eglute | I am new to Ironic, and am going through all the documentation. I found some blueprints for things that are proposed, but I cannot find anything how the current provisioner state machine works. Is there one currently? Also, is there currently anything for discovering nodes? | 00:01 |
---|---|---|
*** naohirot has joined #openstack-ironic | 00:01 | |
*** shakamunyi has quit IRC | 00:01 | |
*** marcoemorais1 has joined #openstack-ironic | 00:02 | |
*** shakamunyi has joined #openstack-ironic | 00:03 | |
*** eglute_ has joined #openstack-ironic | 00:05 | |
*** marcoemorais has quit IRC | 00:06 | |
*** eglute_ has quit IRC | 00:07 | |
*** david-lyle_afk is now known as david-lyle | 00:10 | |
*** Viswanath has joined #openstack-ironic | 00:17 | |
Haomeng|2 | eglute: hi | 00:19 |
Haomeng|2 | eglute: welcome | 00:20 |
*** Viswanath has quit IRC | 00:20 | |
Haomeng|2 | eglute: this is our wiki - https://wiki.openstack.org/wiki/Ironic | 00:21 |
Haomeng|2 | eglute: this is the guide to say how our ironic works - http://docs.openstack.org/developer/ironic/deploy/user-guide.html | 00:21 |
Haomeng|2 | eglute: hope this help you, feel free to ask our question here, you are welcome:) | 00:22 |
eglute | Thanks! I did look at the wiki and the docs, but I did not find the answers I was looking for! | 00:22 |
Haomeng|2 | eglute: you want to know about discovering nodes? | 00:23 |
eglute | yes! | 00:23 |
Haomeng|2 | eglute: let me check, we have some bps about discovering | 00:23 |
eglute | that would be great, thank you | 00:24 |
Haomeng|2 | eglute: try to find this bp list - https://blueprints.launchpad.net/ironic , and search discovery | 00:24 |
JoshNang | eglute: i haven't seen a state transistion diagram for how it works right now. and discovery is in progress for kilo, but not in yet | 00:24 |
eglute | ok... So currently no discovery? | 00:24 |
JoshNang | correct | 00:25 |
*** rainya has quit IRC | 00:25 | |
JoshNang | discoverd is a ramdisk for doing discovery. i'll see if i can find a link | 00:25 |
Haomeng|2 | eglute: yes, we have some discovery bp which are in progress in the bp list | 00:25 |
JoshNang | eglute: https://github.com/Divius/ironic-discoverd we're hopefully more tightly integrating this | 00:26 |
Haomeng|2 | JoshNang: looks like it is under tripleo, right? | 00:27 |
*** rainya has joined #openstack-ironic | 00:27 | |
eglute | JoshNang: thank you! I will take a look at it | 00:28 |
JoshNang | Haomeng|2: yup | 00:28 |
Haomeng|2 | and it just discovery the node properties, not discovering the nodes I think, right | 00:28 |
Haomeng|2 | not sure if we have out-of-band nodes auto-discover bp | 00:28 |
JoshNang | right :) | 00:29 |
Haomeng|2 | JoshNang: ok | 00:29 |
JoshNang | eglute: any specific questions on the state diagram? otherwise I can probably gist a basic flow | 00:29 |
eglute | I am just trying to understand current state of things, and how functional ironic currently is | 00:29 |
JoshNang | it depends on your use case :) | 00:30 |
*** Marga_ has quit IRC | 00:30 | |
eglute | looks like right now, all nodes have to be manually added | 00:30 |
JoshNang | correct | 00:31 |
Haomeng|2 | eglute: yes | 00:31 |
*** Marga_ has joined #openstack-ironic | 00:31 | |
JoshNang | in our env, we wrote a script to translate a spreadsheet from our server oem to commands to the ironic api to create nodes | 00:32 |
Haomeng|2 | JoshNang: cool | 00:32 |
*** penick has quit IRC | 00:33 | |
Haomeng|2 | JoshNang: it is easy to find ipmi ip and mac, but it is diffcult to find other non-ipmi nic's macs I think | 00:34 |
Haomeng|2 | we need both ipmi ip and non-ipmi nic mac as input for node creating | 00:35 |
eglute | JoshNang: good to know that is an option. I am, however, glad to see that there are blueprints for other ways to discover things | 00:35 |
JoshNang | Haomeng|2: we got those from our oem (luckily). it looks like discoverd will create non-ipmi ports | 00:36 |
Haomeng|2 | JoshNang: cool | 00:36 |
JoshNang | eglute: agreed! | 00:36 |
Haomeng|2 | I understand, we have to do in-band discovery to detect the non-ipmi nic macs | 00:36 |
eglute | have anyone run into issues with scale? | 00:37 |
Haomeng|2 | another way, we workthrough the switch snmp to find the macs, assuming all the ports are connecting to our server | 00:37 |
JoshNang | Haomeng|2: i think that's possible too. it sounds like that's what the compass people want to do | 00:38 |
JoshNang | eglute: yes! (i work on rackspace onmetal) | 00:38 |
Haomeng|2 | JoshNang: yes | 00:38 |
Haomeng|2 | JoshNang: xcat has such function already | 00:38 |
eglute | JoshNang: i will email you then! | 00:39 |
Haomeng|2 | JoshNang: we can take as reference | 00:39 |
JoshNang | eglute: one of our other devs (jroll) did a talk on the issues we ran into at the summit, but i can't seem to find the video http://openstacksummitnovember2014paris.sched.org/venue/Room+242AB#.VGP9y1nF-5I | 00:40 |
JoshNang | Haomeng|2: ahh, awesome | 00:40 |
eglute | watching this one at teh moment: https://www.openstack.org/summit/openstack-paris-summit-2014/session-videos/presentation/scaling-ironic | 00:40 |
ChuckC | hi ironic, I'm sending a general plea for help: please take a look at http://lists.openstack.org/pipermail/openstack-dev/2014-November/050329.html | 00:41 |
JoshNang | eglute: that's the one | 00:41 |
ChuckC | especially if you have network boot experience/knowledge... many thanks!! | 00:41 |
Haomeng|2 | ChuckC: ok, welcome | 00:42 |
Haomeng|2 | fyi the reference of xcat discovery - http://sumavi.com/sections/understanding-autodiscovery | 00:44 |
Haomeng|2 | ChuckC: thanks for your mail | 00:45 |
JoshNang | Haomeng|2: awesome, thanks | 00:45 |
*** lazy_prince has quit IRC | 00:52 | |
*** achanda has quit IRC | 00:53 | |
*** achanda has joined #openstack-ironic | 01:00 | |
*** Marga_ has quit IRC | 01:01 | |
*** penick has joined #openstack-ironic | 01:03 | |
*** penick has quit IRC | 01:08 | |
*** shakamunyi has quit IRC | 01:14 | |
*** chenglch has joined #openstack-ironic | 01:23 | |
*** marcoemorais1 has quit IRC | 01:30 | |
*** rloo has quit IRC | 01:34 | |
openstackgerrit | jiangfei proposed openstack/ironic: Enable hacking rule E265 https://review.openstack.org/134098 | 01:50 |
*** ChuckC has quit IRC | 01:53 | |
*** ChuckC has joined #openstack-ironic | 02:04 | |
openstackgerrit | jiangfei proposed openstack/ironic: Enable hacking rule E265 https://review.openstack.org/133921 | 02:05 |
*** nosnos has joined #openstack-ironic | 02:06 | |
*** david-lyle is now known as david-lyle_afk | 02:08 | |
*** achanda has quit IRC | 02:17 | |
openstackgerrit | jiangfei proposed openstack/ironic: Enable hacking rule E265 https://review.openstack.org/133921 | 02:17 |
