| *** openstack has joined #openstack-ironic | 00:04 | |
| *** sdake has quit IRC | 00:06 | |
| *** sdake has joined #openstack-ironic | 00:09 | |
| *** zzzeek has quit IRC | 00:10 | |
| *** rbudden has quit IRC | 00:10 | |
| *** zzzeek has joined #openstack-ironic | 00:14 | |
| *** vishwana_ has quit IRC | 00:14 | |
| *** harlowja has joined #openstack-ironic | 00:19 | |
| *** baoli has joined #openstack-ironic | 00:25 | |
| *** sdake_ has joined #openstack-ironic | 00:28 | |
| *** sdake has quit IRC | 00:30 | |
| *** jaybeale has quit IRC | 00:32 | |
| openstackgerrit | Stephanie Miller proposed openstack/bifrost: Make boolean usage consistent across playbooks https://review.openstack.org/321310 | 00:33 |
|---|---|---|
| *** phschwartz has quit IRC | 00:34 | |
| *** hoangcx has joined #openstack-ironic | 00:34 | |
| openstackgerrit | Bruno Cornec proposed openstack/bifrost: Fix Bug #1583539 - rpm part https://review.openstack.org/321311 | 00:34 |
| openstack | bug 1583539 in bifrost "Missing dependencies in documentation and scripts for a CentOS 7 install of bifrost" [Undecided,New] https://launchpad.net/bugs/1583539 | 00:34 |
| *** YiboCai has joined #openstack-ironic | 00:37 | |
| *** penick has quit IRC | 00:42 | |
| *** vishwanathj has joined #openstack-ironic | 00:44 | |
| *** chlong has quit IRC | 00:46 | |
| *** openstackstatus has quit IRC | 00:46 | |
| *** thrash is now known as thrash|g0ne | 00:48 | |
| *** sdake_ has quit IRC | 00:50 | |
| *** openstack has joined #openstack-ironic | 00:55 | |
| *** rloo has joined #openstack-ironic | 01:02 | |
| *** piet has quit IRC | 01:10 | |
| *** phuongnh has joined #openstack-ironic | 01:13 | |
| *** Egyptian[Home] has joined #openstack-ironic | 01:15 | |
| *** piet has joined #openstack-ironic | 01:18 | |
| *** wuhg has joined #openstack-ironic | 01:22 | |
| *** praneshp has joined #openstack-ironic | 01:23 | |
| *** ChrisAusten has joined #openstack-ironic | 01:24 | |
| *** sdake has quit IRC | 01:25 | |
| *** piet has quit IRC | 01:31 | |
| *** vmud213 has joined #openstack-ironic | 01:33 | |
| *** Egyptian[Home] has quit IRC | 01:41 | |
| *** Egyptian[Home] has joined #openstack-ironic | 01:42 | |
| *** jaybeale has quit IRC | 01:49 | |
| *** openstackgerrit has quit IRC | 01:49 | |
| *** wajdi_ has joined #openstack-ironic | 01:50 | |
| *** Egyptian[Home] has quit IRC | 01:51 | |
| *** wajdi has quit IRC | 01:54 | |
| *** wajdi_ has quit IRC | 01:54 | |
| *** openstackgerrit has joined #openstack-ironic | 01:57 | |
| *** chlong has joined #openstack-ironic | 01:59 | |
| *** rloo has quit IRC | 02:00 | |
| *** sabeen has quit IRC | 02:04 | |
| *** sabeen has joined #openstack-ironic | 02:05 | |
| *** Egyptian[Home] has joined #openstack-ironic | 02:06 | |
| *** rbudden has joined #openstack-ironic | 02:17 | |
| *** sdake has joined #openstack-ironic | 02:19 | |
| *** vmud213 has quit IRC | 02:19 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Gate on ironic.conf.sample generation https://review.openstack.org/238837 | 02:19 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Gate on ironic.conf.sample generation https://review.openstack.org/238837 | 02:20 |
| *** sabeen1 has joined #openstack-ironic | 02:22 | |
| *** sabeen has quit IRC | 02:23 | |
| *** openstack has joined #openstack-ironic | 02:40 | |
| *** vishwanathj has quit IRC | 02:41 | |
| *** rbudden has quit IRC | 02:44 | |
| *** yuanying has quit IRC | 02:50 | |
