| *** rbudden has quit IRC | 00:07 | |
| *** kirankv has joined #openstack-ironic | 00:10 | |
| *** baoli has quit IRC | 00:10 | |
| *** r-daneel has quit IRC | 00:11 | |
| *** wajdi has joined #openstack-ironic | 00:12 | |
| *** jcoufal_ has joined #openstack-ironic | 00:13 | |
| *** ishant|away has joined #openstack-ironic | 00:13 | |
| *** jcoufal has quit IRC | 00:15 | |
| *** lazy_prince has joined #openstack-ironic | 00:18 | |
| *** lazy_prince has quit IRC | 00:22 | |
| *** rama_y has joined #openstack-ironic | 00:22 | |
| *** ishant|away has quit IRC | 00:23 | |
| *** kirankv has quit IRC | 00:23 | |
| *** mtanin___ has joined #openstack-ironic | 00:28 | |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce new RPCs to support volume target operations https://review.openstack.org/285220 | 00:28 |
|---|---|---|
| *** fxpester has joined #openstack-ironic | 00:30 | |
| *** mtanino has quit IRC | 00:30 | |
| *** achanda has joined #openstack-ironic | 00:39 | |
| *** hoangcx has joined #openstack-ironic | 00:39 | |
| *** michauds has quit IRC | 00:39 | |
| *** TuanLuong has joined #openstack-ironic | 00:43 | |
| *** phuongnh has joined #openstack-ironic | 00:46 | |
| *** yuanying has quit IRC | 00:54 | |
| *** yuanying has joined #openstack-ironic | 01:02 | |
| *** TuanLuong has quit IRC | 01:03 | |
| *** fxpester has quit IRC | 01:04 | |
| *** achanda has quit IRC | 01:10 | |
| *** sdake has quit IRC | 01:14 | |
| *** jcoufal_ has quit IRC | 01:28 | |
| *** achanda has joined #openstack-ironic | 01:31 | |
| *** bvandewa has quit IRC | 01:35 | |
| *** jcoufal has joined #openstack-ironic | 01:41 | |
| *** yuanying has quit IRC | 01:45 | |
| *** yuanying has joined #openstack-ironic | 01:46 | |
| *** jhesketh has quit IRC | 01:51 | |
| *** jhesketh has joined #openstack-ironic | 01:51 | |
| *** mtanin___ has quit IRC | 01:52 | |
| *** thrash is now known as thrash|g0ne | 01:53 | |
| *** yuanying has quit IRC | 01:55 | |
| *** jcoufal has quit IRC | 02:04 | |
| *** yuanying has joined #openstack-ironic | 02:04 | |
| *** jcoufal has joined #openstack-ironic | 02:05 | |
| *** hoangcx2 has quit IRC | 02:06 | |
| *** milan has quit IRC | 02:06 | |
| *** hshiina has joined #openstack-ironic | 02:06 | |
| *** bvandewa has joined #openstack-ironic | 02:08 | |
| *** milan has joined #openstack-ironic | 02:11 | |
| *** yuanying has quit IRC | 02:24 | |
| *** yuanying has joined #openstack-ironic | 02:25 | |
| *** amitkqed has quit IRC | 02:30 | |
| *** amitkqed has joined #openstack-ironic | 02:30 | |
| *** achanda has quit IRC | 02:32 | |
| *** yuanying has quit IRC | 02:32 | |
| *** milan has quit IRC | 02:36 | |
| *** milan has joined #openstack-ironic | 02:37 | |
| *** achanda has joined #openstack-ironic | 02:44 | |
| *** harlowja_at_home has joined #openstack-ironic | 02:49 | |
| *** harlowja_at_home has quit IRC | 02:57 | |
| *** lazy_prince has joined #openstack-ironic | 03:00 | |
| *** Sukhdev has quit IRC | 03:01 | |
| *** ishant|away has joined #openstack-ironic | 03:01 | |
| *** lazy_prince has quit IRC | 03:05 | |
| *** sabeen has quit IRC | 03:05 | |
| *** lazy_prince has joined #openstack-ironic | 03:05 | |
| *** sabeen has joined #openstack-ironic | 03:06 | |
| *** ishant|away has quit IRC | 03:09 | |
| *** ishant|away has joined #openstack-ironic | 03:09 | |
| *** sabeen1 has joined #openstack-ironic | 03:10 | |
| *** sabeen has quit IRC | 03:12 | |
| *** achanda has quit IRC | 03:14 | |
| *** ishant|away has quit IRC | 03:15 | |
| *** kirankv has joined #openstack-ironic | 03:15 | |
| *** ishant|away has joined #openstack-ironic | 03:17 | |
| *** amotoki has joined #openstack-ironic | 03:27 | |
| *** Romanenko_K has joined #openstack-ironic | 03:31 | |
| *** yuanying has joined #openstack-ironic | 03:48 | |
| *** jcoufal has quit IRC | 03:50 | |
| *** achanda has joined #openstack-ironic | 03:55 | |
| *** achanda has quit IRC | 03:55 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic: Enhance ilo drivers to do inband-inspection https://review.openstack.org/343577 | 04:23 |
| *** crinkle has quit IRC | 04:35 | |
| *** crinkle has joined #openstack-ironic | 04:35 | |
| *** links has joined #openstack-ironic | 04:36 | |
| *** syed_ has quit IRC | 04:42 | |
| *** harlowja_at_home has joined #openstack-ironic | 04:49 | |
| *** harlowja_at_home has quit IRC | 04:55 | |
| *** milan has quit IRC | 05:07 | |
| *** milan has joined #openstack-ironic | 05:07 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic: Follow-up patch for d36bd6f74aa12f6d460fbf086893150a2dc92d63 https://review.openstack.org/349800 | 05:14 |
| *** sabeen1 has quit IRC | 05:16 | |
| *** bks has joined #openstack-ironic | 05:23 | |
| *** appu103_ has joined #openstack-ironic | 05:28 | |
| *** appu103_ is now known as aparnav | 05:29 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic: Fix iLO drivers inconsistent boot mode default value https://review.openstack.org/345231 | 05:32 |
| *** ChubYann has quit IRC | 05:35 | |
| *** moshele has joined #openstack-ironic | 05:44 | |
| openstackgerrit | Bharath kumar proposed openstack/ironic: adding back the inspect_ports as an optional parameter https://review.openstack.org/325230 | 05:45 |
| *** achanda has joined #openstack-ironic | 05:46 | |
| *** f28d798e4ecff has joined #openstack-ironic | 06:06 | |
| *** rama_y has quit IRC | 06:21 | |
| f28d798e4ecff | Hello, guys. I'm trying to implement a basic mdraid (software raid) support for ironic-python-agent to see if it will work good enough in our environment. | 06:22 |
| f28d798e4ecff | However, since I am not a developer at all, it's hard for me to understand some pieces. Currently I have built a test ramdisk with modified IPA including dummy hardware manager with create_configuration and delete_configuration implemented. | 06:24 |
| f28d798e4ecff | I did see that target_raid_config being passed to the create_configuration. However, since I could not find any create_configuration function signature guides, I believe, my method returns not what it supposed to. So, my test node is stuck at clean_wait state. | 06:26 |
| f28d798e4ecff | Althoigh, I have specified that create_configuration and delete_configuration are abortable, in conductor log I can see that both steps are considered non-abortable and clean abort is possible only after the step will be finished. | 06:27 |
| f28d798e4ecff | Can you, please, guide me on how can I force-abort node cleaning? | 06:27 |
| *** mjura has joined #openstack-ironic | 06:29 | |
| *** lok has joined #openstack-ironic | 06:34 | |
| *** mjura has quit IRC | 06:36 | |
| openstackgerrit | Aparna proposed openstack/ironic: Support 'hardware_disk_erase' as a cleaning step https://review.openstack.org/349484 | 06:36 |
| *** mjura has joined #openstack-ironic | 06:36 | |
| bks | can someone review https://review.openstack.org/#/c/325230/9, please? | 06:37 |
| openstackgerrit | David Lenwell proposed openstack/ironic: Introduce VolumeConnector object https://review.openstack.org/214584 | 06:40 |
| openstackgerrit | David Lenwell proposed openstack/ironic: Introduce new RPCs to support volume connector operation https://review.openstack.org/214585 | 06:40 |
| openstackgerrit | David Lenwell proposed openstack/ironic: Add REST API for volume connector operation https://review.openstack.org/214586 | 06:40 |
| openstackgerrit | David Lenwell proposed openstack/ironic: Add volume_connector table to save connector information https://review.openstack.org/200983 | 06:40 |
| *** tesseract- has joined #openstack-ironic | 06:43 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic-inspector: Use OSC in exercise.sh https://review.openstack.org/344616 | 06:44 |
| *** rbartal has joined #openstack-ironic | 06:48 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic: Follow-up patch for d36bd6f74aa12f6d460fbf086893150a2dc92d63 https://review.openstack.org/349800 | 07:01 |
| *** hshiina has quit IRC | 07:07 | |
| *** fxpester has joined #openstack-ironic | 07:11 | |
| *** achanda has quit IRC | 07:15 | |
| *** jpich has joined #openstack-ironic | 07:20 | |
| *** ccamacho is now known as ccamacho|afk | 07:26 | |
| openstackgerrit | David Lenwell proposed openstack/ironic: Add REST API for volume connector and volume target operation https://review.openstack.org/214586 | 07:29 |
| *** jpich has quit IRC | 07:29 | |
| *** mbound has joined #openstack-ironic | 07:35 | |
| *** hshiina has joined #openstack-ironic | 07:40 | |
| *** jpich has joined #openstack-ironic | 07:41 | |
| *** aarefiev has joined #openstack-ironic | 07:43 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Update devstack plugin with new auth options https://review.openstack.org/297312 | 07:45 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Migrate to using keystoneauth Sessions https://review.openstack.org/236982 | 07:45 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: Use new agent API if available https://review.openstack.org/334523 | 07:49 |
| *** ohamada has joined #openstack-ironic | 07:58 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-ironic | 08:00 | |
| *** Goneri has joined #openstack-ironic | 08:07 | |
| *** hecliunyx has quit IRC | 08:07 | |
| *** tangchen has quit IRC | 08:08 | |
| *** tangchen_ has quit IRC | 08:08 | |
| *** hecliunyx has joined #openstack-ironic | 08:09 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 08:10 |
| *** livelace has joined #openstack-ironic | 08:14 | |
| *** jralph_ has joined #openstack-ironic | 08:14 | |
| *** achanda has joined #openstack-ironic | 08:16 | |
| *** jmiu has quit IRC | 08:16 | |
| *** lekha has quit IRC | 08:16 | |
| *** jmiu_ has joined #openstack-ironic | 08:16 | |
| *** jralph has quit IRC | 08:17 | |
| *** strigazi has quit IRC | 08:18 | |
| *** marulkan has quit IRC | 08:18 | |
| *** cppforlife_ has quit IRC | 08:19 | |
| *** lekha has joined #openstack-ironic | 08:19 | |
| *** clenimar has quit IRC | 08:19 | |
| *** tangchen has joined #openstack-ironic | 08:20 | |
| *** tangchen_ has joined #openstack-ironic | 08:20 | |
| *** marulkan1 has joined #openstack-ironic | 08:20 | |
| *** cppforlife__ has joined #openstack-ironic | 08:21 | |
| *** cppforlife__ has quit IRC | 08:21 | |
| *** cppforlife__ has joined #openstack-ironic | 08:21 | |
| *** fungi has quit IRC | 08:22 | |
| *** cppforlife__ is now known as cppforlife_ | 08:22 | |
| *** achanda has quit IRC | 08:22 | |
| *** dtantsur|afk is now known as dtantsur | 08:22 | |
| *** clenimar has joined #openstack-ironic | 08:23 | |
| dtantsur | Morning Ironic (won't call this morning "good" for sure) | 08:23 |
| lucasagomes | dtantsur, morning :D welcome back | 08:23 |
| dtantsur | hey lucasagomes! | 08:24 |
| *** fungi has joined #openstack-ironic | 08:24 | |
| dtantsur | jroll, essentially yes. I'll get back to it, and +2/-1 depending on how I like them :) | 08:25 |
| *** links has quit IRC | 08:26 | |
| *** strigazi has joined #openstack-ironic | 08:26 | |
| *** greghaynes has quit IRC | 08:29 | |