*** rakesh_hs has joined #openstack-ironic | 02:23 | |
*** jrist has quit IRC | 02:24 | |
*** jrist has joined #openstack-ironic | 02:41 | |
*** jrist has quit IRC | 02:41 | |
*** dlaube has quit IRC | 02:42 | |
*** jrist has joined #openstack-ironic | 02:47 | |
*** killer_prince has joined #openstack-ironic | 02:48 | |
*** killer_prince is now known as lazy_prince | 02:48 | |
openstackgerrit | jiangfei proposed openstack/ironic: Enable hacking rule E265 https://review.openstack.org/133921 | 02:49 |
*** spandhe has quit IRC | 02:51 | |
*** jrist has quit IRC | 02:53 | |
*** lazy_prince has quit IRC | 02:55 | |
*** lazy_prince has joined #openstack-ironic | 02:55 | |
*** nosnos has quit IRC | 03:25 | |
*** naohirot has quit IRC | 03:29 | |
*** rakesh_hs has quit IRC | 03:31 | |
*** lazy_pri- has joined #openstack-ironic | 03:43 | |
*** lazy_prince has quit IRC | 03:44 | |
*** lazy_pri- is now known as lazy_prince | 03:44 | |
openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Automate UEFI-BIOS Iso Creation https://review.openstack.org/134022 | 03:51 |
*** pensu has joined #openstack-ironic | 03:52 | |
*** pensu has quit IRC | 03:56 | |
*** pensu has joined #openstack-ironic | 03:57 | |
*** nosnos has joined #openstack-ironic | 03:57 | |
*** naohirot has joined #openstack-ironic | 03:59 | |
*** Ng has quit IRC | 04:06 | |
*** Ng has joined #openstack-ironic | 04:06 | |
*** cppforlife_ has joined #openstack-ironic | 04:09 | |
*** slagle_ has joined #openstack-ironic | 04:09 | |
*** Isotopp_ has joined #openstack-ironic | 04:11 | |
*** mordred_ has joined #openstack-ironic | 04:12 | |
*** colettecello has joined #openstack-ironic | 04:12 | |
*** JayF has quit IRC | 04:12 | |
*** cohn has quit IRC | 04:12 | |
*** EmilienM has quit IRC | 04:12 | |
*** slagle has quit IRC | 04:12 | |
*** cppforlife has quit IRC | 04:12 | |
*** keekz has quit IRC | 04:12 | |
*** rainya has quit IRC | 04:12 | |
*** chenglch has quit IRC | 04:12 | |
*** mitz_ has quit IRC | 04:12 | |
*** mordred has quit IRC | 04:12 | |
*** Isotopp has quit IRC | 04:12 | |
*** trown has quit IRC | 04:12 | |
*** gothicmindfood has quit IRC | 04:12 | |
*** mitz has quit IRC | 04:12 | |
*** mordred_ is now known as mordred | 04:12 | |
*** mitz has joined #openstack-ironic | 04:12 | |
*** trown has joined #openstack-ironic | 04:12 | |
*** subscope has joined #openstack-ironic | 04:13 | |
*** EmilienM has joined #openstack-ironic | 04:14 | |
*** keekz has joined #openstack-ironic | 04:14 | |
*** pcrews has quit IRC | 04:15 | |
*** rainya has joined #openstack-ironic | 04:15 | |
*** cppforlife_ is now known as cppforlife | 04:16 | |
*** pcrews has joined #openstack-ironic | 04:19 | |
*** JayF has joined #openstack-ironic | 04:20 | |
*** pradipta_away is now known as pradipta | 04:20 | |
*** achanda has joined #openstack-ironic | 04:24 | |
*** pcrews has quit IRC | 04:25 | |
*** rushiagr_away is now known as rushiagr | 04:27 | |
*** lazy_prince is now known as killer_prince | 04:27 | |
*** Haomeng has joined #openstack-ironic | 04:28 | |
*** achanda has quit IRC | 04:29 | |
*** Haomeng|2 has quit IRC | 04:29 | |
*** cohn has joined #openstack-ironic | 04:29 | |
*** cohn is now known as Guest70410 | 04:30 | |
*** rameshg87 has joined #openstack-ironic | 04:32 | |
*** achanda has joined #openstack-ironic | 04:32 | |
*** Guest70410 is now known as cohn | 04:34 | |
*** cohn has joined #openstack-ironic | 04:34 | |
*** pensu has quit IRC | 04:43 | |
*** pensu has joined #openstack-ironic | 04:43 | |
*** eglute has quit IRC | 04:47 | |
*** jrist has joined #openstack-ironic | 04:54 | |
*** jrist has quit IRC | 05:00 | |
*** jrist has joined #openstack-ironic | 05:11 | |
*** rameshg87_ has joined #openstack-ironic | 05:11 | |
*** rameshg87_ has quit IRC | 05:13 | |
*** rameshg87__ has joined #openstack-ironic | 05:13 | |
*** rameshg87 has quit IRC | 05:14 | |
*** marcoemorais has joined #openstack-ironic | 05:15 | |
*** rameshg87__ has quit IRC | 05:16 | |
*** rameshg87_ has joined #openstack-ironic | 05:16 | |
*** vinbs has joined #openstack-ironic | 05:16 | |
*** marcoemorais1 has joined #openstack-ironic | 05:17 | |
*** marcoemorais has quit IRC | 05:20 | |
*** rameshg87_ has quit IRC | 05:22 | |
*** rameshg87 has joined #openstack-ironic | 05:22 | |
*** killer_prince has quit IRC | 05:23 | |
*** rameshg87 has quit IRC | 05:30 | |
*** rameshg87 has joined #openstack-ironic | 05:30 | |
*** subscope has quit IRC | 05:33 | |
*** rameshg87 has quit IRC | 05:35 | |
*** rameshg87__ has joined #openstack-ironic | 05:35 | |
*** chenglch has joined #openstack-ironic | 05:39 | |
*** achanda has quit IRC | 06:01 | |
*** achanda has joined #openstack-ironic | 06:04 | |
*** k4n0 has joined #openstack-ironic | 06:09 | |
*** teju has joined #openstack-ironic | 06:19 | |
*** lazy_prince has joined #openstack-ironic | 06:27 | |
*** mrda is now known as mrda_away | 06:29 | |
*** naohirot has quit IRC | 06:42 | |
*** ChuckC has quit IRC | 06:45 | |
*** kongfy has joined #openstack-ironic | 07:04 | |
*** kongfy has quit IRC | 07:06 | |
*** rameshg87__ has quit IRC | 07:06 | |
*** achanda has quit IRC | 07:07 | |
*** ChuckC has joined #openstack-ironic | 07:18 | |
*** rameshg87 has joined #openstack-ironic | 07:20 | |
*** zhidong_ has quit IRC | 07:21 | |
*** naohirot has joined #openstack-ironic | 07:27 | |
lazy_prince | any idea if https://review.openstack.org/#/c/87825/ is implemented..? | 07:37 |
*** harlowja has joined #openstack-ironic | 07:41 | |
erwan_taf | hi there | 07:43 |
*** harlowja has quit IRC | 07:53 | |
*** erwan_taf has quit IRC | 07:59 | |
*** marcoemorais1 has quit IRC | 08:02 | |
*** harlowja has joined #openstack-ironic | 08:06 | |
*** achanda has joined #openstack-ironic | 08:07 | |
*** achanda has quit IRC | 08:12 | |
*** foexle has joined #openstack-ironic | 08:13 | |
*** rakesh_hs has joined #openstack-ironic | 08:20 | |
*** Nisha has joined #openstack-ironic | 08:23 | |
*** harlowja has quit IRC | 08:23 | |
Haomeng | lazy_prince: hi | 08:28 |
Haomeng | lazy_prince: i think it is neutron patch, right? | 08:29 |
lazy_prince | Haomeng: its a spec for netron and ironic integration.. | 08:33 |
Haomeng | lazy_prince: yes, i see it is merged already | 08:33 |
Haomeng | lazy_prince: you want to check the code? | 08:33 |
lazy_prince | yup.. so is someone working on this spec..? | 08:34 |
lazy_prince | i would love to try it out.. | 08:34 |
Haomeng | lazy_prince: yes, it is implemented, I think, - https://blueprints.launchpad.net/neutron/+spec/neutron-external-attachment-points | 08:35 |