| *** sdake_ has joined #openstack-ironic | 03:00 | |
| *** sdake has quit IRC | 03:02 | |
| *** sdake_ is now known as sdake | 03:05 | |
| *** vishwanathj has joined #openstack-ironic | 03:06 | |
| *** baoli has quit IRC | 03:26 | |
| *** yuanying has joined #openstack-ironic | 03:30 | |
| *** kromanenko__ has joined #openstack-ironic | 03:34 | |
| *** Romanenko_K has joined #openstack-ironic | 03:34 | |
| *** kromanenko_ has joined #openstack-ironic | 03:34 | |
| *** links has joined #openstack-ironic | 03:34 | |
| *** phuongnh has quit IRC | 03:35 | |
| *** phuongnh has joined #openstack-ironic | 03:36 | |
| *** yuanying has quit IRC | 03:45 | |
| *** piet has joined #openstack-ironic | 03:46 | |
| *** yuanying has joined #openstack-ironic | 03:47 | |
| *** Egyptian[Home] has quit IRC | 03:48 | |
| *** Egyptian[Home] has joined #openstack-ironic | 03:49 | |
| *** vmud213 has joined #openstack-ironic | 04:03 | |
| *** piet has quit IRC | 04:03 | |
| *** chlong has quit IRC | 04:15 | |
| *** Egyptian[Home] has quit IRC | 04:24 | |
| *** baoli has joined #openstack-ironic | 04:27 | |
| *** chlong has joined #openstack-ironic | 04:34 | |
| *** wajdi has joined #openstack-ironic | 04:51 | |
| *** ChubYann has quit IRC | 05:01 | |
| *** sacharya has quit IRC | 05:01 | |
| *** stendulker has joined #openstack-ironic | 05:12 | |
| *** _vsaienko has joined #openstack-ironic | 05:19 | |
| *** _vs has joined #openstack-ironic | 05:19 | |
| *** rama_y has joined #openstack-ironic | 05:22 | |
| *** yolanda_ has quit IRC | 05:23 | |
| *** _vsaienko has quit IRC | 05:25 | |
| *** _vs has quit IRC | 05:25 | |
| *** _vsaienko has joined #openstack-ironic | 05:26 | |
| *** _vs has joined #openstack-ironic | 05:27 | |
| *** ishant has joined #openstack-ironic | 05:28 | |
| *** bnemec has quit IRC | 05:30 | |
| *** bnemec has joined #openstack-ironic | 05:31 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic-specs: Enhance Power Interface for Soft Power Off and Inject NMI https://review.openstack.org/186700 | 05:34 |
| *** baoli has quit IRC | 05:39 | |
| *** bnemec has quit IRC | 05:44 | |
| *** chlong has quit IRC | 05:47 | |
| *** rama_y has quit IRC | 05:47 | |
| *** _vs has quit IRC | 05:51 | |
| *** _vsaienko has quit IRC | 05:51 | |
| *** bnemec has joined #openstack-ironic | 05:52 | |
| *** yolanda_ has joined #openstack-ironic | 05:52 | |
| *** sdake has quit IRC | 05:54 | |
| *** rcernin has joined #openstack-ironic | 05:54 | |
| *** chlong has joined #openstack-ironic | 05:59 | |
| *** sdake has joined #openstack-ironic | 05:59 | |
| *** sacharya has joined #openstack-ironic | 06:01 | |
| *** itamarl has joined #openstack-ironic | 06:03 | |
| *** sacharya has quit IRC | 06:06 | |
| *** bnemec has quit IRC | 06:12 | |
| *** ChrisAusten has quit IRC | 06:12 | |
| *** bnemec has joined #openstack-ironic | 06:14 | |
| pas-ha | morning all | 06:18 |
| *** fragatina has quit IRC | 06:18 | |
| *** moshele has joined #openstack-ironic | 06:20 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add a doc about how to append kernel parameters to boot instances https://review.openstack.org/320848 | 06:21 |
| *** bnemec has quit IRC | 06:23 | |
| *** yolanda_ has quit IRC | 06:29 | |
| *** sdake has quit IRC | 06:30 | |
| *** parlos has joined #openstack-ironic | 06:31 | |
| *** yolanda has joined #openstack-ironic | 06:31 | |
| openstackgerrit | Sharat Sharma proposed openstack/ironic: Devstack: configure Ironic to use SSL https://review.openstack.org/257753 | 06:35 |