| fxpester | do I understand correct that ironic provides nodes for Nova-Compute, and then Nova can start VMs on them ? or ironic provides end-user services/Machines ? | 08:35 |
| f28d798e4ecff | Hi, fxpester. | 08:38 |
| f28d798e4ecff | It provides the way to manage bare-metal servers in fashion same to the VMs. | 08:39 |
| f28d798e4ecff | It includes hardware inspection, operating system provisioning and pre-deploy disk secure erase. | 08:39 |
| *** greghaynes has joined #openstack-ironic | 08:40 | |
| vsaienk0 | Morning lucasagomes, dtantsur, vdrok, sambetts, jroll and all Ironic'ers! | 08:41 |
| *** links has joined #openstack-ironic | 08:41 | |
| dtantsur | morning vsaienk0 | 08:44 |
| openstackgerrit | Yuiko Takada Mori proposed openstack/ironic: Add node serial console installation howto https://review.openstack.org/293872 | 08:44 |
| pas-ha | fxpester: as part of OpenStack, Ironic is not really though as cloud end-user facing, by default it is only operable by cloud admin. All end-user interactions should go via Nova | 08:45 |
| fxpester | pas-ha: thx | 08:45 |
| lucasagomes | vsaienk0, hi there | 08:46 |
| *** mbound has quit IRC | 08:47 | |
| *** bvandewa has quit IRC | 08:51 | |
| *** daemontool_ has joined #openstack-ironic | 08:51 | |
| *** TuanLA has joined #openstack-ironic | 08:52 | |
| f28d798e4ecff | Guys, sorry for being so bothersome, but let me repeat my question one more time. Is there any way to manually clear clean_step property of the node? Currently I've brought node to manage state, but since clean_step is not empty, looks like ironic conductor is still waiting for the data in AgentRAID._create_configuration_final. | 08:53 |
| *** stendulker has joined #openstack-ironic | 08:54 | |
| *** daemontool__ has quit IRC | 08:54 | |
| *** electrofelix has joined #openstack-ironic | 08:56 | |
| *** mbound has joined #openstack-ironic | 08:56 | |
| lok | dtantsur vsaienko review please https://review.openstack.org/#/c/272658/ | 08:57 |
| *** bks has quit IRC | 08:58 | |
| *** bks has joined #openstack-ironic | 08:58 | |
| *** bks has quit IRC | 09:00 | |
| *** aparnav has quit IRC | 09:02 | |
| *** appu103_ has joined #openstack-ironic | 09:04 | |
| milan | morning Ironic! | 09:05 |
| milan | morning dtantsur lucasagomes fxpester f28d798e4ecff vsaienk0 sambetts lok pas-ha and all! :) | 09:06 |
| dtantsur | morning milan | 09:06 |
| milan | aarefiev, morning :) | 09:07 |
| lucasagomes | milan, morning | 09:08 |
| sambetts | Morning all | 09:16 |
| sambetts | milan: Re: LLDP patch, I'm not sure if the inspector plugin should care about the format of the data? That should be asserted on the IPA side when we ensure the packet we've received is an LLDP one right? | 09:17 |
| sambetts | milan: although it isn't right now :/ | 09:18 |
| dtantsur | morning sambetts. I'd prefer we do server-side validations, anything on the ramdisk is hard to debug/tune | 09:18 |
| sambetts | dtantsur: I'm not sure testing if the data is in a particular order though gains us anything? | 09:20 |
| dtantsur | I'm not sure, my comment was quite general :) | 09:20 |
| dtantsur | i.e. if we do want to validate things, we'd better do it on the ironic-inspector side, if possible | 09:21 |
| sambetts | sure, although in this case I think if we are adding the validation it would be on the IPA side to ensure we've received a valid LLDP packet, so we can decide not to do any parsing of it on the IPA side, if I was sending the packet raw to inspector I would say we should check it on the inspector side | 09:24 |
| sambetts | dtantsur: in here: https://github.com/openstack/ironic-python-agent/blob/master/ironic_python_agent/netutils.py#L142 | 09:24 |
| dtantsur | yeah, some basic sanity check can definitely stay in IPA | 09:24 |
| *** ccamacho|afk is now known as ccamacho | 09:27 | |
| sambetts | milan: what do you think? ^ | 09:27 |
| milan | sambetts, dtantsur I thought about reusing the generic plugin in a vendor plugin | 09:27 |
| milan | so the vendor might first call the generic to process the mandatory fields | 09:27 |
| dtantsur | which vendor are you referring to? | 09:27 |
| sambetts | sure, thats what I intended | 09:27 |
| sambetts | the generic one won't touch any of the other fields | 09:28 |
| milan | also, if other tool than lldptool was used on IPA some sort of sanity should be done server side | 09:28 |
| milan | sambetts, correct | 09:28 |
| sambetts | milan: its all manually implemented on the IPA side | 09:28 |
| *** links has quit IRC | 09:28 | |
| milan | at minimum, there should be some specific error value to indicate IPA failed to parse/validate the LLDP packets | 09:29 |
| milan | in the port lldb data | 09:29 |
| milan | lldb | 09:29 |
| milan | well. autocorrection. | 09:29 |
| sambetts | it'll return None for a port that failed to process a valid LLDP packet | 09:30 |
| sambetts | but it doesn't enforce the LLDPDU format on those packets | 09:30 |
| sambetts | it just reads the TLVs | 09:30 |
| milan | sambetts, what about parsing issues? shouldn't the error message be posted together with the None value? | 09:31 |
| milan | so the plug-in might indicate: oops something wrong with this node | 09:31 |
| sambetts | milan: as of now, if there is a parsing issues its logged in IPA but its just treated as I received an invalid packet therefore None | 09:31 |
| dtantsur | it depends on whether we can get some meaningful error. and whether we can do something meaningful about it too | 09:32 |
| milan | I think None just doesn't say anything | 09:32 |
| sambetts | I think if the operator see the node is receiving no LLDP information and knows the switch should be sending it, then they know what to investigate | 09:33 |
| milan | lldp : { 'error': 'stdout of lldptool'} might be more useful | 09:33 |
| milan | for instance | 09:33 |
| *** e0ne has joined #openstack-ironic | 09:34 | |
| dtantsur | what are you going to do with it next? | 09:34 |
| milan | keep the node on maintenance perhaps? | 09:34 |
| milan | or raise it as utils.Error | 09:34 |
| sambetts | you going to fail inspection because the switch is derp? | 09:35 |
| dtantsur | maintenance is not an option right away | 09:35 |
| dtantsur | as to failing the introspection. for every error we have there are people that want it to be fatal, and there are people that don't | 09:35 |
| milan | sambetts, security-way it would be better to raise than to pass an half-configured node imo | 09:35 |
| dtantsur | e.g. I've got enough requests to make missing local_gb not fatal | 09:35 |
| dtantsur | milan, the problem here is that there is no way to opt-out of such errors from the API standpoint | 09:36 |
| dtantsur | so if you know that one of your nodes derps, you have to disable the hook completely for all nodes | 09:36 |
| milan | dtantsur, ok, but if you enable this plugin and agree that it should do it's best sanity-checking lldp I think it makes sense to raise an error, doesn't it? | 09:36 |
| dtantsur | I'm not so confident | 09:37 |
| dtantsur | this is a usual problem we have: introspection does several things, but from ironic standpoint only basic properties and MACs are mandatory | 09:38 |
| milan | ok, what would break if it did? | 09:38 |
| dtantsur | I don't quite know, because I don't know what can go wrong in LLDP | 09:38 |
| dtantsur | do we fail if some nodes have a switch without LLDP support? | 09:38 |
| milan | node plugged in wrong port maybe | 09:39 |
| sambetts | no, but in that case it would just not receive a packet | 09:39 |
| dtantsur | then what are these errors we expect? | 09:39 |
| sambetts | I think milan is talking about erroring if we receive a packet on the LLDP port number but it fails to parse | 09:39 |
| milan | or we assume a node should get an lldp packet and it doesn't | 09:39 |
| sambetts | or fails to match the LLDP RFC defined format | 09:39 |
| milan | because someone miss-plugged it | 09:40 |
| sambetts | milan: we can have nodes with mixed LLDP and not LLDP ports so unless we list all the ports we can't know | 09:40 |
| *** links has joined #openstack-ironic | 09:40 | |
| milan | sambetts, yeah, true, my way would enforce that if you enabled the generic plugin you declared all nodes&switches supported this | 09:41 |
| sambetts | right, which isn't always the case | 09:41 |
| milan | sambetts, I see | 09:41 |
| milan | OK, so I'm cool lifting my -1 on that | 09:42 |
| sambetts | I think maybe having an error if we receive a packet which doesn't conform to the LLDP standard might be possible, but we could add that as an enhancment to the IPA netutils LLDP code | 09:43 |
| * milan not sure about the None value then | 09:44 | |
| sambetts | we'd have to return the error in a different field anyway, becuase the lldp_data field is defined as a list of lists, so putting an error message in there wouldn't work anyway | 09:45 |
| milan | so something like lldp_error: None/Text? | 09:45 |
| sambetts | or just putting it in the place we put all the other errors | 09:46 |
| sambetts | then it would be handled automagically by inspector anyway | 09:46 |
| *** mgould|afk is now known as mgould | 09:47 | |
| mgould | morning Ironic! | 09:47 |
| dtantsur | morning mgould | 09:47 |
| sambetts | milan: but I'm think this is an enhancement we need to make on the IPA side | 09:47 |
| sambetts | o/ mgould | 09:47 |
| milan | sambetts, OK, sounds reasonable | 09:47 |
| milan | morning mgould :) | 09:47 |
| mgould | morning dtantsur sambetts milan | 09:47 |
| sambetts | milan: awesome :D | 09:48 |
| milan | sambetts, cool, lemme lift my -1 with some future IPA-enhancements comment then ;) | 09:48 |
| sambetts | sweet :) | 09:48 |
| dtantsur | sambetts, milan, I'll be -1 to reusing "error" field for the same reason as above | 09:48 |
| dtantsur | if you have one derping switch, you'll have to disable lldp support for the whole cluster | 09:48 |
| milan | dtantsur, i'm wondering whether in that case the nodes failing introspection aren't the sane thing to do | 09:49 |
| sambetts | hmm, we'll need to RFE/spec it up and work out if error or lldp_error field or something else works better | 09:50 |
| dtantsur | in theory yes, but trust me in practice it's a nightmare | 09:50 |
| dtantsur | cause you have a customer saying "ok, we have one bad node, how do we introspect it with still supporting lldp for other nodes" | 09:50 |
| milan | dtantsur, go to the switch and correct its config/plugs and re-run introspection? | 09:51 |
| dtantsur | it's not sweet to tell your customer "first introspect this node alone, then change the ramdisk kernel arguments, then introspect all nodes but this one" >_< | 09:51 |
| dtantsur | milan, we're talking about malformed packages. I don't think many switching have option "screw up LLDP packages" :) | 09:52 |
| *** vmud213 has joined #openstack-ironic | 09:52 | |
| milan | dtantsur, so in IPA you enabled lldp collection via cmd line but no lldp packet came, should that be an error reported to inspector? | 09:54 |
| sambetts | no | 09:54 |
| sambetts | because we can have some ports that have LLDP and some ports that don't | 09:54 |
| dtantsur | definitely no | 09:54 |
