| *** shakamunyi has joined #openstack-ironic | 00:02 | |
| *** naohirot has joined #openstack-ironic | 00:02 | |
| *** ryanpetrello has joined #openstack-ironic | 00:08 | |
| *** shakamunyi has quit IRC | 00:09 | |
| *** penick has joined #openstack-ironic | 00:09 | |
| *** shakamunyi has joined #openstack-ironic | 00:10 | |
| *** shakamunyi has quit IRC | 00:15 | |
| *** ryanpetrello has quit IRC | 00:23 | |
| *** Viswanath has joined #openstack-ironic | 00:25 | |
| *** achanda has quit IRC | 00:27 | |
| *** achanda has joined #openstack-ironic | 00:27 | |
| *** Viswanath has quit IRC | 00:29 | |
| *** zyluo has joined #openstack-ironic | 00:29 | |
| *** shakamunyi has joined #openstack-ironic | 00:31 | |
| *** rushiagr is now known as rushiagr_away | 00:44 | |
| *** zhangyang_ has quit IRC | 00:47 | |
| *** penick has quit IRC | 00:49 | |
| *** ryanpetrello has joined #openstack-ironic | 00:56 | |
| *** rloo has quit IRC | 01:16 | |
| *** achanda has quit IRC | 01:19 | |
| *** achanda has joined #openstack-ironic | 01:20 | |
| *** achanda has quit IRC | 01:25 | |
| naohirot | GM ironic team | 01:28 |
|---|---|---|
| *** Haomeng has quit IRC | 01:39 | |
| *** ryanpetrello has quit IRC | 01:39 | |
| *** Haomeng has joined #openstack-ironic | 01:42 | |
| *** Haomeng has quit IRC | 01:47 | |
| *** Viswanath has joined #openstack-ironic | 01:47 | |
| *** Viswanath has quit IRC | 01:51 | |
| *** anderbubble has joined #openstack-ironic | 01:55 | |
| *** nosnos has joined #openstack-ironic | 02:06 | |
| *** dlaube has quit IRC | 02:16 | |
| *** spandhe has quit IRC | 02:16 | |
| *** ChuckC has quit IRC | 02:17 | |
| *** marcoemorais has quit IRC | 02:21 | |
| *** openstackgerrit has quit IRC | 02:34 | |
| *** anderbubble has quit IRC | 02:46 | |
| *** anderbubble has joined #openstack-ironic | 02:50 | |
| *** jrist has joined #openstack-ironic | 03:01 | |
| *** Haomeng has joined #openstack-ironic | 03:17 | |
| *** yuanying has joined #openstack-ironic | 03:18 | |
| *** Viswanath has joined #openstack-ironic | 03:22 | |
| *** Viswanath has quit IRC | 03:25 | |
| *** achanda has joined #openstack-ironic | 03:27 | |
| *** shakamunyi has quit IRC | 03:32 | |
| *** teju has joined #openstack-ironic | 03:33 | |
| *** __mohit__ has joined #openstack-ironic | 03:41 | |
| *** __mohit__ has quit IRC | 03:41 | |
| *** yuanying has quit IRC | 03:46 | |
| *** yuanying has joined #openstack-ironic | 03:46 | |
| *** zyluo has quit IRC | 03:50 | |
| *** rameshg87 has joined #openstack-ironic | 03:53 | |
| naohirot | rameshg87: Hello | 04:14 |
| naohirot | sorry for not being able to reply to you | 04:15 |
| naohirot | yesterday | 04:15 |
| *** pensu has joined #openstack-ironic | 04:15 | |
| naohirot | I certainly looked at the URL https://github.com/hpproliant/proliantutils/ | 04:16 |
| rameshg87 | naohirot, hello | 04:16 |
| naohirot | Hi | 04:17 |
| rameshg87 | naohirot, did i answer your question yesterday ? :) | 04:17 |
| naohirot | Yes, you did :-) thanks! | 04:17 |
| rameshg87 | naohirot, great :) | 04:17 |
| *** Nisha has joined #openstack-ironic | 04:18 | |
| naohirot | The page https://wiki.openstack.org/wiki/Ironic/Drivers/iLODrivers | 04:18 |
| naohirot | is excellent. | 04:18 |
| naohirot | it's very helpful to understand what is iLO driver. | 04:19 |
| naohirot | I'm still in the stage to understand overall picture. | 04:20 |
| naohirot | I submitted 4 blue prints, and right now I'm in the process of witting the spec of the first one, pxe_irmc driver. | 04:22 |
| *** pensu has quit IRC | 04:23 | |
| naohirot | do you think that I should write similar spec as | 04:23 |
| naohirot | https://github.com/openstack/ironic-specs/blob/master/specs/juno/drac-management-driver.rst | 04:24 |
| naohirot | and | 04:24 |
| naohirot | https://github.com/openstack/ironic-specs/blob/master/specs/kilo/ilo-management-interface.rst | 04:24 |
| naohirot | ? | 04:24 |
| *** pcrews has quit IRC | 04:33 | |
| *** subscope has joined #openstack-ironic | 04:36 | |
| rameshg87 | naohirot, sorry, i just saw your messages now | 04:47 |
| rameshg87 | naohirot, may be you can add my name to the messages if it was to me so that i will get a notification | 04:47 |
| naohirot | that's ok, no problem.:-) | 04:47 |
| naohirot | Okay, I'll do that. | 04:48 |
| rameshg87 | naohirot, are you planning to have a management interface for your hardware ? if so, you should write a similar spec | 04:51 |
| naohirot | rameshg87: yes, pxe_irmc implements like pxe_ilo and pxe_drac. | 04:54 |
| naohirot | rameshg87: I believe that pxe part is common. the different part are get/set boot device, and power on/off. | 04:56 |
| *** jrist has joined #openstack-ironic | 04:57 | |
| rameshg87 | naohirot, yes | 04:57 |
| naohirot | rameshg87: So maybe I need to write similar spec with https://github.com/openstack/ironic-specs/blob/master/specs/juno/drac-power-driver.rst too. | 04:57 |
| rameshg87 | naohirot, so you may submit two different specs - one for irmc power interface and one for irmc management interface | 04:57 |
| rameshg87 | naohirot, yes | 04:58 |
| naohirot | rameshg87: I see. that make sense to me. thanks! | 04:58 |
| *** pensu has joined #openstack-ironic | 05:18 | |
| *** loki184 has joined #openstack-ironic | 05:21 | |
| *** rakesh_hs has joined #openstack-ironic | 05:44 | |
| *** lazy_prince has quit IRC | 05:57 | |
| *** lazy_prince has joined #openstack-ironic | 05:57 | |
| *** rushiagr_away is now known as rushiagr | 06:14 | |
| *** pradipta_away is now known as pradipta | 06:28 | |
| *** k4n0 has joined #openstack-ironic | 06:37 | |
| *** subscope has quit IRC | 06:55 | |
| *** achanda has quit IRC | 06:57 | |
| *** achanda has joined #openstack-ironic | 07:01 | |
| *** achanda has quit IRC | 07:10 | |
| *** anderbubble has quit IRC | 07:14 | |
| *** erwan_taf has joined #openstack-ironic | 07:35 | |
| *** chenglch has joined #openstack-ironic | 07:46 | |
| *** dlpartain has joined #openstack-ironic | 07:50 | |
| *** dtantsur|afk is now known as dtantsur | 07:52 | |
| dtantsur | Morning Ironic | 07:52 |
| naohirot | dtantsur: good morning | 07:53 |
| *** erwan_taf has quit IRC | 08:01 | |
| naohirot | dtantsur: isn't there a bug which can be fixed easily? | 08:02 |
| dtantsur | lemme see | 08:02 |
| naohirot | dtantsur: If there were, I'd like to try it to learn the process of the ironic development. | 08:03 |
| *** jistr has joined #openstack-ironic | 08:08 | |
| Haomeng | naohirot: welcome, you can take this as reference - http://docs.openstack.org/developer/ironic/dev/dev-quickstart.html | 08:13 |
| naohirot | Haomeng: Hi Haomeng | 08:14 |
| naohirot | Haomeng: I went through the page. | 08:14 |
| naohirot | Haomeng: pxe_ssh driver worked ok, but agent_ssh always became error due to timeout. | 08:15 |
| Haomeng | naohirot: ok | 08:15 |
| Haomeng | naohirot: have to check agent side log | 08:15 |
| Haomeng | naohirot: so far we have no take agent log back to conductor side, so it it diffcult to debug, have to check physical machine console output | 08:16 |
| Haomeng | naohirot: to see the hang point | 08:16 |