| *** bnemec has joined #openstack-ironic | 06:36 | |
| *** vishwanathj has quit IRC | 06:37 | |
| *** _vsaienko has joined #openstack-ironic | 06:45 | |
| *** _vs has joined #openstack-ironic | 06:45 | |
| *** hoangcx has quit IRC | 06:51 | |
| *** openstack has joined #openstack-ironic | 06:55 | |
| *** parlos has quit IRC | 07:00 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Make sure create_ovs_taps creates unique taps https://review.openstack.org/319101 | 07:03 |
| *** cfarquhar has quit IRC | 07:07 | |
| *** greghaynes has quit IRC | 07:08 | |
| *** sdake has quit IRC | 07:13 | |
| *** parlos has joined #openstack-ironic | 07:15 | |
| *** sdake has joined #openstack-ironic | 07:16 | |
| *** cfarquhar has joined #openstack-ironic | 07:20 | |
| *** cfarquhar has quit IRC | 07:20 | |
| *** cfarquhar has joined #openstack-ironic | 07:20 | |
| *** greghaynes has joined #openstack-ironic | 07:21 | |
| *** sdake has quit IRC | 07:21 | |
| *** athomas has joined #openstack-ironic | 07:22 | |
| *** dtantsur|afk is now known as dtantsur | 07:36 | |
| dtantsur | Morning Ironic | 07:36 |
| vsaienko | morning dtantsur | 07:36 |
| *** hoangcx has joined #openstack-ironic | 07:39 | |
| *** ifarkas has joined #openstack-ironic | 07:42 | |
| *** sdake has joined #openstack-ironic | 07:44 | |
| milan | morning Ironic | 07:44 |
| milan | morning dtantsur vsaienko :) | 07:45 |
| *** sdake has quit IRC | 07:49 | |
| vsaienko | milan o/ | 07:49 |
| *** sdake has joined #openstack-ironic | 07:50 | |
| dtantsur | milan, vsaienko, o/ | 07:52 |
| *** athomas has quit IRC | 07:55 | |
| vsaienko | dtantsur I have reviewed your patch about removing machine-id. Have you checked systemd-firstboot utility? It looks like it can change machine specific settings during first boot https://www.freedesktop.org/software/systemd/man/systemd-firstboot.html | 07:57 |
| *** mgoddard has joined #openstack-ironic | 07:58 | |
| dtantsur | vsaienko, no, I haven't. actually I have little clues about this problem, I just got complains that this file should be deleted, so I came up with a quick patch | 07:59 |
| dtantsur | turns out not so easy actually | 07:59 |
| *** zzzeek has quit IRC | 08:00 | |
| *** athomas has joined #openstack-ironic | 08:00 | |
| *** zzzeek has joined #openstack-ironic | 08:01 | |
| vsaienko | I'm interested how this problem is solved on cloud images, and then I found systemd-firstboot utility. On first look it should help | 08:02 |
| *** sacharya has joined #openstack-ironic | 08:03 | |
| *** chlong has quit IRC | 08:04 | |
| dtantsur | vsaienko, interesting. anyway, I don't think I'll have time to come back to this patch any soon | 08:05 |
| *** dtantsur is now known as dtantsur|bbl | 08:07 | |
| *** sacharya has quit IRC | 08:08 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 08:10 |
| openstackgerrit | Merged openstack/ironic: Introduce provision states: AVAILABLE, ENROLL https://review.openstack.org/321204 | 08:17 |
| *** electrofelix has joined #openstack-ironic | 08:18 | |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: [WIP] [POC] Callback from ramdisk support for Ansible driver https://review.openstack.org/310816 | 08:19 |
| *** lucas-dinner is now known as lucasagmes | 08:22 | |
| *** lucasagmes is now known as lucasagomes | 08:22 | |
| *** dmk0202 has joined #openstack-ironic | 08:23 | |
| lucasagomes | morning all | 08:23 |