| *** bks has joined #openstack-ironic | 09:54 | |
| milan | sambetts, dtantsur OK cool, then ignore my comments on the patch | 09:55 |
| sambetts | :) | 09:56 |
| milan | ;) | 09:56 |
| tangchen_ | Hi guys, need one more +2 for this one. Thanks. :) | 09:59 |
| tangchen_ | https://review.openstack.org/#/c/349196/ | 09:59 |
| * milan relocates | 10:04 | |
| *** hoangcx has quit IRC | 10:09 | |
| aarefiev | morning ironic! | 10:09 |
| aarefiev | morning milan :) | 10:09 |
| milan | aarefiev, o/ :) | 10:10 |
| sambetts | o/ aarefiev | 10:11 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Add manage_boot parameter to introspection API https://review.openstack.org/316801 | 10:11 |
| dtantsur | morning aarefiev | 10:11 |
| aarefiev | o/ sambetts dtantsur | 10:13 |
| *** phuongnh has quit IRC | 10:15 | |
| *** phuongnh has joined #openstack-ironic | 10:16 | |
| *** milan has quit IRC | 10:18 | |
| *** achanda has joined #openstack-ironic | 10:20 | |
| *** phuongnh has quit IRC | 10:22 | |
| vdrok | late morning all, dtantsur lucasagomes sambetts mgould vsaienk0 | 10:22 |
| dtantsur | morning vdrok | 10:23 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Provide meaningful error messages in functional tests https://review.openstack.org/349924 | 10:23 |
| lucasagomes | vdrok, hello there | 10:24 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Collect deployment logs from IPA https://review.openstack.org/336102 | 10:24 |
| *** achanda has quit IRC | 10:26 | |
| *** dtantsur is now known as dtantsur|lunch | 10:28 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add "erase_devices_metadata_priority" config option https://review.openstack.org/344179 | 10:34 |
| *** fxpester has quit IRC | 10:38 | |
| *** tangchen_ has quit IRC | 10:42 | |
| *** tangchen has quit IRC | 10:42 | |
| *** appu103_ is now known as aparnav | 10:43 | |
| *** tangchen_ has joined #openstack-ironic | 10:44 | |
| *** tangchen has joined #openstack-ironic | 10:45 | |
| *** Romanenko_K has quit IRC | 10:46 | |
| *** TuanLA has quit IRC | 10:48 | |
| *** TuanLA has joined #openstack-ironic | 10:50 | |
| *** fxpester has joined #openstack-ironic | 10:52 | |
| *** Lord_Phoenix has joined #openstack-ironic | 10:57 | |
| *** mbound has quit IRC | 10:58 | |
| sambetts | dtantsur|lunch: hey man, sorry to be a pain, but can you undo your +W on the network_interface default thing, I think we can probably do it better a different way | 10:59 |
| sambetts | dtantsur|lunch: and I'd like to explore that before we merge the whole chain that depends on that change | 10:59 |
| dtantsur|lunch | done | 11:00 |
| * dtantsur|lunch gets back to lunch | 11:00 | |
| sambetts | dtantsur|lunch: thanks :) enjoy | 11:00 |
| *** lok has quit IRC | 11:03 | |
| *** Romanenko_K has joined #openstack-ironic | 11:04 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic-python-agent: Parallel erase disk devices https://review.openstack.org/346723 | 11:07 |
| *** dtantsur|lunch is now known as dtantsur | 11:10 | |
| openstackgerrit | Tang Chen proposed openstack/ironic: Split API root unit and functional tests https://review.openstack.org/348965 | 11:11 |
| openstackgerrit | Tang Chen proposed openstack/ironic: Prepare for functional tests https://review.openstack.org/349421 | 11:11 |
| *** stendulker has quit IRC | 11:24 | |
| *** zhenguo has joined #openstack-ironic | 11:25 | |
| *** sdake has joined #openstack-ironic | 11:27 | |
| jroll | sambetts: what's your plan on it? | 11:27 |
| TheJulia | Good morning | 11:28 |
| sambetts | jroll: we can add a default to the network_interface field, and then the getters and setters implemented inside the versioned object handle the processing of the default if the field is ever == None | 11:29 |
| *** sdake_ has joined #openstack-ironic | 11:30 | |
| jroll | sambetts: yeah, I thought about that, hmm | 11:30 |
| mat128 | morning TheJulia and everyone else | 11:30 |
| sambetts | jroll: I'm just trying to work out all the kinks but so far its mostly working locally | 11:30 |
| dtantsur | morning TheJulia | 11:31 |
| jroll | sambetts: does o.vo have mechanisms for that, or would it just use the build-in python stuff | 11:31 |
| jroll | oh, let's see :) | 11:31 |
| jroll | morning everyone | 11:31 |
| sambetts | jroll: o.vo has built in mechanisms for it | 11:31 |
| jroll | where at? | 11:31 |
| sambetts | jroll: https://github.com/openstack/oslo.versionedobjects/blob/master/oslo_versionedobjects/fields.py#L162 | 11:31 |
| *** sdake has quit IRC | 11:31 | |
| *** tangchen has quit IRC | 11:32 | |
| sambetts | jroll: | 11:32 |
| sambetts | https://github.com/openstack/oslo.versionedobjects/blob/master/oslo_versionedobjects/base.py#L72 | 11:32 |
| jroll | oh right, I tried reading the docs | 11:32 |
| jroll | silly me :D | 11:32 |
| *** bigjools has left #openstack-ironic | 11:32 | |
| sambetts | haha ;) I only know because I've done some hacking around in o.ov to fix some bugs | 11:33 |
| jroll | yeah, I've seen this before now that you link it | 11:33 |
| sambetts | the only problem I'm hitting at the moment is that the default is set at module load time, so changing it during the tests e.g. after overriding the config involves a module reload :/\ | 11:36 |
| sambetts | functionaly its working perfectly | 11:36 |
| jroll | can't you have the default be a method? | 11:37 |
| jroll | maybe not | 11:37 |
| jroll | if that's how we have to do the tests then that's fine | 11:37 |
| jroll | we do that in a few other places | 11:37 |
| * jroll brb | 11:38 | |
| sambetts | its not too bad to add to the tests, just a little more complicated than your orignal plan | 11:38 |
| jroll | yeah | 11:41 |
| jroll | sambetts: wonder if it's worth refactoring later, I'd like to get this done asap as we might end up with an FFE on the nova side | 11:42 |
| sambetts | possibly, whats the impact of adding a function to the versioned node object and then removing it again? | 11:45 |
| *** bfournie has quit IRC | 11:46 | |
| jroll | we'd have to deprecate it and do a major version | 11:48 |
| *** thrash|g0ne is now known as thrash | 11:48 | |
| jroll | which isn't a big deal | 11:48 |
| jroll | or just remove it, given we don't technically do rolling upgrades right anyway | 11:48 |
| jroll | but the former is good practice | 11:48 |
| jroll | anyway let's get this up today and see what it looks like | 11:48 |
| *** tangchen has joined #openstack-ironic | 11:48 | |
| aparnav | dtantsur : Hey, I have updated the description for the RFE https://bugs.launchpad.net/ironic/+bug/1593194. Can you take a look? | 11:49 |
| openstack | Launchpad bug 1593194 in Ironic "[RFE]Add hardware disk erase as cleaning step" [Undecided,Incomplete] - Assigned to Aparna (aparnavtce) | 11:49 |
| dtantsur | in a few minutes, busy right now | 11:49 |
| openstackgerrit | Tang Chen proposed openstack/python-ironicclient: Trivial: Remove useless spaces https://review.openstack.org/349949 | 11:49 |
| *** lucasagomes is now known as lucas-hungry | 11:51 | |
| *** sdake_ is now known as sdake | 11:51 | |
| *** Lord_Phoenix has quit IRC | 11:52 | |
| *** dprince has joined #openstack-ironic | 11:55 | |
| aparnav | dtantsur : okay, Thank you | 11:55 |
| dtantsur | aparnav, so, does this RFE essentially boil down to adding a new configuration option for another cleaning mode? | 12:00 |
| aparnav | dtantsur, yes. | 12:02 |
| aparnav | the cleaning step which can be called us the part of both automated and manual cleaning step | 12:02 |
| dtantsur | jroll, do you think we need some generic way to configure step priorities on server side? ^^^ | 12:05 |
| dtantsur | I'd prefer to avoid an RFE for each vendor clean step... | 12:05 |
| openstackgerrit | Yuiko Takada Mori proposed openstack/ironic: Add node serial console installation howto https://review.openstack.org/293872 | 12:07 |
| *** milan has joined #openstack-ironic | 12:09 | |
| *** aparnav has quit IRC | 12:09 | |
| *** TuanLA has quit IRC | 12:09 | |
| *** aparnav has joined #openstack-ironic | 12:09 | |
| *** appu1034 has joined #openstack-ironic | 12:10 | |
| *** aparnav has quit IRC | 12:10 | |
| *** appu1034 is now known as aparnav | 12:11 | |
| openstackgerrit | Bharath kumar proposed openstack/ironic: adding back the inspect_ports as an optional parameter https://review.openstack.org/325230 | 12:13 |
| aparnav | dtantsur: Do you suggest to submit it as bug? | 12:17 |
| *** baoli has joined #openstack-ironic | 12:18 | |
| jroll | dtantsur: hm, probably | 12:18 |
| jroll | dtantsur: though, rfe for each feature is intentional | 12:18 |
| *** baoli_ has joined #openstack-ironic | 12:18 | |
| jroll | even if trivial | 12:18 |
| *** bfournie has joined #openstack-ironic | 12:19 | |
| *** MattMan has quit IRC | 12:19 | |
| *** MattMan has joined #openstack-ironic | 12:19 | |
| dtantsur | jroll, I understand, but it will get annoying if we treat every possible (including 3rdparty) clean step priority as a separate feature | 12:19 |
| jroll | dtantsur: fair enough | 12:20 |
| *** fxpester|2 has joined #openstack-ironic | 12:21 | |
| *** baoli has quit IRC | 12:22 | |
| *** achanda has joined #openstack-ironic | 12:24 | |
| *** fxpester has quit IRC | 12:25 | |
| *** aparnav has quit IRC | 12:28 | |
| *** achanda has quit IRC | 12:29 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Clean nodes stuck in CLEANING state when ir-cond restarts https://review.openstack.org/349971 | 12:31 |
| *** hoangcx has joined #openstack-ironic | 12:32 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic-inspector: Use OSC in exercise.sh https://review.openstack.org/344616 | 12:34 |
| *** fumie has quit IRC | 12:34 | |
| tangchen_ | dtantsur: Hi Dmitry, in this patch https://review.openstack.org/#/c/348965/, I put a test in functional dir, and enabled the functional test in the previous patch. But it seems that the test didn't run. I think we need to add a job to CI. Would you please tell me how to do it ? | 12:36 |
| dtantsur | tangchen_, hi. I think currently this test is in the experimental pipeline and can be run by leaving a comment "check experimental" on your patch | 12:37 |
| tangchen_ | OH, OK, THanks, I'll try it. :) | 12:38 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Move specs for 6.0.0 https://review.openstack.org/349977 | 12:38 |
| jroll | easy one? maybe? | 12:38 |
| jroll | surprised there was only one spec to move | 12:38 |
| *** joprovost has joined #openstack-ironic | 12:39 | |
| *** jjohnson2 has joined #openstack-ironic | 12:41 | |
| *** bks has quit IRC | 12:42 | |
| *** jjohnson2 has quit IRC | 12:43 | |
| dtantsur | jroll, could you please find some time to look at https://review.openstack.org/236982 ? it's huge and twisted, but it's critical for keystone V3 support IIRC.. | 12:43 |
| jroll | dtantsur: heh, I even had it starred | 12:43 |
| dtantsur | cool! I would love to get keystone v3 support by newton, I know tripleo folks a bit sad because of us not supporing it :) | 12:44 |
| jroll | that's a rough one | 12:44 |
| jroll | true | 12:44 |