| naohirot | Haomeng: I checked the log, and I found three error in screen conductor's log. | 08:16 |
| *** erwan_taf has joined #openstack-ironic | 08:16 | |
| Haomeng | naohirot: can you paste the error message via http://paste.openstack.org/ | 08:16 |
| naohirot | Haomeng: Ok, just moment. | 08:17 |
| Haomeng | naohirot: no rush:) | 08:17 |
| dtantsur | naohirot, that's what we use to search for easy bugs: https://bugs.launchpad.net/ironic/+bugs?field.tag=low-hanging-fruit | 08:17 |
| dtantsur | naohirot, have a look at e.g. https://bugs.launchpad.net/ironic/+bug/1277282 | 08:18 |
| dtantsur | or https://bugs.launchpad.net/ironic/+bug/1326269 | 08:18 |
| naohirot | Haomeng: I pasted my log. | 08:19 |
| Haomeng | naohirot: ok | 08:19 |
| naohirot | dtantsur: thanks, I read the bug/1326269. | 08:24 |
| naohirot | dtantsur: Drivers page (doc/source/dev/drivers.rst) should be updated. Do you think how many lines of explanation is expected? | 08:26 |
| dtantsur | oh I don't know :) imagine you just came to Ironic and read this new explanation. what will be sufficient for you? | 08:26 |
| *** jcoufal has joined #openstack-ironic | 08:28 | |
| naohirot | dtantsur: I see. When I went through the quick start page, I also found some places which more explanation should be added. | 08:29 |
| dtantsur | naohirot, feel free to propose changes, that's always very good to have feedback from fresh point of view :) | 08:30 |
| naohirot | dtantsur: So I can submit similar improvement suggestion as a bug, and fix it by myself. | 08:30 |
| dtantsur | naohirot, for small improvements you don't need a bug IMO | 08:31 |
| naohirot | dtanstsur: Okay, I'll try, please assign bug/1326269 to me. | 08:32 |
| dtantsur | naohirot, you can assign bugs to yourself (especially since I don't know your ID) | 08:33 |
| naohirot | dtantsur: really? | 08:34 |
| dtantsur | naohirot, yeah, press edit button next to " Unassigned" in "Assigned to" column | 08:35 |
| naohirot | dtantsur: In case of "TRIAGED", who triaged? | 08:35 |
| dtantsur | naohirot, didn't quite get your question. Assignee is who is working on the bug. Triager is one of us that sets priority and confirms bugs | 08:36 |
| Haomeng | naohirot: yes, just click 'edit' icon in 'assigned to' colume and click 'assing me', done:) | 08:37 |
| naohirot | dtantsur: I see, "TRIAGED" is one of state of a bug toward "FIX COMMITTED". | 08:38 |
| *** Nisha has quit IRC | 08:38 | |
| dtantsur | naohirot, triaged means "we understand how to fix it", while "confirmed" is "we understand it's a bug, but still not sure how to fix it" | 08:39 |
| dtantsur | does it make things clearer? :) | 08:39 |
| naohirot | dtantsur: Haomeng: Yes, very clear. I understood. Thanks! :-) | 08:40 |
| *** ifarkas has joined #openstack-ironic | 08:41 | |
| Haomeng | naohirot: welcome | 08:41 |
| Haomeng | naohirot: feel free to ask question here, all of us can help you, good luck | 08:42 |
| naohirot | Haomeng: Thank you for saying that. Here is easier than mailing list to ask such basic question :-) | 08:44 |
| *** rakesh_hs has quit IRC | 08:44 | |
| *** rakesh_hs has joined #openstack-ironic | 08:46 | |
| *** Nisha has joined #openstack-ironic | 08:46 | |
| Haomeng | naohirot: yes, we are all same way from basic... :) | 08:46 |
| Nisha | dtantsur: hi | 08:47 |
| *** pradipta is now known as pradipta_away | 08:48 | |
| Haomeng | naohirot: based on your log message - FailedToUpdateDHCPOptOnPort: No VIFs found for node ca91bacc-e943-4e35-896a-47c9a275b976 when attempting to update DHCP BOOT options. | 08:49 |
| Haomeng | naohirot: it should be neutron issue | 08:49 |
| Haomeng | naohirot: can you check neutron logs | 08:49 |
| Haomeng | naohirot: looks like the nif is not ready during bm provisioning | 08:49 |
| naohirot | Haomeng: Okay, you got the log. | 08:50 |
| Haomeng | naohirot: have to run neutron command to check the dhcp port information by instance id | 08:50 |
| dtantsur | Nisha, hi | 08:50 |
| Haomeng | naohirot: yes, i got your log | 08:51 |
| Nisha | dtantsur: i was seeing ur comments on discovery spec | 08:51 |
| Nisha | i made management interface to create ports | 08:51 |
| Nisha | as per the spec | 08:51 |
| Nisha | dtantsur: do you expect anything else for in-band? | 08:51 |
| *** romcheg has joined #openstack-ironic | 08:52 | |
| Nisha | dtantsur: yesterday, lucasgomes told that its decided that in-band will not be done via generic, while it will be third party driver | 08:53 |
| Nisha | dtantsur: please give comments in the spec where all you see the changes will be required for in-band | 08:53 |
| naohirot | Haomeng: do you think that there is some relation between first two JSONDecodeError errors and the last VIFs error? | 08:53 |
| naohirot | Haomeng: I put the url here too. http://paste.openstack.org/show/132230/ | 08:55 |
| Nisha | dtantsur: excerpt from yesterday's discussion: "2014-11-11T10:05:33 <lucasagomes> Nisha, no in-band is out of ironic codebase 2014-11-11T10:05:51 <lucasagomes> if people is willing to do in-band discovery they will need a third-party service to" | 08:55 |
| Haomeng | naohirot: dont think so,ok will check your new log | 08:55 |
| naohirot | Haomeng: the url I put here is the same log, not new log. | 08:56 |
| Haomeng | naohirot: yes, same one | 08:58 |
| Haomeng | naohirot: can you try with pxe_ipmitool driver? | 08:58 |
| Haomeng | naohirot: it is easy one then IPA | 08:59 |
| dtantsur | Nisha, that does not contradict, we'll still need discoverd for in-band. But the common API should invoke it. | 08:59 |
| Nisha | dtantsur: the management interface is the one i remeber was your comment for that | 09:00 |
| Haomeng | naohirot: how did you run ironic, with devstack? | 09:00 |
| naohirot | Haomeng: Yes I can try pxe_ipmitool driver. I confirmed that pxe_ssh in my environment. | 09:00 |
| Haomeng | naohirot: yes, pxe_ssh is easy for dev/debug | 09:01 |
| *** rushiagr has quit IRC | 09:01 | |
| Haomeng | naohirot: pxe_ipmitool is used for real physical node I think if you have hardware server | 09:01 |
| *** rushiagr has joined #openstack-ironic | 09:02 | |
| naohirot | Haomeng: I followed the exactly same steps as the page http://docs.openstack.org/developer/ironic/dev/dev-quickstart.html#deploying-ironic-with-devstack. | 09:02 |
| Haomeng | naohirot: yes, that is good | 09:02 |
| Nisha | dtantsur: it will be better if you could provide the comment in spec where and how you feel the interface shall be modified to support in-band. i have less hands on in-band :( | 09:03 |
| Haomeng | the default driver is "IRONIC_DEPLOY_DRIVER" with this document | 09:03 |
| naohirot | Haomeng: Yes, I have a physical server. | 09:03 |
| Haomeng | naohirot: so if you want to run with pxe_ssh driver, have to change the ironic.conf file to enable it | 09:03 |
| Haomeng | naohirot: cool | 09:03 |
| dtantsur | Nisha, readding comments now | 09:03 |
| Haomeng | naohirot: can you check from neutron side | 09:04 |
| Haomeng | naohirot: neutron port-list | 09:04 |
| Haomeng | naohirot: to check if the dhcp port is created already | 09:04 |