| *** praneshp has quit IRC | 08:27 | |
| *** MattMan has quit IRC | 08:28 | |
| *** MattMan has joined #openstack-ironic | 08:29 | |
| vmud213 | morning lucasagomes,dtantsur,vsaienko,milan and all | 08:35 |
| milan | morning lucasagomes vmud213 | 08:35 |
| *** derekh has joined #openstack-ironic | 08:39 | |
| *** moshele has quit IRC | 08:44 | |
| *** moshele has joined #openstack-ironic | 08:47 | |
| *** daemontool has joined #openstack-ironic | 08:58 | |
| *** parlos has quit IRC | 08:58 | |
| *** jtomasek_ has joined #openstack-ironic | 09:00 | |
| lucasagomes | milan, vmud213 hello :-) | 09:01 |
| *** jtomasek_ has quit IRC | 09:02 | |
| *** tesseract has joined #openstack-ironic | 09:02 | |
| *** jtomasek_ has joined #openstack-ironic | 09:02 | |
| *** jtomasek_ has quit IRC | 09:08 | |
| *** kromanenko__ has quit IRC | 09:13 | |
| *** watanabe_isao has quit IRC | 09:13 | |
| *** kromanenko_ has quit IRC | 09:13 | |
| *** Romanenko_K has quit IRC | 09:14 | |
| *** sambetts|afk is now known as sambetts | 09:18 | |
| sambetts | Morning all :D | 09:18 |
| milan | sambetts, morning :) | 09:19 |
| sambetts | Hi milan | 09:19 |
| milan | guys, any idea why CONF.baremetal.active_timeout in a tempest test wouldn't be left default? I'm getting 196 instead of 300 and cannot locate the source of that :-/ | 09:21 |
| sambetts | milan: thats overridden by tempest | 09:23 |
| sambetts | milan: the BUILD_TIMEOUT value gets set to it | 09:23 |
| milan | sambetts, /me gitgreps | 09:23 |
| sambetts | take a look at devstack/extra.d/80-tempest | 09:24 |
| lucasagomes | sambetts, morning | 09:24 |
| lucasagomes | #pixiesay *sigh* | 09:25 |
| PixieBoots | ʕ•͡ᴥ•ʔ: *sigh* | 09:25 |
| sambetts | lucasagomes: whats up? | 09:25 |
| lucasagomes | sambetts, all is well, you? | 09:26 |
| lucasagomes | the *sigh* was about overriding the config heh | 09:26 |
| sambetts | ah :-P hehe yeah I'm all good, thats something that I had to contend with for my CI local.conf | 09:27 |
| milan | sambetts, seems you're right, it does :-/ strange, this started to manifest itself for me just yesterday without andy (unstack; stack) | 09:28 |
| *** Romanenko_K has joined #openstack-ironic | 09:28 | |
| *** kromanenko__ has joined #openstack-ironic | 09:28 | |
| *** kromanenko_ has joined #openstack-ironic | 09:28 | |
| *** dtantsur|bbl is now known as dtantsur | 09:28 | |
| sambetts | milan: I end up doing: | 09:29 |
| sambetts | [[post-extra|$TEMPEST_CONFIG]] | 09:29 |
| sambetts | [DEFAULT] | 09:29 |
| sambetts | debug = false | 09:29 |
| sambetts | [baremetal] | 09:29 |
| sambetts | power_timeout = 180 | 09:29 |
| sambetts | deploywait_timeout = 180 | 09:29 |
| milan | sambetts, which file is that? | 09:30 |
| sambetts | local.conf | 09:30 |
| milan | devstack/local.conf? | 09:30 |
| dtantsur | milan, iirc vsaienko has patches allowing modifying various tempest timeouts via the job configuration | 09:31 |
| sambetts | yeah, it allows you to override any config file you want as the very last thing stack.sh does | 09:31 |
| milan | sambetts, thanks a lot, I've been getting some gray hairs already | 09:31 |
| milan | esp. since i wasn't able to override it in my setUp test methods while debugging that | 09:32 |
| * milan things devstack + tempest == black magic | 09:33 | |
| sambetts | milan: my entire local.conf -> http://paste.openstack.org/show/505592/ | 09:34 |
| milan | sambetts, thanks | 09:35 |