| jroll | we should have prioritized it at beginning of cycle | 12:44 |
| * jroll suddenly wonders what happened to the people that were going to put a v3-only devstack job on ironic | 12:44 | |
| jroll | that is a huge patch, wonder if we can split it up a bit | 12:45 |
| *** moshele has quit IRC | 12:46 | |
| jroll | anyway, gotta run a couple errands, bbiab | 12:46 |
| dtantsur | jroll, probably one service at a time, dunno.. | 12:46 |
| dtantsur | pas-ha, ^^? | 12:46 |
| *** links has quit IRC | 12:48 | |
| milan | sambetts, I'd have a question w/r inspector test_process module: for the introspection state support, should I mock the database or rather use sqlite? | 12:49 |
| *** hshiina has quit IRC | 12:49 | |
| * milan actually struggles to make the mock work | 12:49 | |
| milan | dtantsur, aarefiev also maybe have opinions ^^^ | 12:50 |
| dtantsur | milan, I don't think we mock database anywhere | 12:50 |
| milan | yeah, I checked test_node_info uses sqlite | 12:51 |
| aarefiev | yeah, afair we have dbtestbase class for tests | 12:51 |
| milan | cool, dbtestbase it is then :) better than mock imo ;) | 12:52 |
| *** thiagop has joined #openstack-ironic | 12:54 | |
| *** jcoufal has joined #openstack-ironic | 12:54 | |
| *** rbudden has joined #openstack-ironic | 12:55 | |
| thiagop | good morning Ironic | 12:57 |
| thiagop | dtantsur: around? | 12:58 |
| milan | morning thiagop! :) | 13:00 |
| dtantsur | thiagop, yep | 13:00 |
| dtantsur | morning thiagop | 13:00 |
| *** hwcomcn has joined #openstack-ironic | 13:01 | |
| *** hwcomcn has quit IRC | 13:01 | |
| thiagop | dtantsur: about your comment on 340596, setting boot non-persistently is already a feature of Ironic for quite some time. We are just complying with it. Do you think we need a release note for it anyway? | 13:01 |
| *** hwcomcn has joined #openstack-ironic | 13:02 | |
| dtantsur | thiagop, we do release notes for user-visible bug fixes as well, not only features | 13:02 |
| dtantsur | there is a "fixes" section in the release note template | 13:02 |
| *** mtanino has joined #openstack-ironic | 13:02 | |
| thiagop | dtantsur: okay then. I'll write it now (is there a template? I didn't know.) | 13:03 |
| *** lucas-hungry is now known as lucasagomes | 13:03 | |
| dtantsur | thiagop, when you run "tox -evenv reno new <slug>" it ends up with a new file filled in with a template | 13:03 |
| mat128 | jroll: for that v3 only, I think it got setup as an experimental job | 13:04 |
| thiagop | dtantsur: magic! | 13:04 |
| mat128 | have to drive to the office now, I can pull it up later | 13:04 |
| thiagop | dtantsur: thanks for the tip | 13:04 |
| dtantsur | np :) | 13:08 |
| *** fellypefca has quit IRC | 13:08 | |
| *** nicodemos has quit IRC | 13:08 | |
| *** thiagop has quit IRC | 13:09 | |
| *** clenimar has quit IRC | 13:09 | |
| *** xavierr has quit IRC | 13:09 | |
| *** gabriel-bezerra has quit IRC | 13:09 | |
| vdrok | morning jroll and mat128 | 13:10 |
| *** nicodemos has joined #openstack-ironic | 13:10 | |
| *** clenimar has joined #openstack-ironic | 13:10 | |
| *** xavierr has joined #openstack-ironic | 13:11 | |
| *** gabriel-bezerra has joined #openstack-ironic | 13:11 | |
| *** thiagop has joined #openstack-ironic | 13:11 | |
| *** wajdi has quit IRC | 13:13 | |
| pas-ha | dtantsur: one service per-patch migration might be tricky. right now everything is too bound to the common/keystone.py... but I'll see what could be done | 13:15 |
| jroll | pas-ha: or I can just trust you and land it :D | 13:19 |
| pas-ha | that's how I'd naturally prefer it :) | 13:19 |
| jroll | pas-ha: don't worry too much about splitting it, it already has a +2 and such | 13:19 |
| *** causten has joined #openstack-ironic | 13:19 | |
| jroll | I'll review once dtantsur's -1 is resolved | 13:20 |
| pas-ha | ok, as splitting would require a bit ugly things like if a function is passed a session do the new stuff, if not - do the old stuff (and keep the old stuff around) | 13:20 |
| pas-ha | will do | 13:20 |
| lazy_prince | jroll: can you please review https://review.openstack.org/#/c/272658/ and possibly approve it as it already has 2 +2s.. | 13:21 |
| *** links has joined #openstack-ironic | 13:21 | |
| jroll | lazy_prince: looking | 13:22 |
| lazy_prince | jroll: thanks for spending your precious time on it.. | 13:23 |
| *** afaranha has joined #openstack-ironic | 13:23 | |
| openstackgerrit | Sam Betts proposed openstack/ironic: Move network_interface selection logic to node object https://review.openstack.org/350006 | 13:24 |
| sambetts | jroll: ^ I've pushed a version up under a differnt change ID to see what you think | 13:25 |
| openstackgerrit | Merged openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/348627 | 13:25 |
| * jroll queue.push(sambetts_patch) | 13:26 | |
| jroll | sambetts: hm, I like, one question on the tests in comments | 13:29 |
| *** ipukha has joined #openstack-ironic | 13:30 | |
| fxpester|2 | guys I just followed - http://docs.openstack.org/developer/ironic/dev/dev-quickstart.html#deploying-ironic-with-devstack | 13:31 |
| fxpester|2 | and it looks broken X_X no ironic process running, no baremetal flavor created... | 13:31 |
| thiagop | fxpester|2: what the log of devstack is saying? | 13:32 |
| fxpester|2 | Horizon is now available at http://10.0.2.15/dashboard | 13:33 |
| sambetts | jroll: the reason is the default is set at a class/module level so when we update it in other tests it can bleed over into other tests, and that test is specfic to the flat network interface so if we ever changed the default it would break anyway | 13:33 |
| fxpester|2 | so, it says everything is ok | 13:33 |
| thiagop | fxpester|2: maybe the plugin has failed, but the devstack build continued | 13:33 |
| fxpester|2 | but... | 13:33 |
| jroll | lazy_prince: +W | 13:33 |
| fxpester|2 | I built stable mitaka devstack and ironic on fresh ubuntu 14.04.4 | 13:33 |
| thiagop | fxpester|2: you need to see the whole log. Look for "ir-cond" and errors around this word | 13:34 |
| jroll | sambetts: fair enough | 13:34 |
| sambetts | jroll: I tried using module reloads in the tests to get it to recalculate the defaults too, but apprently reloading the versioned objects on the fly breaks a whole ton of the tests/mocks | 13:35 |
| jroll | lol | 13:35 |
| jroll | yeah, what you did is fine | 13:35 |
| jroll | sambetts: can you reply to that comment on gerrit, I'll +2 and rebase on top? | 13:35 |
| *** mbound has joined #openstack-ironic | 13:35 | |
| sambetts | sure, do you want me to change the change id back to your orignal one? | 13:35 |
| jroll | sambetts: nah, it's fine as is | 13:36 |
| jroll | the second patch needs some editing anyway, it was using my new function | 13:36 |
| jroll | method. | 13:36 |
| jroll | sambetts: oh, update your commit message too :) | 13:37 |
| sambetts | jroll: oh yeah, and I need to update the node version note | 13:37 |
| jroll | sambetts: yeah | 13:38 |
| * jroll rebases now anyway and starts testing | 13:38 | |
| vdrok | sambetts: shame that it cannot just accpet a function to get the default like oslo_config does :( | 13:38 |
| sambetts | vdrok: yeah, it would be nice if it could calc it on the fly if given a function | 13:38 |
| jroll | turns out o.vo is open source, maybe someone can add that feature ;) | 13:39 |
| sambetts | ;) | 13:39 |
| sambetts | although i guess an advantage of doing it this way is we only need to calculate it once on process start | 13:39 |
| jroll | yeah, but it isn't very heavy | 13:39 |
| jroll | oh well | 13:39 |
| sambetts | true | 13:40 |
| *** mbound has quit IRC | 13:40 | |
| *** causten has quit IRC | 13:41 | |
| jroll | sambetts: I do wonder if this is going to be annoying for future devs writing tests | 13:41 |
| *** mtanino has quit IRC | 13:41 | |
| jroll | it isn't obvious at all, that you need to do that reset | 13:41 |
| sambetts | yeah :/ /me was wondering about moving that refrsh logic into the constructor of the node object, so that it would recalc it on each node load WDYT? | 13:42 |
| jroll | hm, maybe? | 13:43 |
| sambetts | meanwhile we can pursue getting function based defaults built into o.vo | 13:43 |
| jroll | yeah, I guess that'll be the best route | 13:44 |
| jroll | eat the pain until we do that | 13:44 |
| jroll | welp | 13:45 |
| jroll | updated my api tests with that reload thing | 13:45 |
| jroll | 91 failed | 13:45 |
| jroll | :| | 13:45 |
| sambetts | yeah, I think many many of the tests rely on the default logic | 13:46 |
| sambetts | to define the network_interface :/ | 13:46 |
| sambetts | even when they aren't testing that dfeature | 13:46 |
| openstackgerrit | Thiago Paiva Brito proposed openstack/ironic: Onetime boot when set_boot_device isn't persistent https://review.openstack.org/340596 | 13:46 |
| thiagop | dtantsur: ^ | 13:46 |
| *** mgoddard_ has joined #openstack-ironic | 13:46 | |
| fxpester|2 | thiagop: hate the moment when you create file local.rc, instead of local.conf with local.rc header in it... :( | 13:46 |
| dtantsur | thnx, on the meeting now, will check later | 13:46 |
| thiagop | dtantsur: ack, thanks to you | 13:47 |
| jroll | sambetts: oh, I was just doing it wrong | 13:47 |
| jroll | woohoo | 13:47 |
| sambetts | :-P | 13:47 |
| thiagop | fxpester|2: I never use local.rc. Only local.conf | 13:48 |
| jroll | _set_config() was blowing things up | 13:48 |
| fxpester|2 | what is first line in local.conf ? | 13:48 |
| thiagop | fxpester|2: in fact, I never create. Only copy from /samples and edit | 13:48 |
| sambetts | jroll: ah :/ | 13:48 |
| *** causten has joined #openstack-ironic | 13:49 | |
| *** mgoddard has quit IRC | 13:50 | |
| *** bks has joined #openstack-ironic | 13:50 | |
| *** hoangcx has quit IRC | 13:50 | |
| *** bks has quit IRC | 13:50 | |
| *** bks has joined #openstack-ironic | 13:50 | |
| *** kumarb has joined #openstack-ironic | 13:53 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Add note about network_interface migration to release notes https://review.openstack.org/349546 | 13:53 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Add a data migration to fill node.network_interface https://review.openstack.org/349528 | 13:53 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Test that network_interface is explicitly set on POST/PATCH https://review.openstack.org/349520 | 13:53 |
| *** bks has quit IRC | 13:53 | |
| jroll | sambetts: that's rebased on your patch, the lack of actual changes in 349520 tells me it works :) | 13:54 |
| afaranha | Hello, I've tried the integration code with the noop and flat network interface and it worked fine, but when I tried with neutron network interface the node stays in waiting callback state. I configured the provisioning_network_uuid and cleaning_network_uuid to use the provision network. Do I need to configure something else in order to work? | 13:54 |
| jroll | afaranha: you'll need to make sure your provisioning network has access to the control plane, for one | 13:55 |
| *** e0ne has quit IRC | 13:55 | |
| *** stendulker has joined #openstack-ironic | 13:56 | |