Haomeng | code is in review, - https://review.openstack.org/#/c/116187/ | 08:35 |
lazy_prince | also i would like to understand how this will actually work in principle.. if there is a ppt/write up elaborating the same, pointers to that will be helpful.. | 08:36 |
Haomeng | lazy_prince: the code is not merged, so have to check with neutron guys, he should be neutron core | 08:37 |
lazy_prince | aha.. so if my understanding is correct, if this merges, then there wont be any need for ironic-neutron-plugin (from rackspace).. right..? | 08:38 |
*** dtantsur|afk is now known as dtantsur | 08:38 | |
dtantsur | Morning Ironic | 08:39 |
Haomeng | lazy_prince: yes, I think so, but not sure with rackspace | 08:40 |
Haomeng | lazy_prince: :) | 08:40 |
*** ifarkas has joined #openstack-ironic | 08:40 | |
lazy_prince | k.. i will bounce this in neutron room then to get more clarity.. | 08:40 |
Haomeng | lazy_prince: ok:) | 08:41 |
lazy_prince | Haomeng: do you know if there is a pointer anywhere on how port create happens for a baremetal in neutron world..? | 08:41 |
Haomeng | lazy_prince: I understand neutron will manage the port, such as ip address | 08:43 |
Haomeng | lazy_prince: so it is controled by neutron | 08:44 |
lazy_prince | i understand that for a general nove boot call, a virtual port is created in neutron by nova and then scheduler schedules to a compute node.. | 08:45 |
lazy_prince | and then the compute node generates a MAC id and then the quantum agent come into picture.. | 08:46 |
*** jcoufal has joined #openstack-ironic | 08:46 | |
lazy_prince | or something similar... how does it work for baremetal.. the sequences is what i looking for.. | 08:46 |
rameshg87 | lazy_prince, i think it works this way | 08:47 |
rameshg87 | lazy_prince, ironic has its port enrolled and nova ironic virt driver queries it and creates neutron ports for them | 08:48 |
rameshg87 | lazy_prince, nova ironic virt driver adds the neutron port id as 'vif_port_id' in the ironic port | 08:49 |
rameshg87 | lazy_prince, ironic goes and updates the pxe/tftp boot parameters into the neutron port | 08:49 |
rameshg87 | lazy_prince, one way that i have been testing is to connect the baremetal provisioning network on the neutron node and connect it to ovs integration bridge | 08:50 |
rameshg87 | lazy_prince, it's documented http://docs.openstack.org/developer/ironic/deploy/install-guide.html#configure-neutron-to-communicate-with-the-bare-metal-server | 08:51 |
lazy_prince | rameshg87, okay.. so when you say ironic has its port enrolled, i do not understand as by then the scheduler has not yet identified what node needs to be provisioned.. | 08:54 |
rameshg87 | lazy_prince, yes, ironic has a separate place in db for enrolling nodes | 08:55 |
rameshg87 | lazy_prince, neutron ports are created only after the scheduler picks up the node and passes it down to nova compute ironic virt driver | 08:56 |
rameshg87 | lazy_prince, was that your question ? | 08:56 |
*** foexle_ has joined #openstack-ironic | 08:57 | |
lazy_prince | rameshg87: well i want to understand the virtual port creation in neutron for baremetal and then how it is updated with the correct port info.. entire sequence.. I am not worried for the configuration part.. i know the steps for configuration.. i just need to understand the sequence in which the the port is created and configured.. | 08:58 |
*** foexle has quit IRC | 08:59 | |
Haomeng | lazy_prince: yes, neutron dhcp port will be configured by the neutron port update api call from ironic | 09:02 |
openstackgerrit | Tan Lin proposed openstack/ironic: Remove unused variable in agent.lookup() https://review.openstack.org/134139 | 09:08 |
*** chenglch has quit IRC | 09:10 | |
*** romcheg has joined #openstack-ironic | 09:16 | |
*** ndipanov_gone is now known as ndipanov | 09:18 | |
openstackgerrit | Ramakrishnan G proposed openstack/ironic: Update iLO driver documentation https://review.openstack.org/134144 | 09:19 |
*** derekh has joined #openstack-ironic | 09:20 | |
*** igordcard has joined #openstack-ironic | 09:29 | |
*** jistr has joined #openstack-ironic | 09:34 | |
*** tteggel has quit IRC | 09:35 | |
*** tteggel has joined #openstack-ironic | 09:35 | |
openstackgerrit | Lucas Alvares Gomes proposed openstack/python-ironicclient: Sane parameters for node and driver vendor_passthru() https://review.openstack.org/133780 | 09:35 |
*** tteggel has quit IRC | 09:39 | |
*** MattMan has quit IRC | 09:40 | |
*** tteggel has joined #openstack-ironic | 09:40 | |
*** MattMan has joined #openstack-ironic | 09:40 | |
*** lucasagomes has joined #openstack-ironic | 09:45 | |
rameshg87 | dtantsur, hi | 09:45 |
dtantsur | rameshg87, hi | 09:45 |
rameshg87 | dtantsur, are you planning to add whole-disk vs partition-disk as a driver capability in your spec ? | 09:46 |
*** rakesh_hs2 has joined #openstack-ironic | 09:46 | |
dtantsur | rameshg87, I'm not really going into details there, just want to create a basic mechanism | 09:46 |
rameshg87 | dtantsur, ah okay | 09:46 |
*** dtantsur is now known as dtantsur|brb | 09:47 | |
*** rakesh_hs has quit IRC | 09:47 | |
rameshg87 | dtantsur|brb, so you will just create an api and the code to query and give all the capabilities. correct ? | 09:47 |
dtantsur|brb | yes | 09:47 |
rameshg87 | dtantsur|brb, okay | 09:48 |
*** athomas has joined #openstack-ironic | 09:49 | |
*** viktors has joined #openstack-ironic | 09:49 | |
*** rameshg87 has quit IRC | 10:01 | |
*** rameshg87_ has joined #openstack-ironic | 10:01 | |
*** rameshg87_ has quit IRC | 10:02 | |
*** rameshg87__ has joined #openstack-ironic | 10:02 | |
*** rameshg87__ has quit IRC | 10:13 | |
*** rameshg87 has joined #openstack-ironic | 10:13 | |
*** rameshg87_ has joined #openstack-ironic | 10:14 | |
*** rameshg87 has quit IRC | 10:17 | |
*** vinbs has quit IRC | 10:27 | |
*** pelix has joined #openstack-ironic | 10:27 | |
openstackgerrit | Merged openstack/ironic: Add sync and async support for passthru methods https://review.openstack.org/129662 | 10:28 |
*** lazy_prince has quit IRC | 10:32 | |
*** lazy_prince has joined #openstack-ironic | 10:33 | |
*** Nisha has quit IRC | 10:33 | |
*** teju has left #openstack-ironic | 10:37 | |
*** naohirot has quit IRC | 10:40 | |
openstackgerrit | Merged openstack/ironic: Enable hacking rule E265 https://review.openstack.org/133921 | 10:44 |
*** lazy_prince has quit IRC | 11:07 | |
openstackgerrit | Tan Lin proposed openstack/ironic: Remove unused variable in agent._get_interfaces() https://review.openstack.org/134139 | 11:28 |
