC | 17:08 | |
TheJulia | okay out of meeting and lacking coffee | 17:13 |
* TheJulia is out of coffee beans | 17:13 | |
gyee | efried, thanks for your help! I have a question to the latest changes to https://review.openstack.org/#/c/640491. This is the 'backportable' version right? And we'll spin up a new one to support 'valid_interfaces' in master? | 17:13 |
patchbot | patch 640491 - python-ironicclient - pass endpoint interface to http client - 6 patch sets | 17:13 |
efried | gyee: Correct, I'm just about to upload the master one now. | 17:14 |
*** arne_wiebalck has quit IRC | 17:14 | |
openstackgerrit | Eric Fried proposed openstack/python-ironicclient master: Accept 'valid_interfaces' in client setup https://review.openstack.org/641444 | 17:14 |
efried | gyee, rloo: there's the master-only one with the "feature" ^ | 17:15 |
TheJulia | rloo: I've +2'ed the client patch | 17:15 |
gyee | efried, we'll need to make sure nova also register the 'interface' option. Right now, it is only registering 'valid_interfaces' option | 17:15 |
efried | I have to run now. Please feel free to respin those if you need to. | 17:15 |
TheJulia | oh, guess I'm redoing | 17:15 |
efried | gyee: It should be registering both, with the 'interface' deprecated in favor of 'valid_interfaces'. | 17:15 |
*** arne_wiebalck has joined #openstack-ironic | 17:16 | |
efried | TheJulia: Your +2 holds (unless you want that reno fixed, which I agree with); the other is a separate patch. | 17:16 |
openstackgerrit | Eric Fried proposed openstack/python-ironicclient master: pass endpoint interface to http client https://review.openstack.org/640491 | 17:16 |
gyee | efried, for some reason 'interface' is not in nova.conf.ironic, maybe I am missing something | 17:16 |
gyee | I was trying to test that last night | 17:17 |
efried | rloo, TheJulia: I edited that reno. | 17:17 |
efried | so *now* it needs redo :) | 17:17 |
efried | tbc, 'interface' is not deprecated on Adapter __init__ - only in ksa config. | 17:17 |
gyee | oh that's confusing | 17:18 |
rpittau | good night o/ | 17:18 |
efried | gyee: I'll have to look into this later; I need to run now. The time pressure was on the ironicclient side, because client lib freeze is tomorrow (I think). | 17:18 |
efried | And I think the ironicclient pieces should be good to go now, or pretty close. | 17:18 |
gyee | awesome | 17:18 |
*** hjensas has quit IRC | 17:18 | |
rloo | thx efried & gyee. looking ... | 17:23 |
gyee | rloo, thanks for the help! Meanwhile, let me rework the nova patch to pass in 'interface'. | 17:29 |
rloo | gyee: thx you for fixing it! | 17:29 |
arne_wiebalck | good night, ironic | 17:30 |
openstackgerrit | Lars Kellogg-Stedman proposed openstack/ironic master: honor ipmi_port in serial console drivers https://review.openstack.org/640930 | 17:32 |
*** dsneddon has quit IRC | 17:33 | |
*** e0ne has quit IRC | 17:34 | |
*** baha has quit IRC | 17:42 | |
*** mmethot has quit IRC | 17:44 | |
*** mmethot has joined #openstack-ironic | 17:45 | |
openstackgerrit | Lars Kellogg-Stedman proposed openstack/ironic master: honor ipmi_port in serial console drivers https://review.openstack.org/640930 | 17:46 |
larsks | Now that ^^^ is actually working; sorry for the spam. | 17:47 |
TheJulia | larsks: no worries | 17:50 |
jroll | efried: +W on those ironicclient patches, thanks for that | 17:50 |
TheJulia | rloo: re python-ironicclient release, we've typically not done a prelude for it | 17:50 |