| *** kumarb has quit IRC | 13:56 | |
| *** kumarb has joined #openstack-ironic | 13:56 | |
| stendulker | dtantsur: Hi | 13:57 |
| afaranha | jroll: I'm testing with devstack, and virtual machines as baremetal, everything configured from the ironic integration local conf. Could you tell me how configure the network to access the control plane? | 13:57 |
| *** joprovost has quit IRC | 13:57 | |
| jroll | afaranha: oh, devstack should do that for you | 13:57 |
| stendulker | dtantsur: Regarding review https://review.openstack.org/#/c/345231 | 13:58 |
| vdrok | sambetts: jroll is it possible that at the time when class level 'network_interface' will be initialized with the default, config won't be parsed yet? | 13:58 |
| *** dtantsur is now known as dtantsur|mtg | 13:58 | |
| *** e0ne has joined #openstack-ironic | 13:58 | |
| jroll | vdrok: don't think so, with the cnofig centralization stuff we initialize all the config at startup now | 13:58 |
| vdrok | oh, right, it has changed now | 13:59 |
| afaranha | jroll: Actually devstack is not creating the provision network, I create it based on the configuration in local conf, I create the network normally, and then create a subnet with the subnet prefix and gateway as indicated. Is there any other step? | 13:59 |
| jroll | afaranha: I'm not sure, I haven't run devstack with neutron driver in a while now | 14:00 |
| jroll | vdrok: ^^ do you know | 14:00 |
| *** Goneri has quit IRC | 14:00 | |
| *** anush_ has joined #openstack-ironic | 14:00 | |
| *** causten has quit IRC | 14:00 | |
| vdrok | afaranha: there are more steps iirc | 14:01 |
| vdrok | let me find this place in devstack plugin | 14:01 |
| *** causten_ has joined #openstack-ironic | 14:02 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Move specs for 6.0.0 https://review.openstack.org/349977 | 14:03 |
| afaranha | vdrok: sure, let me find it then, thank you for the help :) | 14:03 |
| thiagop | stendulker: thanks for reviewing 340596. Hope we're using iLO's rest API in the correct way | 14:03 |
| afaranha | thanks jroll | 14:03 |
| *** vmud213 has quit IRC | 14:03 | |
| vdrok | afaranha: here they are I think https://github.com/openstack/ironic/commit/8fe320753c10ede6445781cc884cadc2bc235ea5 | 14:04 |
| *** baoli_ has quit IRC | 14:04 | |
| stendulker | thiagop: I did not check your library side of code as it was merged when I was reviewing it. Will take a look. | 14:04 |
| *** ametts has joined #openstack-ironic | 14:04 | |
| afaranha | vdrok: thanks, I'm gonna check it | 14:04 |
| *** Goneri has joined #openstack-ironic | 14:04 | |
| sambetts | jroll: I'm just testing out a version of the patch with the default getting updated in the __init__ of the node | 14:04 |
| thiagop | stendulker: It would be awesome, thanks | 14:05 |
| jroll | sambetts: cool, take your time | 14:05 |
| jroll | sambetts: those tests I have in the next patch might be helpful too | 14:05 |
| sambetts | jroll: in the POST/PATCH one? | 14:06 |
| jroll | sambetts: yeah, it now only adds tests, no code modifications | 14:06 |
| sambetts | makes sense to squash them then | 14:07 |
| jroll | probably | 14:07 |
| jroll | I don't mind either way | 14:07 |
| openstackgerrit | Thiago Paiva Brito proposed openstack/ironic: Onetime boot when set_boot_device isn't persistent https://review.openstack.org/340596 | 14:08 |
| stendulker | thiagop: It would be better if you set system['Boot']['BootSourceOverrideEnabled'] to 'Once' explicitly | 14:09 |
| stendulker | thiagop: I'm not sure if thts default behavior, nonethless its safe to set it that way for one-time boot | 14:10 |
| stendulker | thiagop: it should have value of 'Continuous' if you want to set it as persistent | 14:10 |
| thiagop | stendulker: to set it persistently, I'd use OneView directly since it enforce it's state | 14:11 |
| thiagop | its* | 14:11 |
| stendulker | thiagop: ok | 14:12 |
| *** links has quit IRC | 14:12 | |
| thiagop | stendulker: but I'll use "Once" to comply with the model, then | 14:15 |
| thiagop | stendulker: ty | 14:15 |
| NobodyCam | Good morning Ironic'ers | 14:17 |
| thiagop | good morning, NobodyCam | 14:17 |
| NobodyCam | Morning thiagop :) | 14:17 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add openstack baremetal driver commands https://review.openstack.org/350050 | 14:19 |
| *** kumarb has quit IRC | 14:23 | |
| *** kumarb has joined #openstack-ironic | 14:23 | |
| jlvillal | Good morning everyone :) | 14:25 |
| thiagop | morning jlvillal | 14:25 |
| Romanenko_K | hi, jlvillal! | 14:25 |
| openstackgerrit | Merged openstack/ironic: Use 'UUID', not 'uuid' in exception strings https://review.openstack.org/346883 | 14:26 |
| jlvillal | Hi thiagop Romanenko_K :) | 14:26 |
| *** stendulker has quit IRC | 14:28 | |
| *** mgoddard_ has quit IRC | 14:29 | |
| *** mgoddard has joined #openstack-ironic | 14:29 | |
| *** wajdi has joined #openstack-ironic | 14:30 | |
| *** kumarb has quit IRC | 14:31 | |
| *** kumarb has joined #openstack-ironic | 14:31 | |
| *** kumarb has quit IRC | 14:35 | |
| *** kumarb has joined #openstack-ironic | 14:35 | |
| openstackgerrit | Thiago Paiva Brito proposed openstack/python-oneviewclient: Comply with iLO model of usage for one-time boot https://review.openstack.org/350060 | 14:36 |
| *** achanda has joined #openstack-ironic | 14:36 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic-inspector: Check whether action value is string before calling format() https://review.openstack.org/349392 | 14:37 |
| *** sborkows has joined #openstack-ironic | 14:39 | |
| *** hoangcx has joined #openstack-ironic | 14:39 | |
| *** kumarb has quit IRC | 14:39 | |
| *** PollyZ has joined #openstack-ironic | 14:42 | |
| openstackgerrit | Sam Betts proposed openstack/ironic: Move default network_interface logic in node object https://review.openstack.org/350006 | 14:43 |
| openstackgerrit | Sam Betts proposed openstack/ironic: Test that network_interface is explicitly set on POST/PATCH https://review.openstack.org/349520 | 14:43 |
| sambetts | jroll: ^ | 14:43 |
| *** mtanino has joined #openstack-ironic | 14:45 | |
| NobodyCam | Morning jlvillal, Romanenko_K, sambetts :) | 14:47 |
| *** achanda has quit IRC | 14:48 | |
| *** hoangcx has quit IRC | 14:50 | |
| *** hoangcx2 has joined #openstack-ironic | 14:50 | |
| *** hwcomcn has quit IRC | 14:51 | |
| *** hoangcx2 has quit IRC | 14:51 | |
| sambetts | o/ NobodyCam | 14:51 |
| NobodyCam | :) | 14:51 |
| jmiu_ | good morning NobodyCam | 14:51 |
| NobodyCam | morning jmiu_ :) | 14:51 |
| *** hoangcx has joined #openstack-ironic | 14:51 | |
| *** jvgrant has joined #openstack-ironic | 14:52 | |
| *** michauds has joined #openstack-ironic | 14:53 | |
| *** joprovost has joined #openstack-ironic | 14:54 | |
| *** rama_y has joined #openstack-ironic | 14:56 | |
| fxpester|2 | guys, I still have bunch of stupid questions %) I`m testing mitaka devstack now, so I assume KVM host is like `baremetal` machine | 14:57 |
| fxpester|2 | does ironic-node running neutron-agent, nova-agent, etc-agents ? | 14:58 |
| fxpester|2 | how to check it ? | 14:58 |
| *** rama_y has quit IRC | 15:00 | |
| *** hshiina has joined #openstack-ironic | 15:00 | |
| *** jvgrant has quit IRC | 15:03 | |
| *** rama_y has joined #openstack-ironic | 15:04 | |
| *** mjura has quit IRC | 15:04 | |
| *** baoli has joined #openstack-ironic | 15:06 | |
| sambetts | fxpester|2: in devstack we create VMs which simulate baremetal machines | 15:06 |
| sambetts | fxpester|2: if you do a virsh list --all | 15:07 |
| fxpester|2 | sambetts: yes, how to get into that VM ? | 15:07 |
| fxpester|2 | sambetts: yes, I can see it | 15:07 |
| fxpester|2 | sambetts: can`t figure it`s ip for now :D | 15:07 |
| sambetts | fxpester|2: into it? after deploying an image to it? | 15:07 |
| sambetts | ssh or something? | 15:07 |
| *** dtantsur|mtg is now known as dtantsur | 15:08 | |
| fxpester|2 | sambetts: exactly | 15:08 |
| fxpester|2 | ssh | 15:08 |
| jroll | 'nova list' or 'nova show' should give you the IP of the instance | 15:08 |
| *** mbound has joined #openstack-ironic | 15:10 | |
| fxpester|2 | so, in `nova list`, `ironic node-list` and `virsh list` I see exactly same Virtual/Baremetal machine ? no additional virtualization or abstraction ? | 15:13 |
| vgadiraj | mariojv: Are you still working on this bug? https://bugs.launchpad.net/ironic/+bug/1602410 | 15:13 |
| openstack | Launchpad bug 1602410 in Ironic "agent_last_heartbeat type not consistent with other date fields" [Undecided,New] - Assigned to Mario Villaplana (mario-villaplana-j) | 15:13 |
| *** syed_ has joined #openstack-ironic | 15:14 | |
| mariojv | vgadiraj: i was planning on working on it later this week, but feel free to start on it if you're interested | 15:14 |
| lucasagomes | fxpester|2, what you mean by "additional virtualization or abstraction" ? | 15:15 |
| lucasagomes | fxpester|2, the VM (the one you see in virsh list) is the same that is registered in Ironic (ironic node-list) and (after deployed) have a nova instance running on it (nova list) | 15:16 |
| fxpester|2 | lucasagomes: what I was sure about: We call `nova boot`, nova calls Ironic, Ironic bootstrap physical machine, setup it as new Compute node, and creating my new VM on it | 15:16 |
| jroll | fxpester|2: no | 15:17 |
| lucasagomes | fxpester|2, no really... I mean, that sounds like TripleO | 15:17 |
| sambetts | fxpester|2: you call nova boot, nova calls Ironic, Ironic provisions the baremetal node as your instance | 15:17 |
| lucasagomes | but ironic itself stops at the boostrap phase | 15:17 |
| *** Sukhdev has joined #openstack-ironic | 15:20 | |
| *** milan has quit IRC | 15:22 | |
| *** hoangcx has quit IRC | 15:24 | |
| *** Sukhdev has quit IRC | 15:25 | |
| *** hoangcx has joined #openstack-ironic | 15:27 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Update devstack plugin with new auth options https://review.openstack.org/297312 | 15:28 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Migrate to using keystoneauth Sessions https://review.openstack.org/236982 | 15:28 |
| *** Sukhdev has joined #openstack-ironic | 15:30 | |
| *** sabeen has joined #openstack-ironic | 15:32 | |
| *** rbartal has quit IRC | 15:33 | |
| *** hshiina has quit IRC | 15:33 | |
| *** lucasagomes has quit IRC | 15:37 | |
| *** lucasagomes has joined #openstack-ironic | 15:44 | |
| dtantsur | going now, see you tomorrow! | 15:45 |
| sambetts | cya dtantsur | 15:45 |
| *** dtantsur is now known as dtantsur|afk | 15:45 | |
| lucasagomes | dtantsur, see ya, have a great evening | 15:45 |
| *** afaranha has quit IRC | 15:47 | |
| vgadiraj | I had a question. I am running tox -edocs on ubuntu 16.04. When running Sphinx, it states Using openstack theme from /home/ubuntu/ironic/.tox/venv/local/lib/python2.7/site-packages/oslosphinx/theme. It's opening the theme from python2.7. Does anyone know how I can change this to use python3.5? | 15:48 |
| *** sborkows has quit IRC | 15:50 | |
| jroll | hm, docs uses the 'venv' tox target which is always 2.7 right now | 15:58 |