| Haomeng | naohirot: will be back after a while, good luck:) | 09:05 |
| naohirot | Haomeng: Okay, I'll try it, and see the dhcp port. | 09:05 |
| *** jistr has quit IRC | 09:06 | |
| naohirot | Haomeng: Thanks a lot. | 09:06 |
| dtantsur | Nisha, done | 09:07 |
| *** foexle has joined #openstack-ironic | 09:08 | |
| dtantsur | Nisha, tl;dr the only difference between OOB and IB is that IB is _essentially_ async, so you can't rely on it returning values right away | 09:09 |
| dtantsur | Nisha, that may potentially by the case for some OOB too | 09:09 |
| Nisha | dtantsur: ok. So i have proposed discovery APIs and management interface to be both synchronous and asynchronous as per driver choice. | 09:10 |
| *** derekh has joined #openstack-ironic | 09:10 | |
| dtantsur | Nisha, lucas meant that we won't _implement_ IB discovery as part of Ironic (and that's true) | 09:10 |
| Nisha | dtantsur: :) | 09:10 |
| Nisha | ok i will just address the comments in the spec | 09:11 |
| dtantsur | Nisha, they are always async, but the problem is that you expect driver discover_hardware_properties to return something, which may not be possible | 09:11 |
| dtantsur | Nisha, yes please :) | 09:11 |
| Nisha | dtantsur: :) i didnt realize that it requires change | 09:12 |
| Nisha | dtantsur: thanks for the comments. | 09:12 |
| dtantsur | Nisha, see http://kilodesignsummit.sched.org/event/6be748864a8d14ee914d292bd0e8c816#.VGMkfH-LL5s : it's stated as " requires a solution which accommodates both in- and out-of-band methods within a common API. " | 09:13 |
| dtantsur | that's why I'm thinking that your API will need to _potentially_ support both | 09:13 |
| dtantsur | no problem, you're welcome :) | 09:13 |
| Nisha | dtantsur: yes thanks | 09:13 |
| *** igordcard has joined #openstack-ironic | 09:19 | |
| *** athomas has joined #openstack-ironic | 09:20 | |
| *** dlpartain has quit IRC | 09:23 | |
| *** andreykurilin has joined #openstack-ironic | 09:25 | |
| *** jistr has joined #openstack-ironic | 09:26 | |
| *** lucasagomes has joined #openstack-ironic | 09:26 | |
| *** ndipanov_gone is now known as ndipanov | 09:29 | |
| *** rakesh_hs has quit IRC | 09:31 | |
| pensu | Hi dtantsur | 09:32 |
| dtantsur | pensu, hi | 09:33 |
| pensu | dtantsur: I have a question regarding this: http://docs.openstack.org/developer/ironic/deploy/install-guide.html#configure-neutron-to-communicate-with-the-bare-metal-server | 09:33 |
| pensu | dtantsur: what is 'physnet1' here supposed to be? | 09:33 |
| pensu | dtantsur: the same physical interface which is connected to bare metal node? | 09:33 |
| dtantsur | I don't really know neutron, sorry :( | 09:34 |
| dtantsur | Haomeng, any ideas ^^^ | 09:34 |
| *** rakesh_hs has joined #openstack-ironic | 09:35 | |
| *** MattMan has quit IRC | 09:39 | |
| *** MattMan has joined #openstack-ironic | 09:40 | |
| pensu | dtantsur: np..:) lucasagomes: jroll: ^^ | 09:51 |
| dtantsur | brb | 09:51 |
| Haomeng | dtantsur: I am back | 10:00 |
| Haomeng | dtantsur: our team has neutron expert, so I learn with her:) | 10:01 |
| *** sambetts has joined #openstack-ironic | 10:01 | |
| Haomeng | pensu: for physnet1, that is defined with neutron to connect to physical and define the vlan tag numbers | 10:01 |
| Haomeng | pensu: "bridge_mappings = physnet1:br-eth2" means that the br-eth2 will take as physical network connection with eth2, and eth2 will be connected with switch on the vlan trunk/access port with fix vlan tag defined by switch | 10:03 |
| *** andreykurilin has quit IRC | 10:05 | |
| Haomeng | pensu: so, for the example in document, we will take eth2 physical as the external physical network connection with our neutron | 10:05 |
| *** achanda has joined #openstack-ironic | 10:11 | |
| Nisha | dtantsur: i posted the patch with comments addressed and some questions in comments for previous patch. | 10:15 |
| *** achanda has quit IRC | 10:15 | |
| *** krtaylor has quit IRC | 10:16 | |
| Nisha | pensu: the eth2 is the nic/mac address of the cloud controller which connects to the baremetal | 10:19 |
| *** chenglch has quit IRC | 10:22 | |
| *** Nisha has quit IRC | 10:23 | |
| *** krtaylor has joined #openstack-ironic | 10:24 | |
| pensu | Haomeng: Nisha: got it, guys, thanks! :) | 10:32 |
| Haomeng | pensu: wel, any time | 10:33 |
| *** dprince has joined #openstack-ironic | 10:39 | |
| *** nosnos has quit IRC | 10:39 | |
| *** ndipanov has quit IRC | 10:56 | |
| *** ndipanov has joined #openstack-ironic | 10:58 | |
| *** romcheg1 has joined #openstack-ironic | 11:00 | |
| *** romcheg has quit IRC | 11:00 | |
| *** nosnos has joined #openstack-ironic | 11:02 | |
| *** romcheg1 has quit IRC | 11:03 | |
| *** romcheg has joined #openstack-ironic | 11:03 | |
| *** loki184 has quit IRC | 11:08 | |
| *** ndipanov has quit IRC | 11:14 | |
| *** rakesh_hs2 has joined #openstack-ironic | 11:20 | |
| *** rakesh_hs has quit IRC | 11:23 | |
| *** ndipanov has joined #openstack-ironic | 11:27 | |
| rameshg87 | ifarkas, hi | 11:28 |
| ifarkas | rameshg87, hi | 11:28 |
| rameshg87 | ifarkas, got some time to discuss regarding raid configuration ? | 11:29 |
| *** erwan_taf has quit IRC | 11:29 | |
| ifarkas | rameshg87, yeah, I have | 11:29 |
| ifarkas | rameshg87, did you see the drac spec? | 11:29 |
| rameshg87 | ifarkas, yeah i saw that | 11:30 |
| rameshg87 | ifarkas, it talks about vendor passthru right ? are you still planning for a vendor passthru for raid configuration ? | 11:31 |
| ifarkas | rameshg87, I am planning to update it based on our discussion to propose a raid interface | 11:31 |
| rameshg87 | ifarkas, yes | 11:31 |
| rameshg87 | ifarkas, we could go for adding raid support in long-term running tasks (that happens before the node reaches available state) | 11:32 |
| ifarkas | rameshg87, did you have a chance to discuss it with the rackspace folks? | 11:33 |
| rameshg87 | ifarkas, i talked to JayF, he wants us to pursue the hardware capabilities spec https://review.openstack.org/#/c/131272/ | 11:34 |
| ifarkas | rameshg87, I meant if you had a chance to discuss with them whether it should go to zapping or prepare | 11:35 |
| rameshg87 | ifarkas, ah no :( | 11:35 |
| *** pensu has quit IRC | 11:35 | |
| rameshg87 | ifarkas, http://eavesdrop.openstack.org/irclogs/%23openstack-ironic/%23openstack-ironic.2014-11-11.log (search for "2014-11-11T18:21:38") | 11:36 |
| ifarkas | rameshg87, ok, no worries | 11:36 |
| rameshg87 | ifarkas, will check with them today | 11:36 |
| rameshg87 | ifarkas, for us, we are targetting the inband stuff, so would like to put it in zapping | 11:36 |
| rameshg87 | ifarkas, we are planning to do raid configuration through inband utility within ipa | 11:37 |
| ifarkas | rameshg87, oh, so you no longer prefer prepare? | 11:37 |
| rameshg87 | ifarkas, kind of not :) | 11:37 |
| rameshg87 | ifarkas, the reason is our out-of-band approach isn't fully capable of doing raid configuration, there are some challenges | 11:37 |
| *** rakesh_hs2 has quit IRC | 11:37 | |