| sambetts | lucasagomes: I don't know if you've seen, https://review.openstack.org/#/c/314514/, I've just seen your follow up reserve words patch merged | 09:38 |
| lucasagomes | sambetts, will take a look | 09:38 |
| sambetts | has +2/+A | 09:39 |
| *** divya_ has quit IRC | 09:41 | |
| milan | #pixiesay ls -l /etc/tempest: .testr .testrepository ; who would make config "invisible"??? -m angry | 09:42 |
| PixieBoots | ʕᗒᴥᗕʔ՞: ls | 09:42 |
| milan | #pixiesay ls \-l /etc/tempest: .testr .testrepository ; who would make config "invisible"??? -m angry | 09:43 |
| PixieBoots | ʕᗒᴥᗕʔ՞: ls \-l /etc/tempest: .testr .testrepository ; who would make config "invisible"??? | 09:43 |
| lucasagomes | #pixiesay -m meh | 09:43 |
| PixieBoots | ¯\_(ツ)_/¯ | 09:43 |
| *** divya has joined #openstack-ironic | 09:44 | |
| vdrok | morning ironic! | 09:44 |
| milan | morning vdrok ! | 09:44 |
| divya | morning Ironic! | 09:45 |
| vdrok | good morning milan and divya | 09:46 |
| vmud213 | morning vdrok,sambetts | 09:47 |
| milan | vmud213, divya o/ | 09:49 |
| *** parlos has joined #openstack-ironic | 09:50 | |
| vdrok | hi vmud213 | 09:51 |
| aarefiev | morning all! | 09:51 |
| milan | aarefiev, o/ | 09:52 |
| aarefiev | milan: looks like tempest fails due using tempest code - not ironic | 09:52 |
| milan | aarefiev, yeah, black magic :-/ | 09:52 |
| aarefiev | it waits node in nostate I guess | 09:52 |
| milan | aarefiev, my fault | 09:53 |
| vsaienko | morning vmud213, vdrok, sambetts, aarefiev, lucasagomes and all! | 09:53 |
| * milan has to upload another patch | 09:53 | |
| milan | vsaienko, o/ | 09:53 |
| aarefiev | morning vsaienko | 09:53 |
| openstackgerrit | Sam Betts proposed openstack/ironic-inspector: Add generic LLDP processing hook https://review.openstack.org/321082 | 09:53 |
| vmud213 | morning vsaienko | 09:54 |
| dtantsur | hey aarefiev, vdrok | 09:54 |
| vdrok | morning aarefiev vsaienko and dtantsur | 09:55 |
| milan | aarefiev, also, me not sure this was merged when the test ran: https://review.openstack.org/#/c/321204/1 | 09:55 |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Consolidate IPA configuration into a config module https://review.openstack.org/320867 | 09:59 |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Support LLDP data as part of interfaces in inventory https://review.openstack.org/320584 | 09:59 |
| lucasagomes | vsaienko, good morning! | 09:59 |
| vdrok | morning lucasagomes | 09:59 |
| *** YiboCai has quit IRC | 10:00 | |
| sambetts | vsaienko: In response to this: https://review.openstack.org/#/c/318479/8 I've created this: https://review.openstack.org/#/c/320040/ | 10:01 |
| sambetts | vsaienko: Hopefully it fixes the problem instead of just working around it | 10:01 |
| *** sacharya has joined #openstack-ironic | 10:04 | |
| milan | aarefiev, it's because I'm using tempest.scenario.manager.BaremetalManager but should use ironic_tempest_plugin managers instead :-/ | 10:04 |
| aarefiev | milan: looks like, because in logs I see node became available | 10:05 |
| aarefiev | but it continues checks | 10:05 |
| sambetts | milan: how are you running the tempest tests? tox -e all_plugin -- ironic | 10:05 |
| milan | aarefiev, yeah | 10:05 |
| milan | sambetts, yeah, tox -e all_plugin -- inspector | 10:06 |
| sambetts | milan: ah right | 10:06 |
| vdrok | morning sambetts | 10:06 |
| sambetts | o/ vdrok | 10:06 |
| milan | aarefiev, I'm patching it, anyways I've missed one NOSTATE in node_cleanup | 10:07 |