| jroll | vgadiraj: any specific reason you need that using 3.5? | 15:58 |
| jroll | vgadiraj: you can change it with this line, change venv to py35 https://github.com/openstack/ironic/blob/master/tox.ini#L67 | 15:59 |
| jroll | does anyone know where (or if) the docs exist for setting up multitenant networking? vdrok? vsaienk0? | 15:59 |
| *** vmud213 has joined #openstack-ironic | 15:59 | |
| vdrok | jroll: yep, there is, lemme find it | 16:00 |
| jroll | thanks | 16:00 |
| vdrok | jroll: kind of :) https://github.com/openstack/openstack-manuals/blob/935d55f7b8a91f949e8a71adec4e2316f83b48fd/doc/admin-guide/source/baremetal-multitenancy.rst | 16:01 |
| jroll | whoa, in the admin guide? O_o | 16:02 |
| jroll | awesome | 16:02 |
| *** tangchen_ has quit IRC | 16:02 | |
| *** tangchen has quit IRC | 16:03 | |
| vgadiraj | jroll: there is a bug report saying that it is breaking using 3.5 | 16:03 |
| vgadiraj | thanks! | 16:03 |
| vdrok | jroll: also there is this patch, but it was not updated for a while https://review.openstack.org/#/c/228496/85 | 16:03 |
| jroll | vgadiraj: aha, right on. good luck :) | 16:04 |
| jroll | vdrok: oh, I knew about that one but assumed it was about portgroups only... | 16:04 |
| jroll | yay commit messages | 16:04 |
| jroll | we should split that up | 16:04 |
| vdrok | jroll: yup, I'll try to do that today | 16:04 |
| *** jcoufal_ has joined #openstack-ironic | 16:04 | |
| jroll | you rock, ty | 16:04 |
| *** Sukhdev has quit IRC | 16:05 | |
| *** jcoufal has quit IRC | 16:07 | |
| *** Nisha_away has joined #openstack-ironic | 16:07 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add openstack baremetal driver commands https://review.openstack.org/350050 | 16:08 |
| openstackgerrit | Dao Cong Tien proposed openstack/ironic: Follow-up patch of 857372a2269cdd0f8a1ae5b9e9f6e0ee193f01be https://review.openstack.org/349400 | 16:10 |
| *** tangchen_ has joined #openstack-ironic | 16:16 | |
| *** rpioso has joined #openstack-ironic | 16:19 | |
| *** mbound has quit IRC | 16:21 | |
| *** tangchen has joined #openstack-ironic | 16:23 | |
| jroll | so uh, nova multitenant networking patch has a +W https://review.openstack.org/#/c/297895/ | 16:24 |
| jroll | :D | 16:24 |
| jroll | \o/ | 16:24 |
| jroll | I would say thank you to a bunch of people here, but it's basically the whole channel, so thank you all for your hard work on that! | 16:25 |
| JayF | \o/ | 16:25 |
| sambetts | \o/ | 16:25 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Follow-up patch of 857372a2269cdd0f8a1ae5b9e9f6e0ee193f01be https://review.openstack.org/349400 | 16:25 |
| * lucasagomes just updated the commit message ^ | 16:25 | |
| lucasagomes | jroll, w00t! | 16:26 |
| mgould | jroll: \o/ | 16:26 |
| jroll | lucasagomes: can you put a more descriptive first line on that so git log is easily readable? | 16:26 |
| jroll | back to nova: now, that won't land until we land this chain: https://review.openstack.org/#/q/topic:bug/1608511 | 16:26 |
| lucasagomes | jroll, yeah, lets do it... that old behavior of "follow-up of blah..." | 16:27 |
| lucasagomes | 1 sec | 16:27 |
| jroll | so let's get to work on that | 16:27 |
| jroll | lucasagomes: heh, yeah ty | 16:27 |
| *** krotscheck is now known as krot_sickleave | 16:27 | |
| *** Sukhdev has joined #openstack-ironic | 16:27 | |
| *** e0ne has quit IRC | 16:28 | |
| *** amotoki has quit IRC | 16:28 | |
| *** e0ne has joined #openstack-ironic | 16:28 | |
| jroll | sambetts: oh man, some flaky unit tests https://review.openstack.org/#/c/349528/ http://logs.openstack.org/28/349528/6/check/gate-ironic-python27-db-ubuntu-xenial/dfe55aa/console.html#_2016-08-02_15_42_36_838416 | 16:29 |
| jroll | :/ | 16:29 |
| jroll | that's on the old patch though, ignore me | 16:29 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Add a data migration to fill node.network_interface https://review.openstack.org/349528 | 16:29 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Add note about network_interface migration to release notes https://review.openstack.org/349546 | 16:29 |
| * jroll rebases | 16:29 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Minor docstring and unittests fixes for IPMIConsole https://review.openstack.org/349400 | 16:30 |
| lucasagomes | jroll, ^ | 16:30 |
| * lucasagomes +2'd the patch anyway cause he only changed the commit message | 16:31 | |
| lucasagomes | JayF, sorry I overwrote ur +W on ^ because jroll asked me to change the first line of the patch | 16:31 |
| *** causten_ has quit IRC | 16:31 | |
| lucasagomes | of the commit message of the patch* | 16:31 |
| JayF | lucasagomes: I -W'd myself first for that same reason | 16:31 |
| JayF | lucasagomes: b/c I saw that comment here | 16:31 |
| lucasagomes | cool :-) | 16:31 |
| lucasagomes | I forgot to refresh the patch heh | 16:32 |
| jroll | lol | 16:32 |
| *** Goneri has quit IRC | 16:37 | |
| *** jmiu_ is now known as jmiu|lunch | 16:38 | |
| rajinir | thiagop: hi, with two networks , we have made some progress. Need some help from you to debug this. I can see the DHCP Request and Reply from the VM. But the hardware node is not seeing the reply | 16:39 |
| *** Goneri has joined #openstack-ironic | 16:40 | |
| sambetts | rajinir: do you have neutron port security enabled in your node provider? | 16:40 |
| rajinir | sambetts: Are you asking about the security groups on the devstack host? | 16:41 |
| sambetts | rajinir: on the node running the VMs that are your jenkins slaves | 16:42 |
| sambetts | rajinir: you need to disable port security on that host, because it blocks DHCP being served by the VMs | 16:43 |
| rajinir | https://www.irccloud.com/pastebin/Gj2GLk7e/Ironic-debug-drac | 16:43 |
| rajinir | sambetts: Have set the Q_USE_SECGROUP to false | 16:44 |
| lucasagomes | folks, calling it a day | 16:44 |
| sambetts | o/ lucasagomes | 16:44 |
| lucasagomes | have a great evening all, see ye tomorrow | 16:44 |
| rajinir | sambetts: This is the slave VM config | 16:44 |
| *** lucasagomes is now known as lucas-dinner | 16:44 | |
| sambetts | rajinir: thats in the test VM though right? I'm talking about the node that is hosting the test VMs | 16:44 |
| rajinir | sambetts: On the host how I disable? | 16:45 |
| rajinir | sambetts: Is it the same setting for the hosting node? | 16:45 |
| sambetts | rajinir: is your hosting node using devstack? | 16:46 |
| rajinir | sambetts: Yes,host node is devstack | 16:46 |
| *** fxpester|2 has quit IRC | 16:46 | |
| *** jpich has quit IRC | 16:47 | |
| rajinir | sambetts: something like this ? neutron port-update <port_id> ---port-security-enabled=False | 16:49 |
| thiagop | rajinir: back from lunch. Morning! | 16:50 |
| * thiagop reads | 16:50 | |
| sambetts | rajinir: right, you can do it like that but as the ports are created on the fly by node-pool that won't work, you can however disable it at a network level | 16:51 |
| *** causten has joined #openstack-ironic | 16:51 | |
| rajinir | sambetts: net-update | 16:51 |
| *** tesseract- has quit IRC | 16:51 | |
| sambetts | right | 16:51 |
| thiagop | rajinir: what I think sambetts is saying is that you need to disable port_security on the cloud you're using to launch the devstack | 16:51 |
| *** bvandewa has joined #openstack-ironic | 16:53 | |
| rajinir | sambetts: thiagop: got it, let me give it a try. thanks | 16:53 |
| rajinir | sambetts, do I have to restart neutron with this change for it to be effective? | 16:55 |
| sambetts | rajinir: nope you should have too | 16:56 |
| *** jvgrant has joined #openstack-ironic | 16:56 | |
| rajinir | sambetts: ok, may be recreate the vm? | 16:56 |
| rajinir | sambetts: didn't see any reply yet | 16:57 |
| thiagop | rajinir: If the underlying cloud is <=mitaka, you may want to see this: https://bugs.launchpad.net/nova/+bug/1175464 | 16:57 |
| openstack | Launchpad bug 1175464 in OpenStack Compute (nova) "should not add default security group to quantum unless api-request had it" [High,In progress] - Assigned to sahid (sahid-ferdjaoui) | 16:57 |
| thiagop | rajinir: this one fixed it for newton -> https://review.openstack.org/#/c/284095/ | 16:58 |
| *** chlong has quit IRC | 16:58 | |
| thiagop | rajinir: we had to apply it by hand in our kilo cloud's nova... | 16:59 |
| rajinir | thiagop: ok, we built this devstack on the host machine very recently, so it is newton based for sure | 16:59 |
| thiagop | rajinir: are you running a devstack inside a devstack? | 16:59 |
| * thiagop minds blows | 16:59 | |
| thiagop | mind* | 16:59 |
| sambetts | thiagop: ;) | 17:00 |
| rajinir | thiagop: yes :( | 17:00 |
| *** f28d798e4ecff has quit IRC | 17:00 | |
| thiagop | rajinir: but anyway, if DHCP requests are being seen in the devstack's network, I should work | 17:01 |
| *** PollyZ has quit IRC | 17:01 | |
| rajinir | thiagop: ok, so there is hope ;) | 17:01 |
| *** ohamada has quit IRC | 17:02 | |
| thiagop | unless a routing table somewhere are sending this package somewhere else other than back to private_net's gateway.. | 17:02 |
| rajinir | thiagop: ok | 17:02 |
| rajinir | sambetts,thiagop: May be I have to disable on the port associated with the VM or recreate it. I tried to disable on port direct and got this message Port has security group associated. Cannot disable port security or ip address until security group is removed | 17:07 |
| thiagop | rajinir: remove the security group then disable port_security | 17:08 |
| *** anush_ has quit IRC | 17:11 | |
| *** chlong has joined #openstack-ironic | 17:12 | |
| *** ametts has quit IRC | 17:12 | |
| rajinir | thiagop, sambetts: that worked, I see the replies. :) | 17:12 |
| thiagop | rajinir: w00t | 17:12 |
| sambetts | \o/ | 17:12 |
| *** ametts has joined #openstack-ironic | 17:13 | |
| *** Goneri has quit IRC | 17:13 | |
| rajinir | thiagop: sambetts: Yay, thank you. great | 17:14 |
| *** mbound has joined #openstack-ironic | 17:16 | |
| jroll | easy specs review if someone has a moment, just marking one as implemented https://review.openstack.org/#/c/349977/ | 17:17 |
| *** crinkle has quit IRC | 17:17 | |
| JayF | jroll: landed it | 17:18 |
| jroll | thanks | 17:18 |
| * jroll spots devananda | 17:19 | |
| devananda | :) | 17:19 |
| devananda | g'morning | 17:19 |
| jroll | morning :) | 17:19 |
| sambetts | o/ devananda | 17:19 |
| devananda | today has a lot less distractions than yesterday - hopefully I can get to the bottom of hte keystone policy patch failures | 17:19 |
| devananda | I also see a few patches up for migration to keystoneauth that I want to test, and possibly pull into a single patch chain | 17:19 |
| devananda | as they're all touching the auth systems and how we configure users in devstack | 17:20 |
| *** moshele has joined #openstack-ironic | 17:21 | |
| jroll | devananda: ++ | 17:22 |
| openstackgerrit | Merged openstack/ironic-specs: Move specs for 6.0.0 https://review.openstack.org/349977 | 17:24 |
| *** e0ne has quit IRC | 17:24 | |
| *** vmud213 has quit IRC | 17:26 | |
| *** Goneri has joined #openstack-ironic | 17:28 | |