jroll | and thanks to gyee as well! | 17:50 |
TheJulia | rloo: really, only ironic has gotten a prelude with dramatic intent :) | 17:51 |
jroll | "intentionally dramatic prelude" is the name of my next garage band | 17:51 |
TheJulia | I expect it to be epic music | 17:52 |
jroll | naturally | 17:52 |
TheJulia | And what will the band's beer be called, since that is apparently now a thing | 17:52 |
* jroll feels inclined to listen to godspeed you! black emperor now | 17:52 | |
jroll | that is a thing now? | 17:52 |
jroll | weird | 17:53 |
TheJulia | yeah, it is | 17:53 |
*** dustinc has joined #openstack-ironic | 17:57 | |
gyee | when you keep mispronouncing 'beer' with 'bug', that when you know your brain is dead tired :-D | 17:59 |
*** baha has joined #openstack-ironic | 18:01 | |
*** dsneddon has joined #openstack-ironic | 18:03 | |
*** derekh has quit IRC | 18:03 | |
*** dustinc has quit IRC | 18:04 | |
*** dustinc has joined #openstack-ironic | 18:05 | |
*** jtomasek has quit IRC | 18:06 | |
*** dustinc has quit IRC | 18:07 | |
*** dustinc has joined #openstack-ironic | 18:07 | |
*** dustinc has quit IRC | 18:09 | |
*** dustinc has joined #openstack-ironic | 18:09 | |
rloo | TheJulia: ok, I couldn't remember wrt renos. So we're good for tomorrow I think. | 18:14 |
*** dustinc_ has joined #openstack-ironic | 18:26 | |
*** andrein has joined #openstack-ironic | 18:26 | |
*** amoralej is now known as amoralej|off | 18:26 | |
*** dustinc has quit IRC | 18:28 | |
larsks | So, thinking about https://review.openstack.org/#/c/637219/, I think there's still a problem: because we're not managing the child process ourselves, we don't know when it exits, so we're never able to send a 'sol deactivate'. | 18:34 |
patchbot | patch 637219 - ironic - ensure that socat serial proxy keeps running (MERGED) - 3 patch sets | 18:34 |
larsks | jroll: ^^^ | 18:36 |
jroll | blurgh | 18:37 |
jroll | larsks: any suggestions on dealing with that? | 18:37 |
larsks | jroll: not any good ones | 18:40 |
larsks | I mean, actually wait() on the socat process and handle things gracefully when it exits... | 18:40 |
*** dustinc_ is now known as dustinc | 18:41 | |
larsks | ...rather than having socat responsible for restarting itself. | 18:41 |
*** hjensas has joined #openstack-ironic | 18:41 | |
*** jaypipes has quit IRC | 18:44 | |
jroll | yeah. | 18:46 |
*** baha has quit IRC | 18:51 | |
*** mjturek has quit IRC | 18:51 | |
TheJulia | gyee: or a sign you want beer or a bug. Tough to tell :) | 19:02 |
TheJulia | I guess there are the two different use cases, where some operators want SOL online constantly and some only use it sparingly.... | 19:03 |
*** dougsz has quit IRC | 19:05 | |
*** baha has joined #openstack-ironic | 19:12 | |
*** mjturek has joined #openstack-ironic | 19:13 | |
openstackgerrit | Merged openstack/networking-baremetal master: Zuulv3 - Use ironic-base job https://review.openstack.org/635824 | 19:13 |
NobodyCam | Good Morning Ironic'ers | 19:29 |
*** bfournie has joined #openstack-ironic | 19:29 | |
*** baha has quit IRC | 19:34 | |
*** baha has joined #openstack-ironic | 19:34 | |
openstackgerrit | Merged openstack/ironic master: Temporary marking ironic-standalone non-voting https://review.openstack.org/641428 | 19:40 |
*** e0ne has joined #openstack-ironic | 20:03 | |
TheJulia | o/ | 20:04 |