| ifarkas | rameshg87, ok, then I will propose the spec this way and then everyone can comment on it | 11:37 |
| rameshg87 | ifarkas, so you want to continue with vendor passthru ? | 11:38 |
| ifarkas | rameshg87, no, I will try to propose a raid interface | 11:38 |
| rameshg87 | ifarkas, okay .. | 11:38 |
| rameshg87 | ifarkas, and do them as part of zapping or prepare ? | 11:38 |
| ifarkas | rameshg87, zapping is fine for me as well | 11:39 |
| rameshg87 | ifarkas, great | 11:39 |
| *** rakesh_hs has joined #openstack-ironic | 11:39 | |
| rameshg87 | ifarkas, so we could have the same interface to the operator atleast | 11:39 |
| ifarkas | rameshg87, yeah, that would be great | 11:39 |
| rameshg87 | ifarkas, i am thinking about this | 11:39 |
| rameshg87 | ifarkas, the operator would say on what raid he wants the node to be configured | 11:40 |
| rameshg87 | ifarkas, the node would create the maximum possible disk of that raid configuration and then return the disk information back to the properties | 11:40 |
| rameshg87 | ifarkas, i am not sure if we can accept the size of the logical drive to be created | 11:41 |
| ifarkas | rameshg87, I think it should be configurable | 11:41 |
| rameshg87 | ifarkas, because different raid configurations have different space utilizations | 11:41 |
| ifarkas | rameshg87, eg. for storage node, you want to have an array for the os and then another one for the actual storage device | 11:42 |
| rameshg87 | ifarkas, hmm.. yeah i agree | 11:42 |
| rameshg87 | ifarkas, but what worries me is this | 11:42 |
| rameshg87 | ifarkas, if i have 10x50G disks on the RAID controller | 11:43 |
| rameshg87 | ifarkas, if operator says RAID0, i can create logical disks update 500G | 11:43 |
| rameshg87 | ifarkas, if operator says RAID1, i can create only upto 250G | 11:44 |
| rameshg87 | ifarkas, so we will have to do the math after checking the physical drive information. | 11:44 |
| rameshg87 | ifarkas, if not sufficient disks, we might need to error out | 11:44 |
| ifarkas | rameshg87, hmm. don't we want to leave the calculation for the operator? | 11:45 |
| rameshg87 | ifarkas, so we would assume that the input is correct (RAID level and logical disk sizes) and go ahead with the configuration ? | 11:46 |
| ifarkas | ifarkas, yes | 11:46 |
| ifarkas | rameshg87, if it fails, the operator will eventually find that out | 11:46 |
| rameshg87 | ifarkas, hmm yeah that's fine | 11:47 |
| rameshg87 | ifarkas, so we had talked about zapping would happen to move the node from init -> available, right ? | 11:48 |
| ifarkas | rameshg87, cool. and if later we think that would be a useful feature, we can add it later | 11:48 |
| ifarkas | rameshg87, yeah | 11:48 |
| rameshg87 | ifarkas, yeah makes sense .. may be later | 11:48 |
| rameshg87 | ifarkas, when the node is in init, the operator will have to put this information into the node. RAID level and the size of logical disks. probably in driver_info ? | 11:49 |
| ifarkas | rameshg87, yeah, that's a good place for that | 11:50 |
| rameshg87 | ifarkas, okay. so once the zapping task for raid configuration is done, it would update the capabilities | 11:51 |
| rameshg87 | ifarkas, and properties for the disk size | 11:51 |
| rameshg87 | ifarkas, and that information would go to the nova scheduler to allow user to choose a node | 11:51 |
| ifarkas | rameshg87, yeah, devananda also added this comment to my spec | 11:52 |
| rameshg87 | ifarkas, in flavor we could set disk=50 and in flavor's extra spec will have raid_level=1 | 11:52 |
| rameshg87 | ifarkas, if user never cared about raids, it wouldn't become a scheduling criteria and the node can still be allocated to the user | 11:53 |
| ifarkas | rameshg87, and if it is specified then nova would be able to pick a correct one | 11:53 |
| rameshg87 | ifarkas, yeah .. | 11:54 |
| rameshg87 | ifarkas, there we go .. we have the solution :) | 11:54 |
| ifarkas | rameshg87, excellent ;-) | 11:54 |
| rameshg87 | ifarkas, i can still make use of the raid interface that you would be using | 11:55 |
| ifarkas | rameshg87, what do you mean? | 11:55 |
| rameshg87 | ifarkas, if node is already booted to ipa ramdisk, i can use the interface to talk to ipa and get it done | 11:56 |
| rameshg87 | ifarkas, for example create_virtual_disk() call, i could talk to ipa to get it done | 11:56 |
| ifarkas | rameshg87, yeah , right | 11:57 |
| rameshg87 | ifarkas, will you doing the creation/deletion synchronously in oob drac approach ? | 11:59 |
| *** dlpartain has joined #openstack-ironic | 11:59 | |
| rameshg87 | ifarkas, but there wouldn't be a chance of call-back in oob approach, right. that was a stupid question i think :) | 12:00 |
| ifarkas | rameshg87, right, and synchronous is not possible with the drac | 12:00 |
| rameshg87 | ifarkas, as part of zapping, so would you trigger a job in drac controller, and keep polling the status of the job ? | 12:01 |
| ifarkas | rameshg87, yes | 12:01 |
| rameshg87 | ifarkas, okay | 12:01 |
| rameshg87 | ifarkas, i could trigger the job in ipa, and ipa could notify it back once it is done | 12:02 |
| ifarkas | rameshg87, cool | 12:03 |
| rameshg87 | ifarkas, i was just thinking what all should be part of the raid interface | 12:03 |
| rameshg87 | ifarkas, should it be just logical disk size and raid level ? | 12:03 |
| rameshg87 | ifarkas, the driver could select the raid controller and the physical disks to use for creating the logical disk | 12:04 |
| ifarkas | rameshg87, oh, right. What if there are multiple raid controllers? | 12:04 |
| rameshg87 | ifarkas, yeah, there could be multiple raid controllers | 12:05 |
| ifarkas | rameshg87, according to my original proposal the workflow would be for the operator to first list the physical disks and raid controllers and then create the virtual disk | 12:05 |
| rameshg87 | ifarkas, yeah but that would be too much for the operator to do when there are 100 nodes | 12:05 |
| ifarkas | rameshg87, true | 12:06 |
| rameshg87 | ifarkas, i guess we could just take raid level and logical disk sizes as input from operator for raid configuration | 12:06 |
| ifarkas | rameshg87, the input should also contain the controller, shouldn't it? | 12:07 |
| rameshg87 | ifarkas, but wouldn't that vary across different nodes ? | 12:07 |
| rameshg87 | ifarkas, does the operator really care which controller it is configured in ? | 12:08 |
| ifarkas | rameshg87, hmm, not sure | 12:08 |
| ifarkas | rameshg87, ok, so we can pick the first one if there are multiple controllers | 12:08 |
| rameshg87 | ifarkas, yeah | 12:08 |
| rameshg87 | ifarkas, and we could allow them to specify optional inputs like stripe size (which you have mentioned in your spec) | 12:09 |
| ifarkas | rameshg87, yeah | 12:10 |
| rameshg87 | ifarkas, and then there would be a delete_disk which again takes the logical disk size (and optional raid level ??) to delete the disk | 12:11 |
| ifarkas | rameshg87, delete_disk should just have the disk id as an input, not? | 12:11 |
| ifarkas | rameshg87, why does it care about disk size? | 12:12 |
| rameshg87 | ifarkas, but where is this disk id ? | 12:12 |