| *** bvandewa has quit IRC | 17:31 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Operator documentation for multitenancy https://review.openstack.org/228496 | 17:31 |
| vdrok | morning devananda | 17:31 |
| vdrok | jroll: ^^ first try to fix that :) | 17:32 |
| jroll | vdrok: nice, ty | 17:32 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: [WIP] add test to ensure policy is always authorized https://review.openstack.org/350177 | 17:32 |
| *** bvandewa has joined #openstack-ironic | 17:34 | |
| *** bvandewa has quit IRC | 17:34 | |
| * sambetts -> home | 17:34 | |
| sambetts | night all | 17:34 |
| *** sambetts is now known as sambetts|afk | 17:35 | |
| thiagop | night sambetts|afk | 17:35 |
| *** tangchen has quit IRC | 17:35 | |
| openstackgerrit | Michael Turek proposed openstack/ironic: Add portgroups to support LAG interfaces - API https://review.openstack.org/332177 | 17:37 |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Add volume_connector table to save connector information https://review.openstack.org/200983 | 17:37 |
| *** tangchen has joined #openstack-ironic | 17:39 | |
| * vdrok leaves too | 17:40 | |
| vdrok | good night o/ | 17:40 |
| thiagop | good night vdrok | 17:40 |
| JayF | with the new conf reorganization stuff, I'm really confused about the roles of ironic/conf/default.py vs the other files | 17:43 |
| JayF | who might be a good person to ask about what? | 17:43 |
| jroll | JayF: what's your question? | 17:44 |
| JayF | WTF is that file for | 17:44 |
| JayF | more or less | 17:44 |
| jroll | things that don't fit elsewhere, iirc? | 17:44 |
| JayF | it appears to have a random amalgamation of configs in there | 17:44 |
| JayF | that seem to be duplicated from groups in other files | 17:44 |
| JayF | like there are api opts in there, despite having an api.py defining that group | 17:44 |
| jroll | yeah | 17:44 |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce VolumeConnector object https://review.openstack.org/214584 | 17:44 |
| jroll | maybe it's everything in the default group? | 17:45 |
| jroll | JayF: rama_y did that work, btw | 17:45 |
| JayF | yep, you're right | 17:45 |
| jroll | Sukhdev: in case you missed it, the nova patch for multitenant networking is approved, and is waiting for some dependent patches in ironic :D | 17:45 |
| JayF | just confusing how it's laid out | 17:45 |
| JayF | thanks, I thought it was something silly I wasn't getting | 17:45 |
| jroll | yay for guessing correctly \o/ | 17:45 |
| JayF | lol | 17:46 |
| *** Sukhdev has quit IRC | 17:46 | |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce new RPCs to support volume connector operation https://review.openstack.org/214585 | 17:50 |
| *** electrofelix has quit IRC | 17:54 | |
| *** PollyZ has joined #openstack-ironic | 17:54 | |
| jroll | api meeting in 5 minutes | 17:55 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Pass agent metrics config via conductor https://review.openstack.org/301919 | 17:57 |
| JayF | ^ now updated for config centralization | 17:57 |
| JayF | reviews appreciated :D | 17:57 |
| *** Goneri has quit IRC | 18:00 | |
| *** crinkle has joined #openstack-ironic | 18:00 | |
| *** tangchen has quit IRC | 18:02 | |
| *** mgould is now known as mgould|afk | 18:02 | |
| *** mbound has quit IRC | 18:02 | |
| *** tangchen has joined #openstack-ironic | 18:04 | |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Add volume_targets table to save target information https://review.openstack.org/285218 | 18:08 |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce VolumeTarget object https://review.openstack.org/285219 | 18:09 |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce new RPCs to support volume target operations https://review.openstack.org/285220 | 18:09 |
| *** daemontool_ has quit IRC | 18:12 | |
| *** e0ne has joined #openstack-ironic | 18:17 | |
| *** Nisha_away has quit IRC | 18:17 | |
| *** ChubYann has joined #openstack-ironic | 18:18 | |
| *** Nisha_away has joined #openstack-ironic | 18:18 | |
| *** moshele has quit IRC | 18:24 | |
| *** bvandewa has joined #openstack-ironic | 18:25 | |
| *** jrist has quit IRC | 18:27 | |
| *** jrist has joined #openstack-ironic | 18:28 | |
| *** jrist has quit IRC | 18:30 | |
| *** hoangcx has quit IRC | 18:32 | |
| *** devananda has quit IRC | 18:38 | |
| jroll | bbiab | 18:40 |
| *** devananda has joined #openstack-ironic | 18:46 | |
| *** devananda has quit IRC | 18:46 | |
| *** devananda has joined #openstack-ironic | 18:47 | |
| openstackgerrit | David Lenwell proposed openstack/ironic: Add REST API for volume connector and volume target operation https://review.openstack.org/214586 | 18:48 |
| *** Sukhdev has joined #openstack-ironic | 18:50 | |
| *** devananda has quit IRC | 18:55 | |
| *** hoangcx has joined #openstack-ironic | 18:57 | |
| *** mbound has joined #openstack-ironic | 19:03 | |
| *** anush_ has joined #openstack-ironic | 19:04 | |
| *** PollyZ has quit IRC | 19:08 | |
| *** mbound has quit IRC | 19:08 | |
| *** Nisha_away has quit IRC | 19:08 | |
| *** livelace has quit IRC | 19:11 | |
| *** afaranha has joined #openstack-ironic | 19:13 | |
| *** devananda has joined #openstack-ironic | 19:14 | |
| JayF | https://review.openstack.org/#/c/344164/ is a pretty contained IPA change if someone wants to take a look | 19:15 |
| JayF | I already have a +2 on it, plus 2x+1s from other folks | 19:16 |
| JayF | CI passing, etc | 19:16 |
| openstackgerrit | David Lenwell proposed openstack/ironic: Add REST API for volume connector and volume target operation https://review.openstack.org/214586 | 19:18 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: Implement rescue mode https://review.openstack.org/171878 | 19:21 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Timing metrics for agent deploy classes https://review.openstack.org/350243 | 19:32 |
| JayF | ^ that's a super easy review if someone wants a quick thing to look at, also it's a priority this cycle so \o/ | 19:32 |
| devananda | JayF: question on that ... | 19:41 |
| devananda | JayF: the metric names are just class.function | 19:41 |
| devananda | however, metrics are, as I see it, an API we are exposing | 19:41 |
| *** PollyZ has joined #openstack-ironic | 19:42 | |
| devananda | anyone who uses these metrics is going to create local configuration (ie, integration with their monitoring stack) based on the metric name | 19:42 |
| devananda | if we refactor any of this code, we won't want to change the metric names -- because it'll negatively affect users | 19:43 |
| devananda | SO - why name them based on the class.function in the first place? | 19:43 |
| *** fxpester|2 has joined #openstack-ironic | 19:43 | |
| JayF | devananda: If we refactored, most likely the new metrics would be appreciably different than the old ones, right? | 19:43 |
| JayF | devananda: I'd argue that the class/method structure impacts performance, and that by changing where things are, you'd be changing performance, and would *want* it exposed as a new metric | 19:44 |
| devananda | JayF: that's orthogonal. we could change the content of these functions significantly while keeping the names the same, too | 19:44 |
| JayF | Sure; but like what I'm trying to say is | 19:44 |
| devananda | JayF: what are we measuring? the class/function structure, or the time it takes to do $thing ? | 19:44 |
| JayF | any refactor would change the way the work is organized | 19:45 |
| JayF | right? | 19:45 |
| devananda | (where $thing is an operator-understandable unit of work) | 19:45 |
| JayF | so if we removed the idea of AgentDeploy.prepare, and split prepare up into 5 pieces, then AgentDeploy.prepare ceases to exist as a metric | 19:45 |
| devananda | for instance, deploy_has_started() isn't an operator-facing unit of work | 19:45 |
| devananda | it's a ping-check to the ramdisk, basically | 19:45 |
| devananda | hmm | 19:46 |
| JayF | Like the question I'd want to ask metrics are things like | 19:46 |
| JayF | "What takes the longest in my Agent deploy driver?" | 19:46 |
| JayF | and dig from there | 19:46 |
| JayF | And seeing, for instance, that all the methods that ping out to the ramdisk take a long time, that'd an indicator as to performance | 19:46 |
| JayF | I don't think the metrics alone are ever worthwhile; metrics + devdocs (generated from docstrings) are, metrics + code being obviously the most useful | 19:47 |
| devananda | ahh - I see your use case now | 19:47 |
| devananda | thanks | 19:47 |
| devananda | *thanks for reminding me :) | 19:47 |
| JayF | For troubleshooting, I see things like notifications and logs as more useful | 19:47 |
| JayF | this is for finding things you wouldn't know to troubleshoot otherwise | 19:47 |
| JayF | if that makes sense? | 19:47 |
| devananda | yup | 19:47 |
| devananda | I was just looking at a notifications patch and superimposed that use case on this review | 19:48 |
| devananda | my bad :) | 19:48 |
| JayF | like I'd def want to see something like | 19:48 |
| JayF | my meta metrics say deployments take 20s longer on average when I went from 6.0.0 -> 7.0.0 | 19:48 |
| JayF | then you go looking at metrics/graphs to see where that extra time is coming from | 19:48 |
| JayF | this use case makes a hell of a lot more sense when you think about it for the power/management drivers though/ tbh | 19:49 |
| devananda | JayF: though the actual metrics names may well have changed from 6 -> 7 | 19:49 |
| JayF | sure; but I'm not sure there's a way to avoid that if it gets reordered around | 19:49 |
| *** fxpester|2 has quit IRC | 19:49 | |
| JayF | you're right; but I think it's an unavoidable side effect of having these kind of fine-grained metrics | 19:49 |
| JayF | in fact; it's something to consider for release notes if/when someone changes something that considerably, for operators to note metrics may have moved | 19:50 |
| devananda | JayF: indeed | 19:50 |
| devananda | this is the first release with metrics, but going forward, we should think about that | 19:50 |
| JayF | yes; very much so | 19:50 |
| JayF | especially if we moved a whole class | 19:50 |
| JayF | b/c if you're feeding it to graphite, and we changed the name of like 100 metrics | 19:50 |
| JayF | that would have an appreciable impact on graphite performance and disk usage | 19:50 |
| *** dprince has quit IRC | 19:51 | |
| JayF | <aside> I talked to some folks inside Rackspace, our Cloud Metrics is driven by an open source project called "blueflood", and we might get a contribution to ironic-lib soon with a blueflood backend for metrics \o/ </aside> | 19:51 |
| JayF | (blueflood is also used by square, and a couple of other big operators to store their metrics) | 19:52 |
| devananda | :) | 19:53 |
| JayF | FWIW, despite all ^^ that, that doesn't mean I'm aganist NOT using class.method; just that it's easy/obvious, and I'm not sure there's a better naming scheme to use | 19:54 |
| devananda | you're right - the goal of these is to allow them to be graphed for performance insights | 19:54 |
| devananda | someone digging that deeply into a performance problem will want to quickly identify the method that's slow | 19:54 |