*** nosnos has quit IRC | 11:34 | |
*** nosnos has joined #openstack-ironic | 11:34 | |
*** pensu has quit IRC | 11:39 | |
*** nosnos has quit IRC | 11:39 | |
*** rameshg87_ has quit IRC | 11:47 | |
*** naohirot has joined #openstack-ironic | 11:53 | |
*** pradipta is now known as pradipta_away | 12:00 | |
openstackgerrit | Tan Lin proposed openstack/ironic: Remove unused variable in agent._get_interfaces() https://review.openstack.org/134139 | 12:07 |
*** lucasagomes is now known as lucas-hungry | 12:15 | |
*** dprince has joined #openstack-ironic | 12:38 | |
openstackgerrit | Tan Lin proposed openstack/ironic: Remove unused variable in agent._get_interfaces() https://review.openstack.org/134139 | 12:39 |
openstackgerrit | Tan Lin proposed openstack/ironic: Add driver_validate() https://review.openstack.org/134184 | 12:39 |
*** lucas-hungry is now known as lucasagomes | 12:43 | |
*** dtantsur|brb is now known as dtantsur | 12:55 | |
*** dlpartain has joined #openstack-ironic | 12:58 | |
*** sambetts has joined #openstack-ironic | 13:12 | |
*** Marga_ has joined #openstack-ironic | 13:14 | |
*** pensu has joined #openstack-ironic | 13:22 | |
*** dlpartain has quit IRC | 13:43 | |
*** Marga_ has quit IRC | 13:44 | |
*** Isotopp_ is now known as Isotopp | 13:44 | |
*** Marga_ has joined #openstack-ironic | 13:45 | |
*** ryanpetrello has joined #openstack-ironic | 13:49 | |
*** rushiagr is now known as rushiagr_away | 13:51 | |
*** jistr has quit IRC | 13:53 | |
*** ryanpetrello has quit IRC | 13:54 | |
*** jistr has joined #openstack-ironic | 13:54 | |
*** rloo has joined #openstack-ironic | 13:55 | |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Make some periodic tasks run in parallel https://review.openstack.org/133976 | 13:56 |
openstackgerrit | Naohiro Tamura proposed openstack/ironic: dev-quickstart.rst update to add required packages https://review.openstack.org/134216 | 13:58 |
naohirot | dtantsur: Hello | 14:00 |
*** rakesh_hs2 has quit IRC | 14:00 | |
dtantsur | naohirot, hi | 14:00 |
naohirot | dtantsur: I just have submitted a very very trivial bug and committed a fix to experience the whole process once. | 14:01 |
dtantsur | good! | 14:02 |
naohirot | dtantsur: so what to do next? should I assign a reviewer by myself? | 14:04 |
dtantsur | naohirot, no, you should just wait | 14:04 |
naohirot | dtantsur: or should I just wait? | 14:04 |
naohirot | dtantsur: Okay | 14:04 |
dtantsur | naohirot, it may take a while, if the team is busy (it took a couple of minutes this time :) | 14:04 |
dtantsur | naohirot, if you review stays for days already, you may come here and ask for review (do not use mail list for it please) | 14:05 |
naohirot | dtantsur: Yes, I know :-) | 14:05 |
dtantsur | naohirot, but generally the team has dashboards with all the patches requiring attention, so if nobody is reviewing your patch, we're probably too busy right now | 14:05 |
naohirot | dtantsur: Yes, absolutely here is comfortable for me than mailing list everybody look at. :-) | 14:06 |
dtantsur | yeah :) | 14:06 |
dtantsur | naohirot, also some people (including myself) receive emails about all the new review requests | 14:07 |
naohirot | dtantsur: I really get tired to set up all tools. there are a lot. :-) | 14:08 |
dtantsur | oh I hear you, it took me weeks to bootstrap when I started | 14:08 |
naohirot | dtantsur: I see. | 14:08 |
rloo | hello Ironickers :) | 14:08 |
dtantsur | that's the price to pay for keeping such a big and lively project on a good quality level | 14:09 |
dtantsur | rloo, o/ | 14:09 |
naohirot | rloo: Hi | 14:09 |
rloo | hi dtantsur, I just read your email. What/why aren't you fond of the weekly meeting format? | 14:09 |
rloo | hi naohirot | 14:09 |
rloo | dtantsur: the 'proposed' weekly meeting format? | 14:09 |
dtantsur | rloo, I'm not found of any complication of the process :) | 14:10 |
*** k4n0 has quit IRC | 14:10 | |
rloo | dtantsur: what process are you referring to, that will get complicated? | 14:10 |
dtantsur | e.g. for the reason naohirot stated above: the more processes we have, the harder it is for new people. That's mostly about having more formal structure of the meeting and having topics set in advance | 14:11 |
naohirot | dtantsur: rloo: I go to bed, little tired :) see you tomorrow! | 14:12 |
rloo | dtantsur: so the 'formal' structure of the meeting. There is already a structure to the meeting, whether you consider it formal or not. The proposal is to change that format. | 14:12 |
dtantsur | naohirot, see you | 14:12 |
rloo | dtantsur: wrt topics set in advance. yeah, that could be a problem | 14:12 |
rloo | night NobodyCam | 14:12 |
naohirot | bye! | 14:12 |
rloo | oops, night naohirot | 14:13 |
*** naohirot has quit IRC | 14:13 | |
dtantsur | lol | 14:13 |
dtantsur | rloo, I do hope NobodyCam is sleeping well right now :D | 14:13 |
rloo | dtantsur: yeah, we were once a small group of folks where only j* mattered ;) | 14:13 |
dtantsur | :) | 14:13 |
rloo | dtantsur: ha ha. | 14:13 |
dtantsur | well, our meeting does not work well right now, so we do need to change | 14:14 |
rloo | dtantsur: maybe what is missing in that email is the rationale for the 48-hour cut off. Being that if the discussion of a topic is going to be worthwhile, people may need to spend time beforehand, looking/thinking about the issue. | 14:14 |
dtantsur | rloo, yeah, it make a lot of sense, if we really start doing it ;) | 14:15 |
rloo | dtantsur: well, we can only try to make things better, and see what happens, and iterate :-) | 14:16 |
dtantsur | rloo, agile! AGILE!! \o/ | 14:16 |
dtantsur | :D | 14:16 |
* dtantsur is a certified scrum master, though never actually did it IRL | 14:17 | |
rloo | dtantsur: slow and steady is my motto ;) | 14:17 |
*** jjohnson2_ has joined #openstack-ironic | 14:17 | |
*** Marga_ has quit IRC | 14:17 | |
lucasagomes | rloo, morning :) | 14:22 |
lucasagomes | dtantsur, yup yeah, it would require us to actually take a look at the agenda 2 days before the discussions | 14:23 |
lucasagomes | maybe we can drop that freeze, that's why I put on the mail list, so people can comment about restructuring the meeting | 14:23 |
rloo | hi lucasagomes | 14:23 |
dtantsur | lucasagomes, the hardest thing will be to get used to it | 14:23 |
lucasagomes | it's even a good place to us to reflect whether we need a meeting or not | 14:24 |
lucasagomes | as a distributed team, maybe only having discussions on the mail list is enough | 14:24 |
dtantsur | oh that's going to extremes :D | 14:24 |
lucasagomes | it's a cultural thing that we may want to discuss about | 14:24 |