| rameshg87 | ifarkas, from raid controller ? is it applicable to all logical disks created ? | 12:12 |
| *** dprince has quit IRC | 12:13 | |
| *** dlpartain1 has joined #openstack-ironic | 12:13 | |
| ifarkas | rameshg87, how about adding the id to capabilities? | 12:14 |
| rameshg87 | ifarkas, yeah that makes sense | 12:14 |
| ifarkas | rameshg87, the problem is what ironic will do if there's to logical disk with the same size | 12:14 |
| rameshg87 | ifarkas, but thinking about cloud use-case, i don't know if we will ever delete a single logical disk | 12:14 |
| ifarkas | s/to/two | 12:14 |
| rameshg87 | ifarkas, just imagining the use-cases | 12:15 |
| *** dlpartain has quit IRC | 12:15 | |
| *** teju has left #openstack-ironic | 12:15 | |
| rameshg87 | ifarkas, when node is enrolled, operator says i need two logical disks of 100G each with RAID1 on this node | 12:15 |
| rameshg87 | ifarkas, and we create it as part of zapping | 12:15 |
| rameshg87 | ifarkas, the node gets scheduled to someone, then at some point later is teared down, and disks are erased as part of decommissioning | 12:16 |
| ifarkas | rameshg87, right, so you would prefer something like clear_raid_config? | 12:16 |
| rameshg87 | ifarkas, exactly | 12:16 |
| rameshg87 | ifarkas, :) | 12:16 |
| ifarkas | rameshg87, ok | 12:16 |
| *** lucasagomes is now known as lucas-hungry | 12:17 | |
| rameshg87 | ifarkas, unless operator wants to change the raid configuration | 12:17 |
| rameshg87 | ifarkas, i don't see why they would delete a single disk | 12:17 |
| ifarkas | rameshg87, I agree that's not a very likely usecase | 12:17 |
| rameshg87 | ifarkas, okay | 12:17 |
| rameshg87 | ifarkas, so we would have 2 method for raid interface - create_raid_configuration and delete_raid_configuration | 12:18 |
| rameshg87 | ifarkas, ?? | 12:18 |
| ifarkas | rameshg87, yes | 12:18 |
| rameshg87 | ifarkas, create_raid_configuration accepts a list of logical disk information [ {disk_size:50, raid_level:1}, {disk_size:100, raid_level:5}, .. ] | 12:19 |
| rameshg87 | ifarkas, does the above make sense ? | 12:19 |
| ifarkas | rameshg87, absolutely ;-) | 12:19 |
| rameshg87 | ifarkas, and clear_raid_configuration wouldn't anyway accept anything :) | 12:19 |
| ifarkas | rameshg87, yeah, basically none of them has input because they are using the driver_info field, right? | 12:20 |
| rameshg87 | ifarkas, yeah .. | 12:21 |
| rameshg87 | ifarkas, would we overload driver_info with raid configuration data ? | 12:21 |
| ifarkas | rameshg87, otherwise how would ironic do it during zapping? | 12:22 |
| rameshg87 | ifarkas, or may be extra ? | 12:22 |
| ifarkas | rameshg87, yeah, that would also work | 12:22 |
| rameshg87 | ifarkas, yeah, may be we can start with extra and see what others think. what do you say ? | 12:23 |
| ifarkas | rameshg87, not sure which one is prefer for this type of data | 12:23 |
| ifarkas | rameshg87, ack | 12:23 |
| rameshg87 | ifarkas, just one question | 12:24 |
| rameshg87 | ifarkas, isn't swapping and decommissioning one and the same ? | 12:25 |
| ifarkas | rameshg87, I guess you meant zapping | 12:26 |
| rameshg87 | ifarkas, oh i meant zapping | 12:26 |
| rameshg87 | ifarkas, yeah :) | 12:26 |
| ifarkas | rameshg87, decommissioning is supposed to be part of zapping | 12:26 |
| rameshg87 | ifarkas, got it | 12:26 |
| ifarkas | so yes ;-) | 12:26 |
| rameshg87 | ifarkas, and we missed one part | 12:26 |
| rameshg87 | ifarkas, get_current_raid_configuration should also be part of the interface | 12:27 |
| *** pensu has joined #openstack-ironic | 12:27 | |
| rameshg87 | ifarkas, everytime the node is returned back, we could make sure that the node is still configured the same raid configuration that the operator wishes | 12:27 |
| rameshg87 | ifarkas, if it is not, we could just delete_raid_configuration and then do create_raid_configuration again | 12:28 |
| rameshg87 | ifarkas, so that we don't have to do this again and again until it really changes | 12:28 |
| rameshg87 | ifarkas, what do you say ? | 12:28 |
| ifarkas | rameshg87, ok, how about naming it as check_current_raid_config because it must be async and get presumes synchronicity? | 12:29 |
| *** rushiagr is now known as rushiagr_away | 12:31 | |
| rameshg87 | ifarkas, and how would it stay async ? | 12:32 |
| *** romcheg has quit IRC | 12:32 | |
| *** romcheg has joined #openstack-ironic | 12:33 | |
| rameshg87 | ifarkas, i am really sorry, i would need to leave now :( | 12:33 |
| ifarkas | rameshg87, okay, we can discuss it later, or tomorrow | 12:34 |
| rameshg87 | ifarkas, i will catch you later, may be one round of discussion more, we should have something | 12:34 |
| ifarkas | rameshg87, I need to think about your q | 12:34 |
| ifarkas | rameshg87, definietly | 12:34 |
| rameshg87 | ifarkas, see you then .. bye :) | 12:34 |
| ifarkas | rameshg87, bye ;-) | 12:34 |
| *** rameshg87 has quit IRC | 12:34 | |
| *** ryanpetrello has joined #openstack-ironic | 12:38 | |
| *** erwan_taf has joined #openstack-ironic | 12:40 | |
| *** Haomeng|2 has joined #openstack-ironic | 12:46 | |
| *** Haomeng has quit IRC | 12:47 | |
| *** naohirot has quit IRC | 13:00 | |
| *** pelix has joined #openstack-ironic | 13:01 | |
| *** krtaylor has quit IRC | 13:01 | |
| *** dprince has joined #openstack-ironic | 13:01 | |
| *** nosnos has quit IRC | 13:10 | |
| *** krtaylor has joined #openstack-ironic | 13:15 | |
| *** lucas-hungry is now known as lucasagomes | 13:18 | |
| *** dlpartain1 has quit IRC | 13:35 | |
| *** k4n0 has quit IRC | 13:46 | |
| *** lucasagomes_ has joined #openstack-ironic | 13:50 | |
| *** lucasagomes has quit IRC | 13:54 | |
| *** sambetts has quit IRC | 13:55 | |
| *** sambetts has joined #openstack-ironic | 13:56 | |
| NobodyCam | morning Ironic | 13:59 |
| vdrok | hi everyone! | 14:10 |
| romcheg | Morning Ironic! | 14:10 |
| vdrok | morning NobodyCam | 14:10 |
| romcheg | Hi vdrok! | 14:10 |
| vdrok | afternoon romcheg ! | 14:10 |
| NobodyCam | mornig vdrok romcheg | 14:11 |
| * NobodyCam just got back home lastnight | 14:11 | |
| vdrok | lucasagomes_, hi | 14:11 |
| dtantsur | morning folks :) | 14:12 |
| vdrok | hi dtantsur | 14:12 |
| *** rakesh_hs has quit IRC | 14:17 | |
| *** lucasagomes_ is now known as lucasagomes | 14:17 | |
| lucasagomes | NobodyCam, morning | 14:17 |
| lucasagomes | vdrok, hi there | 14:18 |
| vdrok | lucasagomes, yuriyz told that there is some agreement on this one https://review.openstack.org/#/c/130228/ ? | 14:18 |
| lucasagomes | vdrok, yeah I brought it up on the "making ironic easier to use" session | 14:19 |
| lucasagomes | overall I think people agree it's a good idea | 14:19 |
| lucasagomes | it was just a mention, I'm looking fwd to see the comments from others on that review | 14:20 |
| NobodyCam | morning lucasagomes | 14:20 |
| NobodyCam | mornig dtantsur | 14:20 |
| vdrok | lucasagomes, aha, ok, thanks | 14:21 |
| lucasagomes | vdrok, np, I think there's an etherpad on that session with what was proposed/discussed for reference | 14:23 |