| milan | should be [AVAILABLE, NOSTATE] instead | 10:08 |
| vdrok | lucasagomes: so you think we should not remove parameters and retry at all, or maybe to log that the port changed to a default in the log warning message? or retry only in case there is no custom port set? (regarding https://review.openstack.org/#/c/319183/ and the next one in chain) | 10:09 |
| *** sacharya has quit IRC | 10:09 | |
| lucasagomes | vdrok, I've been thinking... Maybe the way to do it IMO, is to check if it's the default port. If so, do not pass it via the ramdisk parameters | 10:09 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 10:10 |
| lucasagomes | in case it's a different value we should pass and if it's an old ramdisk it will fail. Then we log something saying it's not supported | 10:10 |
| vdrok | yeah, and in case of wipe metadata, leave things as is | 10:11 |
| aarefiev | milan: yeah, nice, also I don't like ironic checks node status with interval=1 | 10:11 |
| milan | aarefiev, where you see it? | 10:12 |
| aarefiev | milan: it's hard to read logs | 10:12 |
| * milan would fix that | 10:12 | |
| milan | aarefiev, +1 | 10:12 |
| aarefiev | milan: looks like default is used https://github.com/openstack/ironic/blob/master/ironic_tempest_plugin/tests/scenario/baremetal_manager.py#L78 | 10:13 |
| milan | aarefiev, right, should be part of config group baremetal IMHO | 10:14 |
| milan | and all the wait* methods could possibly propagate a reasonable default, too | 10:14 |
| openstackgerrit | Merged openstack/ironic: Follow-up patch of 8e5e69869df476788b3ccf7e5ba6c2210a98fc8a https://review.openstack.org/314149 | 10:15 |
| milan | such as wait_provision_state(...., interval=10 | 10:15 |
| sambetts | vsaienko: I'm looking at https://review.openstack.org/#/c/319101/7 I wonder if we would do better to just hook our VMs into the OVS_PHYSICAL_BRIDGE / PUBLIC_BRIDGE instead of each of the networks specifically | 10:15 |
| milan | it's 30 actually and that one does it :D | 10:15 |
| milan | wrong example | 10:15 |
| milan | aarefiev, ignore my last comments :D | 10:16 |
| * milan facepalm | 10:16 | |
| lucasagomes | vdrok, yeah... At least that way we are honest with what have been requested | 10:16 |
| sambetts | vsaienko: because all networks (if routed external, which they should be) will go through the public bridge | 10:16 |
| aarefiev | milan: :) | 10:17 |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Support LLDP data as part of interfaces in inventory https://review.openstack.org/320584 | 10:31 |
| vsaienko | sambetts in general it should be fixed by using Flat network | 10:31 |
| vsaienko | which is only one officially supported case by Ironic :) | 10:31 |
| vsaienko | I'm going to rebase multitenancy patches, if anyone has any comments please add them | 10:34 |
| vsaienko | sambetts: just hook our VMs into the OVS_PHYSICAL_BRIDGE / PUBLIC_BRIDGE it will not work, since you have to set a tags for traffic from/to VMs | 10:37 |
| openstackgerrit | Vaishnavi proposed openstack/ironic: Config variable to configure [glance] section https://review.openstack.org/266803 | 10:46 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add tests for --wrap option for driver-properties output https://review.openstack.org/297077 | 10:49 |
| sambetts | vsaienko: once the traffic leaves through the PUBLIC_BRIDGE its already been NATed and should be untagged | 10:49 |
| *** ptrlv has joined #openstack-ironic | 10:50 | |