dtantsur | lucasagomes, 2 problems with ML: easy to ignore, long to receive response | 14:24 |
rloo | I think that discussions on the mail list are probably the 'fairest' way to have people contribute | 14:25 |
lucasagomes | dtantsur, yeah, I also like the idea of the meeting, that's why I think we should try make the process better | 14:25 |
rloo | dtantsur: if you're involved in ironic, you shouldn't ignore the emails to ironic ;) | 14:25 |
lucasagomes | rloo, yeah, cause of the different TZ's | 14:25 |
dtantsur | rloo, they're easy to miss though | 14:25 |
lucasagomes | I understand the weekly meeting is something openstackish | 14:25 |
dtantsur | lucasagomes, I don't think so | 14:26 |
rloo | dtantsur: why easy to miss? I filter the ironic ones into a separate folder ;) | 14:26 |
lucasagomes | dtantsur, I don't know any other opensource project that requires people to be online at the same time to discuss stuff | 14:26 |
lucasagomes | every week | 14:26 |
dtantsur | I know very few project that are so large and diverse | 14:27 |
lucasagomes | right, and they do have a weekly meeting? | 14:27 |
lucasagomes | (just to make it clear, I'm not against the weekly meeting, I just want to have a discussion about it) | 14:28 |
dtantsur | I don't know them to this details. we can research. | 14:28 |
lucasagomes | sure yeah | 14:28 |
rloo | dtantsur, lucasagomes: we can research, and assume that others have figured out a good way to work. but I also think that we should also think about what will/might work for us ;) | 14:28 |
lucasagomes | +1 | 14:28 |
lucasagomes | Ironic still fairly small... But very distributed, and finding a way to better serve everyone is good. Because you know some folks in Asia/Oceania they have a lot of problems in participating of the meeting | 14:30 |
lucasagomes | so we are already excluding some folks | 14:30 |
NobodyCam | Good Morning Ironic | 14:33 |
lucasagomes | NobodyCam, morning | 14:33 |
dtantsur | NobodyCam, morning :) | 14:34 |
NobodyCam | morning lucasagomes and dtantsur :) | 14:34 |
NobodyCam | morning rloo :) | 14:34 |
rloo | morning NobodyCam | 14:34 |
NobodyCam | :) | 14:37 |
*** mjturek has joined #openstack-ironic | 14:39 | |
dtantsur | brb | 14:42 |
*** jjohnson2_ has quit IRC | 14:51 | |
*** ryanpetrello has joined #openstack-ironic | 15:04 | |
*** jistr has quit IRC | 15:06 | |
MattMan | Hey lucasagomes : enjoying the torrential rain in Dublin :( | 15:06 |
*** jistr has joined #openstack-ironic | 15:07 | |
lucasagomes | MattMan, hi there! Oh man yeah, I went out to get some lunch and it was raining a lot | 15:07 |
MattMan | quick question for you... | 15:09 |
MattMan | driver_info, a key/value dictionary, from what I can see the values are just free format strings, has there ever been a request to support other data type here such as lists ? | 15:10 |
lucasagomes | MattMan, yeah... I believe it would be easy to add, right now the values can be a string or integer | 15:17 |
lucasagomes | I don't know if someone ever asked to support list, but we should because essentially that's a json blob | 15:17 |
lucasagomes | and you can have list with json | 15:17 |
lucasagomes | MattMan, https://github.com/openstack/ironic/blob/master/ironic/api/controllers/v1/node.py#L423-L424 | 15:18 |
lucasagomes | and wsme seems to have a wsme.ArrayType, so it may be a question of adding it to the list | 15:19 |
MattMan | I'll have a play thanks for the pointer | 15:19 |
lucasagomes | MattMan, np | 15:19 |
*** killer_prince has joined #openstack-ironic | 15:21 | |
*** killer_prince is now known as lazy_prince | 15:21 | |
*** pcrews has joined #openstack-ironic | 15:29 | |
*** dlpartain has joined #openstack-ironic | 15:36 | |
*** rushiagr_away is now known as rushiagr | 15:38 | |
*** Marga_ has joined #openstack-ironic | 15:41 | |
*** dlpartain has quit IRC | 15:42 | |
openstackgerrit | Lucas Alvares Gomes proposed openstack/python-ironicclient: VendorPassthru commands to support different HTTP methods https://review.openstack.org/134253 | 15:44 |
*** jistr has quit IRC | 15:53 | |
NobodyCam | ugh so much catch up email :-p | 15:56 |
*** viktors is now known as viktors|afk | 15:57 | |
NobodyCam | brb | 15:57 |
*** dlpartain has joined #openstack-ironic | 16:02 | |
*** dlpartain has quit IRC | 16:05 | |
*** ppalucki has joined #openstack-ironic | 16:13 | |
NobodyCam | I've updated the agenda slightly for the next meeting section. Please feel free to add to or overwright any changes I've made. | 16:16 |
*** Marga_ has quit IRC | 16:26 | |
*** Marga_ has joined #openstack-ironic | 16:27 | |
NobodyCam | dumb question: what is the best way to get a refference link for an email chain? | 16:34 |
dtantsur | I always struggle to do it as well... | 16:35 |
*** Marga_ has quit IRC | 16:37 | |
*** Marga_ has joined #openstack-ironic | 16:38 | |
NobodyCam | :) I generally just search through the cloest month, just thought there may be a better way | 16:38 |
NobodyCam | :-p | 16:39 |
rloo | NobodyCam: not a dumb question and yeah, that's what I do too :-( | 16:42 |
JayF | If it's old enough that it'd be on google, you can do a "words from subject site:lists.openstack.org" | 16:43 |
JayF | domain may be wrong from memory :) | 16:43 |
NobodyCam | :) | 16:45 |
NobodyCam | morning JayF :) | 16:45 |
JayF | morning | 16:45 |
* JayF hopes nobody else from Ironic got sick with France drugs | 16:45 | |
JayF | I joked with some of the guys that I managed to make an end-run around jet lag by being sick enough to sleep on demand regardless of what my internal clock said :P | 16:46 |
JayF | s/drugs/germs/ | 16:46 |
NobodyCam | :) /me managed to avoid the summit plage this time. (knocks on wood) | 16:47 |
*** jistr has joined #openstack-ironic | 16:51 | |
JayF | why knock on wood? I don't think you're going to get it now if you haven't yet :P | 16:52 |
NobodyCam | lol | 16:52 |
*** derekh has quit IRC | 16:58 | |
*** david-lyle_afk is now known as david-lyle | 16:58 | |
*** Marga_ has quit IRC | 17:01 | |
*** marcoemorais has joined #openstack-ironic | 17:02 | |
*** rushiagr is now known as rushiagr_away | 17:07 | |
dtantsur | do we have an official tag for backport candidates? https://bugs.launchpad.net/ironic/+bug/1391960 | 17:08 |
lucasagomes | https://review.openstack.org/#/c/129942/ has 2 +2 already :) someone wants to take a look ? | 17:08 |
JayF | ugh that's unplesant | 17:08 |
adam_g | dtantsur, $rel-backport-potential is whats used across projects | 17:09 |
adam_g | ie, juno-backport-potential | 17:09 |