| lucasagomes | https://wiki.openstack.org/wiki/Summit/Kilo/Etherpads#Ironic | 14:23 |
| vdrok | oh, thanks, will read | 14:23 |
| *** anderbubble has joined #openstack-ironic | 14:25 | |
| *** anderbubble has quit IRC | 14:27 | |
| *** shakamunyi has joined #openstack-ironic | 14:31 | |
| *** rloo has joined #openstack-ironic | 14:31 | |
| *** krtaylor has quit IRC | 14:33 | |
| lucasagomes | [offtopic | 14:35 |
| *** pensu has quit IRC | 14:35 | |
| * lucasagomes hits enter too quick | 14:36 | |
| lucasagomes | [offtopic] someone here uses google play music? | 14:36 |
| *** adam_g` is now known as adam_g | 14:37 | |
| dtantsur | tried | 14:37 |
| dtantsur | didn't work for me | 14:37 |
| *** adam_g has quit IRC | 14:37 | |
| *** adam_g has joined #openstack-ironic | 14:37 | |
| *** athomas has quit IRC | 14:38 | |
| lucasagomes | dtantsur, oh, I mean the paid service | 14:38 |
| lucasagomes | (full access I think that's how the call it) | 14:38 |
| dtantsur | lucasagomes, I was subscribed for a couple of months | 14:39 |
| dtantsur | it sucks | 14:39 |
| lucasagomes | gotcha | 14:39 |
| lucasagomes | right I will stick to spotify then :) | 14:39 |
| lucasagomes | thanks | 14:39 |
| *** openstackgerrit has joined #openstack-ironic | 14:40 | |
| *** krtaylor has joined #openstack-ironic | 14:42 | |
| *** athomas has joined #openstack-ironic | 14:42 | |
| openstackgerrit | Adam Gandelman proposed openstack/ironic: Testing forward grenade migration (Do not merge) https://review.openstack.org/124474 | 14:43 |
| adam_g | JayF, http://logs.openstack.org/21/133921/1/check/check-tempest-dsvm-ironic-agent_ssh-nv/afb2c10/ passing now | 14:44 |
| lucasagomes | woah | 14:46 |
| *** vdrok has quit IRC | 14:52 | |
| *** vdrok has joined #openstack-ironic | 14:52 | |
| *** krtaylor has quit IRC | 14:58 | |
| dtantsur | folks are already planning their midcycles. should we? | 15:03 |
| NobodyCam | I just got home. Thou we did chat about it at the summit | 15:03 |
| rloo | dtantsur: take a look at the etherpad - deva put some notes there about midcycle. | 15:04 |
| NobodyCam | I vote for palm springs CA, usa :-p | 15:04 |
| NobodyCam | morning rloo | 15:04 |
| rloo | dtantsur: bottom of https://etherpad.openstack.org/p/IronicWhiteBoard | 15:04 |
| dtantsur | yeah, I've seen it, I'm interesting in the follow-up :) | 15:04 |
| rloo | hiya NobodyCam! How was your trip/site-seeing? | 15:05 |
| rloo | dtantsur: I'm not interested. ha ha. I see that the etherpad doesn't mention possible dates, although that was discussed. | 15:05 |
| NobodyCam | it was Great I got to meet *most* all of johns family ) | 15:06 |
| rloo | dtantsur: maybe ping devananda about it. or Put it on the agenda for our meeting? or should it be an email? | 15:06 |
| rloo | NobodyCam: are they like John? :-) | 15:06 |
| NobodyCam | :) | 15:06 |
| *** anderbubble has joined #openstack-ironic | 15:07 | |
| NobodyCam | they are a really nice group of folks, lol, they sent us home with like 25 lbs of different english candy | 15:08 |
| rloo | NobodyCam: ha ha. Better make a dentist appointment! | 15:10 |
| NobodyCam | lol ...yup :-P | 15:10 |
| *** romcheg has quit IRC | 15:11 | |
| *** romcheg has joined #openstack-ironic | 15:12 | |
| *** romcheg has quit IRC | 15:12 | |
| *** ChuckC has joined #openstack-ironic | 15:12 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Make some periodic tasks run in parallel https://review.openstack.org/133976 | 15:13 |
| *** romcheg has joined #openstack-ironic | 15:16 | |
| *** romcheg has left #openstack-ironic | 15:16 | |
| *** romcheg has joined #openstack-ironic | 15:17 | |
| *** anderbubble has quit IRC | 15:17 | |
| *** romcheg has left #openstack-ironic | 15:18 | |
| *** dprince has quit IRC | 15:18 | |
| *** romcheg has joined #openstack-ironic | 15:19 | |
| *** ryanpetrello has quit IRC | 15:19 | |
| *** pensu has joined #openstack-ironic | 15:20 | |
| lucasagomes | dtantsur, NobodyCam rloo about the mid-cycle, tho I love it I think that us as a open source project should also discuss a bit | 15:23 |
| lucasagomes | how much of the face to face thing is required for an open source project | 15:24 |
| openstackgerrit | Victor Lowther proposed openstack/ironic-specs: New Ironic provisioner state machine. https://review.openstack.org/133828 | 15:24 |
| dtantsur | lucasagomes, seems like a lot :-/ | 15:24 |
| *** ryanpetrello has joined #openstack-ironic | 15:25 | |
| lucasagomes | dtantsur, yeah, that's the line I'm looking here, cause seems there's something broken in our development process if that's the case | 15:25 |
| rloo | lucasagomes: yeah. There was a discussion on the list awhile ago. I personally think 4 times/year is a lot (not that I don't like seeing you but...) However, others have said the midcycles are more useful than the summits (or something like that) | 15:25 |
| lucasagomes | rloo, I think midcycle are more productives in terms of coding | 15:25 |
| lucasagomes | it's like a hackaton | 15:26 |
| lucasagomes | but yeah I feel that 4x year, for an opensource project is a bit too much. I also worry about excluding folks that can't attend for whatever reason (visa, budget etc) | 15:26 |
| rloo | lucasagomes: is it possible to do a hackathon via eg google-whatever? Cuz not everyone at the midcycle are hacking on the same thing? | 15:26 |
| lucasagomes | rloo, yeah it would be possible I believe | 15:27 |
| lucasagomes | this midcycle we may try the tele-conference thing for those who can't attend | 15:27 |
| lucasagomes | I hope that work out well | 15:27 |
| *** penick has joined #openstack-ironic | 15:27 | |
| rloo | we'll see. I think that as long as these meet ups work for people and as long as people want to go, then fine. Just don't make it mandatory for others. | 15:28 |
| rloo | i like the idea of the meetup being in Europe this time (if there has to be a meetup) because that makes it easier for Europeans to attend. | 15:29 |
| lucasagomes | yeah sounds reasonable, but I'd like to have us to talk to see what we can improve in the development in a distributed way | 15:29 |
| rloo | lucasagomes: agreed. | 15:29 |
| devananda | something seems odd with my cell ph here ... | 15:30 |
| lucasagomes | devananda, morning there | 15:30 |
| dtantsur | devananda, o/ | 15:30 |
| devananda | lucasagomes: afternoon, actually | 15:30 |
| rloo | It's Wednesday, devananda, do you know where you are? :-) | 15:31 |
| lucasagomes | devananda, oh yeah indeed, is it 13:00ish now? | 15:31 |
| rloo | dtantsur: I have a question for you wrt the Ironic Inbox. Is there any (easy) way to list patches to Nova's ironic driver? | 15:32 |
| dtantsur | rloo, I doubt it honestly | 15:33 |
| devananda | there are patches up for the nova driver? | 15:33 |
| rloo | devananda: I have no idea. That's why it'd be good to have some way to know. | 15:33 |
| devananda | rloo: in principle, you can set a file-based watch in gerrit | 15:33 |
| ChuckC | hi ironic | 15:34 |
| rloo | devananda: hmm. thx. I'll keep that in mind. If I decide to look into it. | 15:35 |
| rloo | hi ChuckC | 15:35 |