| *** yolanda has quit IRC | 10:57 | |
| sambetts | vsaienko: I've been experiementing with running a flat network in devstack with VMs, and its a difficult problem to solve because a jenkins slave only has 1 nic, and turning that nic into the NIC used by neutron for the flat network breaks the hosts networking on that nic :( | 10:59 |
| sambetts | which then means that jenkins can't talk to the slave any more | 10:59 |
| vsaienko | sambetts: we may create additional interface | 11:00 |
| vsaienko | there is no need to use real network, virtual interface should be enough | 11:00 |
| *** Egyptian[Home] has joined #openstack-ironic | 11:01 | |
| vsaienko | sambetts: I have a question about lldp patches. Why don't use lldpd daemon to collect lldp data, why we need to dump packets and parse them? | 11:01 |
| sambetts | can we create a virtual interface not in ovs and not in linux bridge thats not tied to a real NIC? | 11:01 |
| sambetts | vsaienko: Not sure, I'm just reusing the code that already exists for retreiving the LLDP packets | 11:02 |
| vsaienko | sambetts: sudo ip link set name eth10 dev dummy0 | 11:03 |
| sambetts | vsaienko: I've used that code in my downstream work and I know it works, so I havn't fixed what isn't broken | 11:03 |
| sambetts | vsaienko: Ah dummy links | 11:03 |
| sambetts | vsaienko: we need an extra package for that though right? | 11:03 |
| *** yolanda has joined #openstack-ironic | 11:04 | |
| sambetts | vsaienko: it could totally work though I think | 11:04 |
| vsaienko | sambetts: lo also should work as for me | 11:04 |
| sambetts | vsaienko: the VMs won't have an internet connection though right? does that matter? | 11:04 |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Tempest: add basic test https://review.openstack.org/293926 | 11:05 |
| vsaienko | they will have connection via public neutron network | 11:05 |
| sambetts | vsaienko: with flat network there is no public neutron network | 11:05 |
| sambetts | vsaienko: there is only one network | 11:05 |
| sambetts | the flat network | 11:05 |
| vsaienko | might be it is not created by default in devstack, but it is possible to have a public and flat networks at the same time | 11:06 |
| sambetts | vsaienko: right, enabling provider/flat networking in devstack disables L3 so you can't route through another network | 11:08 |
| *** links has quit IRC | 11:08 | |
| sambetts | vsaienko: not sure if we need internet for our tests though right? | 11:11 |
| vsaienko | sambetts: it will be needed for smoke tests | 11:12 |
| vsaienko | sambetts: I had a patch that did the trick when played with real hardware servers | 11:12 |
| sambetts | vsaienko: why? Do they test pinging an IP on the internet? | 11:12 |
| vsaienko | sambetts: there are tests that assigns floating IP to instance | 11:13 |
| *** _vs has quit IRC | 11:13 | |
| *** _vsaienko has quit IRC | 11:13 | |
| sambetts | if neutron L3 is disabled we can't do that right? | 11:13 |
| vsaienko | I don't think that l3 is disabled | 11:13 |
| *** stendulker_ has joined #openstack-ironic | 11:13 | |
| sambetts | like I said to support flat networking in devstack you have to disable neutron l3 | 11:14 |
| vsaienko | https://github.com/openstack-dev/devstack/blob/master/lib/neutron_plugins/services/l3#L206 | 11:14 |
| vsaienko | flat network is configured by l3 plugin, it can't be disabled it makes no sense | 11:14 |