*** mjturek has quit IRC | 20:06 | |
*** mjturek has joined #openstack-ironic | 20:07 | |
*** mjturek has quit IRC | 20:11 | |
*** mjturek has joined #openstack-ironic | 20:12 | |
*** e0ne has quit IRC | 20:21 | |
openstackgerrit | Merged openstack/python-ironicclient master: pass endpoint interface to http client https://review.openstack.org/640491 | 20:28 |
*** jlvillal has joined #openstack-ironic | 20:30 | |
*** betherly has joined #openstack-ironic | 20:31 | |
*** betherly has quit IRC | 20:41 | |
*** betherly has joined #openstack-ironic | 20:51 | |
*** fdegir has quit IRC | 20:53 | |
*** fdegir has joined #openstack-ironic | 20:54 | |
*** betherly has quit IRC | 20:55 | |
*** irclogbot_1 has joined #openstack-ironic | 21:08 | |
*** betherly has joined #openstack-ironic | 21:11 | |
*** e0ne has joined #openstack-ironic | 21:13 | |
*** betherly has quit IRC | 21:15 | |
*** mjturek has quit IRC | 21:16 | |
*** baha has quit IRC | 21:18 | |
*** whoami-rajat has quit IRC | 21:18 | |
*** irclogbot_1 has quit IRC | 21:28 | |
*** betherly has joined #openstack-ironic | 21:31 | |
*** betherly has quit IRC | 21:35 | |
*** mjturek has joined #openstack-ironic | 21:39 | |
*** baha has joined #openstack-ironic | 21:40 | |
*** dustinc has quit IRC | 21:46 | |
*** betherly has joined #openstack-ironic | 21:46 | |
*** dustinc has joined #openstack-ironic | 21:47 | |
*** dustinc has quit IRC | 21:48 | |
*** dustinc has joined #openstack-ironic | 21:48 | |
*** dustinc_ has joined #openstack-ironic | 21:50 | |
*** dustinc has quit IRC | 21:50 | |
*** dustinc_ has quit IRC | 21:50 | |
*** dustinc has joined #openstack-ironic | 21:50 | |
*** betherly has quit IRC | 21:51 | |
*** e0ne has quit IRC | 21:58 | |
*** jcoufal has quit IRC | 21:59 | |
*** e0ne has joined #openstack-ironic | 22:01 | |
*** e0ne has quit IRC | 22:01 | |
*** Skyraven has joined #openstack-ironic | 22:05 | |
Skyraven | hi all again, I'm still struggling a bit with baremetal service / ironic :) I managed to get all up to the cleaning phase to work but during cleaning..I get "failed to reset secure boot keys on the server" | 22:06 |
Skyraven | I have an HP DL360 G10, using uefi_boot (set as capabilities) | 22:06 |
Skyraven | ilo driver used | 22:06 |
Skyraven | is there something I might have forgotten or something in the bios that needs to be set?..googling for the exact message is not helping me out | 22:07 |
Skyraven | found a bugzilla though from 2015...rather far back | 22:08 |
*** MattMan_1 has quit IRC | 22:11 | |
*** MattMan_1 has joined #openstack-ironic | 22:11 | |
*** betherly has joined #openstack-ironic | 22:17 | |
*** betherly has quit IRC | 22:22 | |
Skyraven | I disabled some of the cleaning steps from ironic.conf, all good :) | 22:28 |
*** baha has quit IRC | 22:42 | |
*** mjturek has quit IRC | 22:51 | |
*** rcernin has joined #openstack-ironic | 22:52 | |
*** andrein has quit IRC | 23:02 | |
*** dustinc has quit IRC | 23:04 | |
*** dustinc has joined #openstack-ironic | 23:07 | |
*** dustinc has quit IRC | 23:12 | |
*** andrein has joined #openstack-ironic | 23:17 | |
*** andrein has quit IRC | 23:18 | |
*** dustinc has joined #openstack-ironic | 23:23 | |
*** dustinc has quit IRC | 23:25 | |
*** dustinc has joined #openstack-ironic | 23:26 | |
*** dustinc has joined #openstack-ironic | 23:26 | |
*** rh-jelabarre has quit IRC | 23:43 | |
TheJulia | Skyraven: I guess automatic reset for the ilo bios was set up in your config? | 23:49 |
*** w14161_1 has quit IRC | 23:52 | |
*** w14161_1 has joined #openstack-ironic | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!