| ChuckC | rloo: hi | 15:35 |
| ChuckC | rloo: I have a network boot question | 15:36 |
| ChuckC | rloo: do you know who I should bug? | 15:36 |
| lucasagomes | ChuckC, ask it here in the channel someone may know and answer that | 15:36 |
| rloo | ChuckC: best not to bug anyone, but if you want to, I think you can just ask and hopefully someone will jump in. | 15:36 |
| lucasagomes | and it also benefit other people reading the channel logs | 15:36 |
| ChuckC | rloo: lucasagomes: sure :) | 15:37 |
| rloo | ChuckC: don't worry about 'bugging'. | 15:37 |
| devananda | ChuckC: can you be more precise w.r.t. network boot? | 15:38 |
| ChuckC | if neutron were to support using dhcp client id, would that be helpful in allowing ironic to boot from any of a set of nics? | 15:38 |
| devananda | oh. that's _totally_ not what I thought you meant. glad I asked :) | 15:39 |
| ChuckC | rloo: I don't really worry about bugging folks ;) | 15:39 |
| ChuckC | boot a server, that is | 15:39 |
| ChuckC | What I don't know is whether the boot protocols can use dhcp client id | 15:39 |
| NobodyCam | morning devananda | 15:40 |
| devananda | ChuckC: ironic/dhcp/neutron.py L160 -- looks like Ironic will attempt to update the dhcp options for every port belongong to a node | 15:41 |
| devananda | it aborts the deploy if all attempts fail; continues if at least one port update completed | 15:41 |
| devananda | NobodyCam: g;morning | 15:41 |
| NobodyCam | :) off to go pickup the kids | 15:42 |
| devananda | ChuckC: i'm not immediately seeing how DHCP Client ID would change things. can you explain what you're thinking? | 15:42 |
| ChuckC | devananda: well, DHCP client id can be used instead of mac address to vend and IP, so multiple nics with the same dhcp client id could get the same IP address | 15:45 |
| ChuckC | devananda: which would be handy in case a nic fails, an alternate nic could be used | 15:45 |
| ChuckC | devananda: but I have no idea about boot protocols, so I don't know if the client side can be set up to use dhcp client id | 15:46 |
| devananda | ChuckC: ah, related to port bonding? | 15:46 |
| ChuckC | devananda: ya, similar concept | 15:46 |
| devananda | I'm similarly unsure about the client side -- what ever is called by cloud-init, in this case | 15:47 |
| ChuckC | devananda: that's kind of why I was hoping there was a network boot guru in your ranks | 15:48 |
| ChuckC | devananda: seems like very specific knowledge/experience is needed | 15:49 |
| devananda | yep. lifeless has dug into that stuff in the past. not sure if he wants to any more :) | 15:49 |
| *** lazy_prince is now known as killer_prince | 15:50 | |
| devananda | ChuckC: if it's a matter of Ironic informing Neutron that "this and that VIF belong to teh same host" -- I don't see, offhand, how the current API would support that | 15:50 |
| devananda | but IANANE | 15:51 |
| devananda | (i am not a neutron expert) | 15:51 |
| ChuckC | devananda: the proposal is to add dhcp_client_id to the port resource | 15:51 |
| ChuckC | devananda: right the current API does not | 15:52 |
| *** penick has quit IRC | 15:52 | |
| devananda | ChuckC: here is where Nova gets the MACs from Ironic, passes them to Neutron: http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/manager.py#n1408 | 15:53 |
| *** erwan_taf has quit IRC | 15:53 | |
| JayF | adam_g: quick, make it vote :) | 15:55 |
| *** rushiagr_away is now known as rushiagr | 15:56 | |
| ChuckC | devananda: ok, maybe I'll poke around a bit | 15:57 |
| ChuckC | devananda: do you think an ML thread might be helpful? | 15:57 |
| *** pcrews has joined #openstack-ironic | 15:58 | |
| devananda | ChuckC: there's a second path whereby ironic also updates those options here -- http://git.openstack.org/cgit/openstack/ironic/tree/ironic/dhcp/neutron.py#n137 | 15:59 |
| devananda | ChuckC: yep, definitely | 15:59 |
| ChuckC | devananda: ok, thanks! | 15:59 |
| *** david-lyle has joined #openstack-ironic | 16:03 | |
| *** erwan_taf has joined #openstack-ironic | 16:12 | |
| *** krtaylor has joined #openstack-ironic | 16:17 | |
| *** anderbubble has joined #openstack-ironic | 16:27 | |
| *** Haomeng|2 has quit IRC | 16:28 | |
| *** Haomeng|2 has joined #openstack-ironic | 16:29 | |
| *** mjturek has joined #openstack-ironic | 16:30 | |
| *** Viswanath has joined #openstack-ironic | 16:36 | |
| *** achanda has joined #openstack-ironic | 16:37 | |
| *** Viswanath has quit IRC | 16:39 | |
| *** dtantsur is now known as dtantsur|afk | 16:43 | |
| openstackgerrit | Merged openstack/ironic: Use wsme.Unset as default value for API objects https://review.openstack.org/131139 | 16:47 |
| *** shakamunyi has quit IRC | 16:53 | |
| *** rloo has quit IRC | 16:54 | |
| *** ifarkas has quit IRC | 16:54 | |
| *** achanda has quit IRC | 17:00 | |
| *** lucasagomes is now known as lucas-brb | 17:01 | |
| *** romcheg has quit IRC | 17:06 | |
| *** marcoemorais has joined #openstack-ironic | 17:14 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add put method to API to create or update a node https://review.openstack.org/130228 | 17:16 |
| *** igordcard has quit IRC | 17:28 | |
| *** igordcard has joined #openstack-ironic | 17:33 | |
| *** igordcard has quit IRC | 17:36 | |
| *** jistr has quit IRC | 17:38 | |
| *** athomas has quit IRC | 17:39 | |
| *** Nisha has joined #openstack-ironic | 17:47 | |
| *** spandhe has joined #openstack-ironic | 17:51 | |
| *** anderbubble has quit IRC | 17:52 | |
| *** shakamunyi has joined #openstack-ironic | 17:56 | |
| *** shakamunyi has quit IRC | 17:56 | |
| *** dprince has joined #openstack-ironic | 17:57 | |
| *** ndipanov is now known as ndipanov_gone | 17:58 | |
| *** derekh has quit IRC | 17:58 | |
| *** shakamunyi has joined #openstack-ironic | 18:01 | |
| *** rloo has joined #openstack-ironic | 18:03 | |
| *** sambetts has quit IRC | 18:05 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Discover node properties using new CLI node-discover-properties https://review.openstack.org/100951 | 18:09 |
| *** penick has joined #openstack-ironic | 18:10 | |
| *** andreykurilin_ has joined #openstack-ironic | 18:13 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Automate UEFI-BIOS Iso Creation https://review.openstack.org/134022 | 18:15 |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Discover node properties using new CLI node-discover-properties https://review.openstack.org/100951 | 18:27 |
| Shrews | lucas-brb: ping | 18:27 |
| *** pensu has quit IRC | 18:30 | |
| *** subscope has joined #openstack-ironic | 18:36 | |
| *** achanda has joined #openstack-ironic | 18:47 | |
| *** rloo has quit IRC | 19:03 | |
| *** rloo has joined #openstack-ironic | 19:03 | |
| *** pelix has quit IRC | 19:04 | |
| spandhe | Is there a known issue in Ironic where Ironic never finds out that the node was rebooted? | 19:06 |
| lucas-brb | Shrews, pong, sorry the delay | 19:11 |
| *** lucas-brb is now known as lucasagomes | 19:11 | |
| *** shakamunyi has quit IRC | 19:12 | |
| *** shakamunyi has joined #openstack-ironic | 19:13 | |