| devananda | not have to disambiguate some string we've decided means something to them | 19:55 |
| JayF | I'm kinda more excited about metrics in the agent than I am in Ironic too | 19:55 |
| JayF | b/c with our downstream hardware manager doing so much with hardware | 19:56 |
| JayF | I wanna get that data fed in and start asking questinos like "does taking longer to erase disks predict hardware failure?" | 19:56 |
| JayF | similar with ipmi calls | 19:56 |
| devananda | darn - it's TC meeting time and I forgot to go get lunch :-/ | 19:56 |
| *** jrist has joined #openstack-ironic | 19:58 | |
| JayF | devananda: ...while you're thinking about metrics; https://review.openstack.org/#/c/301919/16 is pretty simple too :) | 20:02 |
| rajinir | thiagop:sambetts: I made great progress, pxe boot happens, DHCP traffic goes to and back. IP gets assigned too. Next step, the tftp which times out. | 20:05 |
| *** RomanenkoK has joined #openstack-ironic | 20:11 | |
| *** RomanenkoK has quit IRC | 20:14 | |
| openstackgerrit | Stephanie Miller proposed openstack/python-ironicclient: Add volume_target support https://review.openstack.org/285233 | 20:19 |
| *** hj-hp has joined #openstack-ironic | 20:20 | |
| *** hj-hp has quit IRC | 20:23 | |
| *** hj-hpe has joined #openstack-ironic | 20:23 | |
| *** RomanenkoK has joined #openstack-ironic | 20:26 | |
| *** hj-hpe has quit IRC | 20:27 | |
| *** hj-hpe has joined #openstack-ironic | 20:27 | |
| *** e0ne has quit IRC | 20:32 | |
| vgadiraj | when i run "tox -epy27" and then later want to run "tox -epy35", I'm getting this error "db type could not be determined | 20:33 |
| vgadiraj | ERROR: InvocationError: '/home/ubuntu/ironic/.tox/py35/bin/ostestr'" | 20:33 |
| vgadiraj | the workaround for this is removing the file in .testrepository/times.dbm\ | 20:34 |
| vgadiraj | Is this something that should be a note in our developer guide? | 20:34 |
| *** daemontool has joined #openstack-ironic | 20:35 | |
| *** openstackgerrit_ has joined #openstack-ironic | 20:35 | |
| *** openstackgerrit_ has quit IRC | 20:36 | |
| openstackgerrit | Stephanie Miller proposed openstack/python-ironicclient: Add volume_connector support https://review.openstack.org/214786 | 20:36 |
| openstackgerrit | Stephanie Miller proposed openstack/python-ironicclient: Add volume_target support https://review.openstack.org/285233 | 20:38 |
| *** bvandewa has quit IRC | 20:38 | |
| *** bvandewa has joined #openstack-ironic | 20:40 | |
| openstackgerrit | Stephanie Miller proposed openstack/python-ironicclient: Add volume_target support https://review.openstack.org/285233 | 20:41 |
| openstackgerrit | Stephanie Miller proposed openstack/python-ironicclient: Add volume_connector support https://review.openstack.org/214786 | 20:42 |
| *** daemontool has quit IRC | 20:42 | |
| *** ametts has quit IRC | 20:44 | |
| jroll | vgadiraj: it wouldn't hurt, I guess | 20:46 |
| openstackgerrit | Stephanie Miller proposed openstack/python-ironicclient: Add volume_target support https://review.openstack.org/285233 | 20:47 |
| rajinir | devstack-gate question, the VM is coming up, DHCP'ing but never hitting the TFTP server , the tftp doesn't seem to be running on the VM. Can anyone help? | 20:47 |
| *** fragatina has joined #openstack-ironic | 20:50 | |
| *** fragatina has quit IRC | 20:50 | |
| *** fragatina has joined #openstack-ironic | 20:51 | |
| Sukhdev | jroll : saw your note about nova patch - great news!!! | 20:53 |
| jroll | :) | 20:55 |
| *** rama_y has quit IRC | 20:56 | |
| *** bvandewa has quit IRC | 20:57 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add baremetal port list command to OSC plugin https://review.openstack.org/346722 | 21:03 |
| *** jmiu|lunch is now known as jmiu | 21:06 | |
| *** bvandewa has joined #openstack-ironic | 21:07 | |
| *** jcoufal_ has quit IRC | 21:08 | |
| *** mbound has joined #openstack-ironic | 21:11 | |
| *** joprovost has quit IRC | 21:12 | |
| openstackgerrit | Xavier proposed openstack/python-oneviewclient: Add auditing instrumentation for python-oneviewclient https://review.openstack.org/337281 | 21:18 |
| openstackgerrit | Chris Krelle proposed openstack/bifrost: Initial work to add ISC-DHCP https://review.openstack.org/343947 | 21:20 |
| *** bvandewa has quit IRC | 21:20 | |
| *** bvandewa has joined #openstack-ironic | 21:27 | |
| *** PollyZ has quit IRC | 21:28 | |
| mrda | Morning Ironic | 21:29 |
| *** PollyZ has joined #openstack-ironic | 21:30 | |
| JayF | o/ | 21:30 |
| jroll | heya mrda | 21:30 |
| *** rama_y has joined #openstack-ironic | 21:30 | |
| mrda | o/ | 21:30 |
| thiagop | morning mrda | 21:30 |
| RomanenkoK | Hi! | 21:30 |
| *** baoli has quit IRC | 21:31 | |
| NobodyCam | morning mrda | 21:32 |
| *** PollyZ has quit IRC | 21:34 | |
| *** fragatina has quit IRC | 21:38 | |
| mrda | o/ | 21:43 |
| RomanenkoK | I noticed here a strange layout of driver-related commands http://specs.openstack.org/openstack/ironic-specs/specs/not-implemented/ironicclient-osc-plugin.html#openstack-baremetal-driver | 21:49 |
| *** hj-hpe has quit IRC | 21:50 | |
| RomanenkoK | i mean it for commands: •openstack baremetal driver show <driver_name> | 21:50 |
| RomanenkoK | •openstack baremetal driver show properties <driver_name> | 21:50 |
| RomanenkoK | •openstack baremetal driver show passthru <driver_name> | 21:50 |
| RomanenkoK | Second and third look like completion of first. I did not find any another example of this case in other OSC commands | 21:52 |
| JayF | RomanenkoK: you're suggesting it should be openstack baremetal driver show <name> properties? | 21:55 |
| openstackgerrit | Alexander Bashmakov proposed openstack/ironic: Rewrite devstack/tools/ironic/scripts/create-node.sh in Python. https://review.openstack.org/346832 | 22:00 |
| RomanenkoK | @JayF, possibly it should be something like: baremetal driver show, baremetal driver properties, baremetal driver passthru ..etc | 22:04 |
| JayF | RomanenkoK: ah, I gotcha | 22:04 |
| JayF | RomanenkoK: I'm not an expert on OSC or the UX thereof so I can't make a statement one way or the other, but I understand the concern now | 22:05 |
| RomanenkoK | otherwise "properties" and "passthru" will be accepted as driver name for "driver show" command | 22:05 |
| JayF | RomanenkoK: I'd suggest filing a bug about it against python-ironicclient, that'll get it seen by the right folks | 22:07 |
| RomanenkoK | thanks, i`ll do it :) | 22:07 |
| openstackgerrit | David Lenwell proposed openstack/ironic: Add REST API for volume connector and volume target operation https://review.openstack.org/214586 | 22:09 |
| abashmak | Hi JayF | 22:10 |
| abashmak | https://review.openstack.org/346832/ | 22:11 |
| JayF | abashmak: yeah; I looked at that and ran the experimental jobs on it | 22:12 |
| abashmak | JayF: I'm new to Gerrit/Jenkins, was going to ask you to show me how to do it | 22:13 |
| JayF | just the comment 'check experimental' does it | 22:13 |
| JayF | it'll come in as a separate comment from jenkins | 22:13 |
| JayF | (the actual comments from jenkins are hidden by default, but you can see them by clicking "toggle ci" at the bottom) | 22:13 |
| *** bfournie has quit IRC | 22:13 | |
| abashmak | so by virture of your posting that comment, Jenkins will pick it up? | 22:14 |
| JayF | exactly | 22:14 |
| JayF | and my comment more or less indicates that as a core I'd rather it not land until all the jobs (inlcuding the ones we expect to pass in experimental) are passing | 22:14 |
| JayF | just because they're more likely to exercise that script in weird ways | 22:14 |
| jroll | what experimental jobs do we have that we expect to pass, just curious | 22:15 |
| JayF | don't we have the multi-tenant job? | 22:15 |
| * JayF may be out of date there, memory-wise | 22:15 | |
| JayF | I honestly never remember until I see the list :x | 22:15 |
| jroll | that's non-voting in check now | 22:17 |
| JayF | aha | 22:17 |
| JayF | that's the only one I was thinking of specifically | 22:17 |
| JayF | just sorta a natural reaction to run exp jobs and see how they go when a big devstack change is going | 22:18 |
| jroll | totally | 22:18 |
| abashmak | I also agree :) | 22:18 |
| *** RomanenkoK has quit IRC | 22:26 | |
| *** hoangcx has quit IRC | 22:28 | |
| *** sdake has quit IRC | 22:40 | |
| *** mbound has quit IRC | 22:41 | |
| *** hoangcx has joined #openstack-ironic | 22:47 | |
| *** mbound has joined #openstack-ironic | 22:49 | |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Introduce new RPCs to support volume target operations https://review.openstack.org/285220 | 22:49 |
| *** david-lyle has quit IRC | 22:50 | |
| *** david-lyle has joined #openstack-ironic | 22:51 | |
| openstackgerrit | Merged openstack/ironic: Minor docstring and unittests fixes for IPMIConsole https://review.openstack.org/349400 | 22:51 |
| *** michauds has quit IRC | 22:52 | |
| *** mbound has quit IRC | 22:54 | |
| *** lucas-dinner has quit IRC | 22:54 | |
| *** wajdi_ has joined #openstack-ironic | 22:54 | |
| openstackgerrit | Varun Gadiraju proposed openstack/ironic: Fix for "db type could not be determined" error message https://review.openstack.org/350302 | 22:58 |
| *** wajdi has quit IRC | 22:58 | |
| *** wajdi_ has quit IRC | 22:59 | |
| *** lucasagomes has joined #openstack-ironic | 22:59 | |
| *** syed_ has quit IRC | 23:02 | |
| openstackgerrit | Varun Gadiraju proposed openstack/ironic: Fix for "db type could not be determined" error message https://review.openstack.org/350302 | 23:03 |
| mrda | pas-ha: Hey Pavlo, can you please reverify that https://bugs.launchpad.net/nova/+bug/1544195 is still a problem? Nova think they might have fixed it, and it'd be great to see if it's still a problem. Thanks! | 23:13 |
| openstack | Launchpad bug 1544195 in heat "User can not provision ironic node via nova when providing pre-created port" [Undecided,Confirmed] | 23:13 |
| *** sdake has joined #openstack-ironic | 23:15 | |
| *** sdake_ has joined #openstack-ironic | 23:17 | |
| openstackgerrit | Alexander Bashmakov proposed openstack/ironic: Rewrite devstack/tools/ironic/scripts/create-node.sh in Python. https://review.openstack.org/346832 | 23:18 |
| *** sdake has quit IRC | 23:20 | |
| *** fumie has joined #openstack-ironic | 23:20 | |
| *** sdake_ is now known as sdake | 23:21 | |
| *** bfournie has joined #openstack-ironic | 23:29 | |
| *** sabeen has quit IRC | 23:32 | |
| *** thiagop has quit IRC | 23:39 | |
| *** causten has quit IRC | 23:40 | |
| *** rpioso has quit IRC | 23:49 | |
| *** hoangcx2 has joined #openstack-ironic | 23:51 | |
| *** hoangcx has quit IRC | 23:53 | |
| *** dprince has joined #openstack-ironic | 23:56 | |
| *** anush_ has quit IRC | 23:57 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Timing metrics: pxe boot and iscsi deploy driver https://review.openstack.org/350317 | 23:58 |
| JayF | devananda: jroll ^ in case you feel like landing another one of these tomorrow morning :D | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!