openstackgerrit | Ruby Loo proposed openstack/ironic: ConductorManager catches Exceptions https://review.openstack.org/122526 | 17:10 |
lucasagomes | adam_g, +1 | 17:10 |
*** ifarkas has quit IRC | 17:11 | |
rloo | lucasagomes, what's wrong? folks afraid to approve your stuff? :-) I'll take a look in a few minutes (going to grab some lunch first) | 17:11 |
NobodyCam | dtantsur: there is this: https://wiki.openstack.org/wiki/Documentation/HowTo#Git_commit_messages_and_backports | 17:11 |
lucasagomes | rloo, hah idk seems so lol | 17:12 |
NobodyCam | lucasagomes: the BP still has Needs approval | 17:14 |
NobodyCam | could that be why? | 17:14 |
lucasagomes | NobodyCam, I don't think so, I mean 2 other patches were merged already | 17:15 |
NobodyCam | lol | 17:15 |
lucasagomes | but that's a good point we should sync the bps with the specs | 17:15 |
lucasagomes | look what was merged and mark it as approved | 17:15 |
lucasagomes | look at* | 17:15 |
* lucasagomes writes it down | 17:15 | |
NobodyCam | +++ | 17:15 |
lucasagomes | day is almost ending here, but tomorrow I can take a look at it | 17:15 |
dtantsur | see you tomorrow! | 17:18 |
NobodyCam | ahh then you'll need to have a pint for /me | 17:18 |
*** dtantsur is now known as dtantsur|afk | 17:18 | |
NobodyCam | night dtantsur|afk | 17:18 |
lucasagomes | dtantsur|afk, see yah | 17:19 |
lucasagomes | NobodyCam, +1 | 17:19 |
rloo | night lucasagomes | 17:19 |
dtantsur|afk | discoverd moving to stackforge \o/ https://review.openstack.org/#/c/133473/ | 17:20 |
lucasagomes | dtantsur|afk, w00t! | 17:20 |
lucasagomes | rloo, will go soonish | 17:21 |
* lucasagomes hope it's not raining | 17:21 | |
rloo | snowing here | 17:21 |
dtantsur|afk | now gone for real :) | 17:21 |
JayF | dtantsur|afk: was it deliberate to not include it in bare metal program? | 17:22 |
lucasagomes | JayF, I think that's the next step, tho yeah it would make sense to do it directly | 17:23 |
lucasagomes | rloo, oh snow <3 | 17:23 |
JayF | for ipa it was done in one swoop | 17:23 |
JayF | slurp in teeth-agent from github --> turned into openstack/ironic-python-agent | 17:24 |
*** dlaube has joined #openstack-ironic | 17:24 | |
lucasagomes | I see, yeah we def should have done that | 17:26 |
NobodyCam | dtantsur|afk: awesome news for discoverd!!! | 17:29 |
*** foexle has joined #openstack-ironic | 17:31 | |
*** jistr has quit IRC | 17:32 | |
*** marcoemorais has quit IRC | 17:32 | |
*** marcoemorais has joined #openstack-ironic | 17:33 | |
*** openstackgerrit has quit IRC | 17:34 | |
*** ppalucki has quit IRC | 17:34 | |
*** openstackgerrit has joined #openstack-ironic | 17:34 | |
*** pawel-palucki has joined #openstack-ironic | 17:35 | |
NobodyCam | brb | 17:35 |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Make vendor methods discoverable via the Ironic API https://review.openstack.org/133533 | 17:36 |
*** achanda has joined #openstack-ironic | 17:37 | |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Make vendor methods discoverable via the Ironic API https://review.openstack.org/133533 | 17:38 |
rloo | lucasagomes: wrt 129942. I'm not crazy about 'supported' to indicate supported http methods. Wouldn't 'http_methods' make more sense? | 17:38 |
rloo | lucasagomes: 'supported' makes me think of a Boolean value. | 17:39 |
lucasagomes | rloo, http_methods would be more explicit indeed | 17:40 |
lucasagomes | maybe supported_http_methods? or too long? | 17:40 |
* lucasagomes is bad naming things >.< | 17:40 | |
rloo | lucasagomes: I think that's too long. I think 'http_methods' or 'supported_http'. hmm, I prefer 'http_methods' | 17:41 |
rloo | lucasagomes: don't worry, the other reviewers didn't comment on it either ;) | 17:41 |
lucasagomes | rloo, right, yeah I'm fine with that | 17:41 |
lucasagomes | mind adding a comment there so I change it tomorrow asap? | 17:41 |
lucasagomes | or today later on at home | 17:41 |
rloo | lucasagomes: yeah, I'm going to add a comment. wanted to check with you first, cuz didn't want to -1 otherwise. | 17:42 |
lucasagomes | rloo, right, yeah it's all good | 17:42 |
rloo | thx lucasagomes! | 17:42 |
lucasagomes | but feel free to -1 any of the patches, I think that naming things consistently is a hard thing and we should try to get it right :D | 17:43 |
lucasagomes | so def a valid -1 :) | 17:43 |
rloo | :-) | 17:43 |
*** colettecello is now known as gothicmindfood | 17:50 | |
*** romcheg has quit IRC | 17:52 | |
*** achanda has quit IRC | 17:52 | |
*** spandhe has joined #openstack-ironic | 17:56 | |
*** achanda has joined #openstack-ironic | 17:56 | |
lucasagomes | alright I will call it a day | 17:58 |
lucasagomes | have a good night everyone | 17:58 |
lucasagomes | rloo, I marked that patch as WIP I will fix it after I get home | 17:59 |
rloo | night lucasagomes. | 17:59 |
rloo | thx lucasagomes; I just added my comments. | 17:59 |
lucasagomes | rloo, awesome alright, thanks! | 17:59 |
*** athomas has quit IRC | 17:59 | |
lucasagomes | see ya | 17:59 |
*** lucasagomes has quit IRC | 17:59 | |
*** marcoemorais has quit IRC | 18:01 | |
*** marcoemorais has joined #openstack-ironic | 18:02 | |
*** marcoemorais has quit IRC | 18:03 | |
*** marcoemorais1 has joined #openstack-ironic | 18:03 | |
*** openstackgerrit has quit IRC | 18:03 | |
*** openstackgerrit has joined #openstack-ironic | 18:04 | |
*** pawel-palucki has quit IRC | 18:05 | |
*** sambetts has quit IRC | 18:08 | |
*** ndipanov is now known as ndipanov_gone | 18:13 | |
yjiang5_away | JoshNang: Hi, just a quick question, I'm reading https://review.openstack.org/#/c/102685/14/specs/kilo/support-zapping-of-nodes.rst , can I understand that zap is mostly similar to decommission, but not in the deployment stage? | 18:27 |
*** eglute has joined #openstack-ironic | 18:27 | |
JoshNang | yjiang5_away: yup! | 18:27 |
JayF | yjiang5_away: zapping is just the new name for decommissioning | 18:27 |
*** yjiang5_away is now known as yjiang5 | 18:27 | |
JayF | yjiang5_away: because many people associate "decommissioning" with permanently deleting and removing a server from their infra | 18:27 |
JoshNang | but also includes more "ready-state" type things, like configuring a baseline raid array | 18:28 |
JoshNang | basically any long running task you want done prior to scheduling | 18:28 |
yjiang5 | JoshNang: I like this :) | 18:28 |
JoshNang | heh me too | 18:28 |
yjiang5 | JoshNang: JayF: to a ESL guy, it's not easy to get the *exact* meaning of these words. Thanks for your explainations. | 18:29 |
JayF | yjiang5: it's not easy for those of us with English as a first language either ;) | 18:30 |