| Shrews | lucasagomes: no worries. i hadn't considered your TZ either. :) i left a comment in one of your reviews | 19:16 |
| lucasagomes | Shrews, ah ta much for the reviews I will take a look | 19:17 |
| Shrews | lucasagomes: i was wondering about dmitry's comment in your client change | 19:17 |
| lucasagomes | Shrews, oh yeah, I will reply there... cause that args_array_to_dict() functions is quite weird | 19:18 |
| lucasagomes | it needs a dict as input | 19:18 |
| Shrews | yeah. that is a weird func | 19:18 |
| lucasagomes | you're correct on the comment I'm not passing that fileds dict to the vendor_passthru() method I'm using it because it's required by that method | 19:18 |
| *** Marga_ has joined #openstack-ironic | 19:20 | |
| *** achanda has quit IRC | 19:30 | |
| *** achanda has joined #openstack-ironic | 19:31 | |
| *** achanda has quit IRC | 19:34 | |
| *** achanda has joined #openstack-ironic | 19:35 | |
| *** zhidong_ has joined #openstack-ironic | 19:36 | |
| *** dlaube has joined #openstack-ironic | 19:36 | |
| *** zhidong has quit IRC | 19:38 | |
| *** lucasagomes is now known as lucas-dinner | 19:38 | |
| *** marcoemorais has quit IRC | 19:44 | |
| *** marcoemorais has joined #openstack-ironic | 19:44 | |
| *** Marga_ has quit IRC | 19:53 | |
| *** igordcard has joined #openstack-ironic | 19:54 | |
| *** achanda has quit IRC | 19:56 | |
| *** foexle has quit IRC | 19:59 | |
| *** subscope has quit IRC | 20:01 | |
| *** rushiagr is now known as rushiagr_away | 20:01 | |
| *** tatyana has joined #openstack-ironic | 20:02 | |
| *** ChuckC has quit IRC | 20:11 | |
| *** jcoufal has quit IRC | 20:15 | |
| *** andreykurilin__ has joined #openstack-ironic | 20:17 | |
| *** andreykurilin_ has quit IRC | 20:17 | |
| rloo | devananda, NobodyCam, Shrews, JayF and anyone else: I did a first stab at release notes for Juno. Please contribute, etc. I'm not quite sure what one puts in release notes. | 20:18 |
| rloo | oh yeah: https://wiki.openstack.org/wiki/Ironic/ReleaseNotes/Juno | 20:18 |
| Shrews | rloo: LOL at the upgrade notes | 20:19 |
| *** mjturek has quit IRC | 20:20 | |
| rloo | Shrews: you noticed! ;) | 20:20 |
| rloo | Shrews: makes me wonder if we should have an etherpad to note down upgrade stuff and deprecated stuff. Or is the DocImpact flag supposed to capture all that. | 20:21 |
| Shrews | rloo: the specs template has a section for that | 20:24 |
| Shrews | though i guess that won't cover a bug fix that may impact upgrading | 20:25 |
| rloo | Shrews: right, but is someone going to go through all the specs, or bugs? ie, I didn't even want to look at the bugs/specs for Juno. | 20:25 |
| Shrews | rloo: an operator won't, surely | 20:25 |
| rloo | Shrews: so I suspect I won't want to go through the specs/bugs for Kilo. Maybe someone will volunteer ;) | 20:26 |
| *** andreykurilin_ has joined #openstack-ironic | 20:26 | |
| *** andreykurilin__ has quit IRC | 20:26 | |
| *** krtaylor has quit IRC | 20:31 | |
| *** Marga_ has joined #openstack-ironic | 20:34 | |
| *** Marga_ has quit IRC | 20:37 | |
| *** Marga_ has joined #openstack-ironic | 20:37 | |
| *** ChuckC has joined #openstack-ironic | 20:38 | |
| *** anderbubble has joined #openstack-ironic | 20:38 | |
| *** achanda has joined #openstack-ironic | 20:39 | |
| *** achanda has quit IRC | 20:39 | |
| *** krtaylor has joined #openstack-ironic | 20:43 | |
| mrda | Morning Ironic | 20:43 |
| *** achanda has joined #openstack-ironic | 20:43 | |
| rloo | morning mrda! | 20:44 |
| Shrews | Morning mrda | 20:44 |
| mrda | \o | 20:49 |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Discover node properties for iLO drivers https://review.openstack.org/103007 | 20:50 |
| *** tatyana has quit IRC | 20:51 | |
| *** tatyana has joined #openstack-ironic | 20:51 | |
| *** Marga_ has quit IRC | 20:53 | |
| *** Nisha has quit IRC | 20:53 | |
| *** mjturek has joined #openstack-ironic | 20:55 | |
| *** dprince has quit IRC | 20:56 | |
| *** mjturek has quit IRC | 20:56 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-specs: Support Decommissioning of Nodes https://review.openstack.org/102685 | 20:57 |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-specs: Support Zapping of Nodes https://review.openstack.org/102685 | 20:57 |
| JoshNang | whoops, forgot to update the commit message | 20:57 |
| JoshNang | <3 "zapping" | 20:58 |
| *** Marga_ has joined #openstack-ironic | 21:10 | |
| *** shakamun_ has joined #openstack-ironic | 21:10 | |
| *** shakamunyi has quit IRC | 21:10 | |
| *** killer_prince has quit IRC | 21:14 | |
| *** eglute has joined #openstack-ironic | 21:16 | |
| *** shakamun_ has quit IRC | 21:26 | |
| *** killer_prince has joined #openstack-ironic | 21:28 | |
| *** killer_prince is now known as lazy_prince | 21:28 | |
| *** shakamunyi has joined #openstack-ironic | 21:30 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-specs: Support Zapping of Nodes https://review.openstack.org/102685 | 21:30 |
| *** achanda has quit IRC | 21:32 | |
| *** Viswanath has joined #openstack-ironic | 21:36 | |
| *** Viswanath has quit IRC | 21:40 | |
| *** Viswanath has joined #openstack-ironic | 21:41 | |
| *** Viswanath has quit IRC | 21:44 | |
| *** anderbubble has quit IRC | 21:57 | |
| *** achanda has joined #openstack-ironic | 22:00 | |
| *** anderbubble has joined #openstack-ironic | 22:02 | |
| *** achanda has quit IRC | 22:06 | |
| *** achanda has joined #openstack-ironic | 22:14 | |
| *** tatyana has quit IRC | 22:27 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add sync and async support for passthru methods https://review.openstack.org/129662 | 22:29 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Make vendor methods discoverable via the Ironic API https://review.openstack.org/133533 | 22:29 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Vendor endpoints to support different HTTP methods https://review.openstack.org/129942 | 22:29 |
| *** andreykurilin_ has quit IRC | 22:34 | |
| *** Marga_ has quit IRC | 22:38 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-specs: Adjust the extended-vendor-passthru spec https://review.openstack.org/134066 | 22:38 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-specs: Adjust the extended-vendor-passthru spec https://review.openstack.org/134066 | 22:40 |
| *** Marga_ has joined #openstack-ironic | 22:58 | |
| *** enterprisedc has joined #openstack-ironic | 22:59 | |
| *** marcoemorais1 has joined #openstack-ironic | 23:05 | |
| *** lucas-dinner has quit IRC | 23:06 | |
| *** marcoemorais has quit IRC | 23:09 | |
| *** david-lyle is now known as david-lyle_afk | 23:12 | |
| *** ChuckC has quit IRC | 23:16 | |
| *** igordcard has quit IRC | 23:23 | |
| *** marcoemorais1 has quit IRC | 23:30 | |
| *** marcoemorais has joined #openstack-ironic | 23:31 | |
| *** anderbubble has quit IRC | 23:32 | |
| *** ryanpetrello has quit IRC | 23:39 | |
| *** ChuckC has joined #openstack-ironic | 23:40 | |
| *** shakamunyi has quit IRC | 23:58 | |
| *** shakamunyi has joined #openstack-ironic | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!