JayF | yjiang5: sometimes I think it'd be easier if we just made up new words and assigned meaning to them | 18:30 |
yjiang5 | JayF: :) | 18:30 |
JoshNang | yjiang5: we spend a lot of time arguing over which word to use a lot :) | 18:30 |
*** foexle has quit IRC | 18:30 | |
NobodyCam | starting to see question about migration support in ironic | 18:31 |
* Shrews wants a PEWPEW state | 18:31 | |
JoshNang | Shrews: haha! | 18:31 |
NobodyCam | lol | 18:31 |
yjiang5 | JoshNang: Noticed the difference of the state machine between your latest spec and your old version decommission spec. | 18:31 |
yjiang5 | Shrews: PEWPEW? what's that? | 18:32 |
JayF | yjiang5: you should def read the big spec victor_lowther put up about the new state maching | 18:32 |
JayF | yjiang5: the sound a zapper would make :P | 18:32 |
JoshNang | yjiang5: is there a difference? i just subbed decom for zap | 18:32 |
JayF | yjiang5: i.e. if you were a kid and were shooting a "fake gun", you'd say "pew pew" | 18:32 |
yjiang5 | JayF: hehe. | 18:32 |
Shrews | yjiang5: http://omglasergunspewpewpew.com | 18:33 |
Shrews | (highly entertaining for a juvenile mind such as mine) | 18:34 |
yjiang5 | Shrews: Got it. like the cute squirrel | 18:34 |
yjiang5 | JoshNang: Aha, I will check again. Thanks. | 18:36 |
*** foexle has joined #openstack-ironic | 18:37 | |
JoshNang | yjiang5: i still envision it as a part of the node lifecycle (post delete, but before scheduling), but you could definitely disable it and only zap when you want via the API | 18:37 |
NobodyCam | lol reading victors spec.. all the impact sections: Probably, Yep, Definitly... lol | 18:39 |
JayF | Impact: Everything evar | 18:39 |
yjiang5 | JoshNang: In latest version and also per victor's spec, It's DELETED->ZAPPING, and in your previous version, IIRC, it's DELETING->DECOMMISSIONING. Am my understanding right? | 18:40 |
JayF | exactly | 18:40 |
JayF | just imagine every instance of "DECOMMISSIONING" goes to "ZAPPING" and "DECOMMISSIONED" to "ZAPPED" | 18:41 |
JayF | just a name change that happened at the summit | 18:41 |
JoshNang | right | 18:41 |
JoshNang | i don't know that the node really ever goes to "DELETED" in actuality. the conductor moves it along as soon as it reports deleting is done | 18:42 |
yjiang5 | JoshNang: JayF, thanks. I will read victor's spec. | 18:43 |
Shrews | JoshNang: is it intended that zapping will be a part of the rebuild process, too? | 18:44 |
Shrews | i.e., nova rebuild | 18:45 |
JoshNang | Shrews: hmm. i'd think not. i haven't looked too deeply at what all happens during rebuild though | 18:45 |
Shrews | JoshNang: ok. leaving comment on the spec for you then | 18:45 |
JayF | Shrews: I'd say zapping will generally occur during a tenant-change | 18:45 |
JoshNang | Shrews: thanks :) | 18:45 |
Shrews | JayF: ++ | 18:46 |
*** pelix has quit IRC | 18:47 | |
*** openstackgerrit has quit IRC | 18:49 | |
*** openstackgerrit has joined #openstack-ironic | 18:49 | |
*** marcoemorais1 has quit IRC | 18:52 | |
*** marcoemorais has joined #openstack-ironic | 18:53 | |
*** marcoemorais has quit IRC | 18:53 | |
*** marcoemorais has joined #openstack-ironic | 18:53 | |
*** achanda has quit IRC | 18:58 | |
*** achanda has joined #openstack-ironic | 18:59 | |
*** marcoemorais has quit IRC | 19:00 | |
*** marcoemorais has joined #openstack-ironic | 19:00 | |
*** marcoemorais has quit IRC | 19:00 | |
*** marcoemorais has joined #openstack-ironic | 19:01 | |
*** marcoemorais has quit IRC | 19:01 | |
yjiang5 | Just wondering what's the nick name of Ramakrishnan in IRC. | 19:02 |
*** marcoemorais has joined #openstack-ironic | 19:02 | |
NobodyCam | yjiang5: I think rameshg87 | 19:03 |
yjiang5 | NobodyCam: thanks. | 19:03 |
*** achanda has quit IRC | 19:07 | |
yjiang5 | NobodyCam: A basic question to https://review.openstack.org/#/c/133828/2/specs/kilo/new-ironic-state-machine.rst , seems all of these states will be kepts in the provision states, right? | 19:07 |
NobodyCam | yjiang5: or target state | 19:10 |
NobodyCam | is the question on the nomenclature of "provision state"? | 19:11 |
yjiang5 | NobodyCam: Yes. I just think we push too much into the provision/target_provision state. | 19:12 |
*** igordcard has quit IRC | 19:13 | |
yjiang5 | NobodyCam: I will try to leave comments on the changeset. If we understanding the provision state as an interface to Nova or other orchestration tools, we should keep that as simple/stable as possible. Discovering IMHO is not part of provision, Zap, depends on the vision, maybe not part either. | 19:13 |
NobodyCam | I can see your point... Yes please leave comments on the review :) | 19:14 |
yjiang5 | NobodyCam: thanks and will do that. | 19:14 |
NobodyCam | this may be a good review for the new meeting format. | 19:15 |
NobodyCam | Are / Would there be objections to me adding the new state machine as a topic for the next meeting? | 19:15 |
JoshNang | NobodyCam: +1000 | 19:16 |
yjiang5 | NobodyCam: +1. | 19:16 |
* NobodyCam updates agenda | 19:16 | |
yjiang5 | JoshNang: you added too much ...... :) | 19:16 |
JoshNang | NobodyCam: we should get that decided and out of the way ASAP. its a blocker for zapping, long running agent, configuring nodes during deploy, etc | 19:17 |
NobodyCam | JoshNang: I agree I have added it to the agenda | 19:18 |
*** openstackgerrit has quit IRC | 19:18 | |
JoshNang | NobodyCam: awesome! | 19:18 |
*** openstackgerrit has joined #openstack-ironic | 19:19 | |
*** dprince has quit IRC | 19:20 | |
*** achanda has joined #openstack-ironic | 19:26 | |
*** marcoemorais has quit IRC | 19:27 | |
NobodyCam | speaking of the new state machine, do we ever expect to support migration of nodes, and if so would we want a state for that? | 19:27 |
openstackgerrit | Sergey Lupersolsky proposed openstack/ironic: Adds help for installing prerequisites on RHEL https://review.openstack.org/134337 | 19:29 |
*** marcoemorais has joined #openstack-ironic | 19:29 | |
yjiang5 | NobodyCam: would it be better to discuss when migration support added? Otherwise, we may have to consider shelf also ( will we support shelf)? | 19:29 |
*** marcoemorais1 has joined #openstack-ironic | 19:31 | |
*** marcoemorais1 has quit IRC | 19:31 | |
NobodyCam | yjiang5: how would you use the shelf state? kinda like maintenance mode? | 19:31 |
JayF | shelving a node makes a lot less sense for ironic | 19:32 |
*** tatyana has joined #openstack-ironic | 19:32 | |
yjiang5 | NobodyCam: No idea :( That's a argument that possibly we can't finish the state machine for all potential states now. We have to update the state machines case by case in future. | 19:33 |
JayF | unless you'd do something like boot a ramdisk, image the disk, store that image somewhere encrypted, and "unshelve" is putting that back on another "compatible" node | 19:33 |
JayF | I think the state engine as proposed gets us through K | 19:33 |
JayF | and is easy to add things to | 19:33 |
yjiang5 | JayF: yes, agree. | 19:33 |
*** marcoemorais has quit IRC | 19:34 | |
*** dprince has joined #openstack-ironic | 19:38 | |
NobodyCam | I've lost the capiablities etherpad, anyone have a link handy? | 19:39 |
*** pensu has quit IRC | 19:40 | |
rloo | NobodyCam: do you mean https://etherpad.openstack.org/p/kilo-ironic-exposing-different-capabilities? | 19:42 |
NobodyCam | rloo: thats the one... Thank you :) | 19:43 |
rloo | NobodyCam: yw | 19:44 |
NobodyCam | back in a bit ... /me heads afk to find food stuffs | 19:57 |
*** eglute has quit IRC | 20:01 | |
*** openstackgerrit has quit IRC | 20:04 | |
*** openstackgerrit has joined #openstack-ironic | 20:05 | |
*** igordcard has joined #openstack-ironic | 20:07 | |
*** eglute_ has joined #openstack-ironic | 20:12 | |
*** mrda_away is now known as mrda | 20:25 | |
mrda | Morning Ironic | 20:25 |
NobodyCam | morning mrda | 20:46 |
*** achanda has quit IRC | 20:47 | |
*** dprince has quit IRC | 20:49 | |
mrda | hey NobodyCam | 20:51 |
*** Marga_ has joined #openstack-ironic | 20:53 | |
*** Marga_ has quit IRC | 20:58 | |
*** marcoemorais has joined #openstack-ironic | 21:02 | |
*** Haomeng|2 has joined #openstack-ironic | 21:11 | |
*** Haomeng has quit IRC | 21:11 | |
*** openstackgerrit has quit IRC | 21:19 | |
*** openstackgerrit has joined #openstack-ironic | 21:19 | |
*** Marga_ has joined #openstack-ironic | 21:20 | |
*** foexle has quit IRC | 21:23 | |
NobodyCam | any one know if sindhusha.yadavalli is in channel? | 21:27 |
*** romcheg has joined #openstack-ironic | 21:41 | |
*** achanda has joined #openstack-ironic | 21:48 | |
*** romcheg has quit IRC | 21:48 | |
*** yjiang5 is now known as yjiang5_away | 21:50 | |
*** achanda has quit IRC | 21:52 | |
*** achanda has joined #openstack-ironic | 21:52 | |
*** tatyana has quit IRC | 21:59 | |
NobodyCam | brb | 22:03 |
*** igordcard has quit IRC | 22:06 | |
*** achanda_ has joined #openstack-ironic | 22:08 | |
openstackgerrit | Josh Gachnang proposed openstack/ironic: Improve IPA client library https://review.openstack.org/111118 | 22:10 |
*** achanda has quit IRC | 22:12 | |
*** Marga_ has quit IRC | 22:12 | |
openstackgerrit | Josh Gachnang proposed openstack/ironic: Improve IPA client library https://review.openstack.org/111118 | 22:13 |
*** Marga_ has joined #openstack-ironic | 22:14 | |
*** romcheg has joined #openstack-ironic | 22:15 | |
*** ryanpetrello has quit IRC | 22:16 | |
*** igordcard has joined #openstack-ironic | 22:21 | |
*** romcheg has quit IRC | 22:22 | |
openstackgerrit | Merged openstack/ironic: ConductorManager catches Exceptions https://review.openstack.org/122526 | 22:23 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic: Updated from global requirements https://review.openstack.org/134383 | 22:28 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/114058 | 22:28 |
NobodyCam | any one have an idea or example of a root-only flavor? | 22:50 |
*** jcoufal has quit IRC | 22:51 | |
JayF | What do you mean by that? | 22:51 |
NobodyCam | line 32 of https://review.openstack.org/#/c/97150/14/specs/kilo/whole-disk-image.rst | 22:51 |
JayF | Hmm. | 22:56 |
JayF | I'm not sure what that means either. | 22:56 |
*** trown is now known as trown|outttypeww | 22:59 | |
*** marcoemorais has quit IRC | 23:00 | |
*** marcoemorais has joined #openstack-ironic | 23:01 | |
NobodyCam | left a comment :-p | 23:05 |
JayF | hmm. https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/ipmitool.py line 677 | 23:09 |
JayF | set_boot_device requires a lock | 23:09 |
JayF | so does raw_bytes() | 23:09 |
JayF | so is it possible for me to call raw_bytes() from within set_boot_device()? | 23:09 |
* JayF working on https://bugs.launchpad.net/ironic/+bug/1390667 | 23:09 | |
*** foexle_ has quit IRC | 23:09 | |
JayF | I have it all implemented inside set_boot_device() but it's a lot of dup'd code | 23:10 |
*** igordcard has quit IRC | 23:11 | |
JayF | JoshNang: ^ do you know? | 23:12 |
JoshNang | lemme lok | 23:13 |
JayF | I'm thinking I might want to rip out most of send_raw() into a private method that I can reuse in both the passthru/public version of send_raw and set_boot_device | 23:14 |
JayF | but I'm not sure | 23:14 |
JoshNang | look, even. but it looks like it'd work? | 23:15 |
*** romcheg has joined #openstack-ironic | 23:15 | |
JayF | Even though they both want a lock? The decorator is smart enough to not deadlock? | 23:15 |
JoshNang | i think it just checks that you have a lock, not takes out a lock | 23:16 |
JoshNang | https://github.com/openstack/ironic/blob/master/ironic/conductor/task_manager.py#L116 | 23:16 |
JoshNang | it just grabs task from args and checks if task.shared is False, otherwise raises an exception | 23:16 |
JayF | aha | 23:17 |
JayF | okay, I'll just do that then. Rockin'. I just wrote a bunch of code for nothign. | 23:17 |
JayF | lol | 23:17 |
JoshNang | :P | 23:18 |
JoshNang | i think we've all been there | 23:18 |
*** romcheg has quit IRC | 23:20 | |
Haomeng|2 | morning | 23:24 |
JayF | mornign | 23:24 |
NobodyCam | morning Haomeng|2 :) | 23:24 |
Haomeng|2 | NobodyCam: :) | 23:26 |
*** Haomeng|2 has quit IRC | 23:31 | |
*** marcoemorais has quit IRC | 23:33 | |
*** marcoemorais has joined #openstack-ironic | 23:34 | |
mrda | Hi Hao | 23:36 |
JayF | JoshNang: aha, I can't call it directly | 23:41 |
JayF | JoshNang: because it's in the VendorPassthru class | 23:41 |
JoshNang | i thought you could still do that...something like task.driver.vendor.$method | 23:43 |
JoshNang | as terribly hacky as that is.. | 23:44 |
JayF | hm | 23:45 |
JoshNang | i think we did something like that in the agent driver to call raw_bytes | 23:46 |
JayF | downstream, you mean? or is that upstream too? | 23:46 |
JoshNang | upstream | 23:46 |
JayF | Aha, so my patch should pull that too. | 23:46 |
JoshNang | hmm maybe not | 23:46 |
JayF | yeah I'm like 90% sure that's only downstream, because we thought it was h/w specific | 23:47 |
JoshNang | yeah you're probably right | 23:47 |
JoshNang | "task.driver.ipmi_vendor.vendor_passthru" | 23:50 |
*** ChuckC has quit IRC | 23:52 | |
*** Haomeng|2 has joined #openstack-ironic | 23:57 | |
*** Haomeng|2 has quit IRC | 23:58 | |
*** Haomeng has joined #openstack-ironic | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!