| vsaienko | sambetts: after multitenancy is merged I have in plans to switch devstack setup to flat networking | 11:15 |
| sambetts | vsaienko: :/ this has all changed since like 2 months ago :/ | 11:16 |
| *** e0ne has joined #openstack-ironic | 11:16 | |
| *** e0ne has quit IRC | 11:16 | |
| vsaienko | at the moment it will take a lot of time | 11:16 |
| *** parlos has quit IRC | 11:16 | |
| *** stendulker has quit IRC | 11:17 | |
| sambetts | vsaienko: I still think that for now we can simplify -> https://review.openstack.org/#/c/319101/ by putting all our VMs onto br-ex right? | 11:18 |
| *** e0ne has joined #openstack-ironic | 11:18 | |
| sambetts | :/ not sure | 11:18 |
| sambetts | I guess the DHCP etc does work like that :( | 11:19 |
| vsaienko | sambetts: I think that it is the best way | 11:19 |
| vsaienko | switching to flat network will take a lot of time, since it is huge change in testing environment setup | 11:20 |
| sambetts | vsaienko: yeah, definatly, and I think we need to do it soon, but leave that patch as it is for now | 11:20 |
| vsaienko | I doubt that jroll would approve it. As for me we can add a new job with flat networking, make sure that it is stable, and only than switch to it completely | 11:21 |
| *** stendulker_ has quit IRC | 11:21 | |
| jroll | ¯\_(ツ)_/¯ | 11:22 |
| vsaienko | morning jroll | 11:23 |
| jroll | morning vsaienko, sambetts :) | 11:23 |
| sambetts | Hey | 11:23 |
| jroll | vsaienko: "Why don't use lldpd daemon" the coreos version of the ramdisk just runs a single process, no init system, so it's... annoying to run lldpd | 11:24 |
| jroll | or that's why we did it to start | 11:24 |
| jroll | hashtag containers | 11:24 |
| jroll | now that we have the code, no sense in an extra daemon :) | 11:24 |
| *** ishant has quit IRC | 11:25 | |
| vsaienko | jroll got it | 11:27 |
| dtantsur | morning jroll | 11:28 |
| vsaienko | jroll, I saw thread about bonding support in Neutron. I'm interested how it will affect multitenancy? | 11:28 |
| jroll | vsaienko: well, the thread said they aren't doing it | 11:29 |
| jroll | sambetts was concerned both teams were doing it | 11:29 |
| jroll | but they aren't, so long live portgroups | 11:29 |
| *** Egyptian[Home] has quit IRC | 11:30 | |
| *** wendar has quit IRC | 11:31 | |
| *** wendar has joined #openstack-ironic | 11:31 | |
| sambetts | \o/ seems like it, wasn't it jaypipes or the tuba guy that came in here asking about bonding for virt? | 11:32 |
| vsaienko | jroll, yeah I know that they working on vlan aware VMs https://specs.openstack.org/openstack/neutron-specs/specs/liberty/vlan-aware-vms.html | 11:32 |
| vdrok | morning jroll | 11:33 |
| jroll | vsaienko: yeah, trunk ports are a thing that's happening, but not bonding I guess | 11:33 |
| jroll | sambetts: no clue and I refuse to grep my logs for "bond" | 11:34 |
| jroll | :D | 11:34 |
| sambetts | haha yeah :-P | 11:34 |
| sambetts | vsaienko: it shouldn't affect us at all, because its not happening, and our next steps are all based around the VLAN aware VMs stuff: https://review.openstack.org/#/c/277853/ | 11:36 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Use name randomizer from tempest https://review.openstack.org/307904 | 11:39 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Use name randomizer from tempest https://review.openstack.org/307904 | 11:39 |
| vsaienko | sambetts: I need to read both specs carefully, but my feeling that we should rely on neutron spec and implementation | 11:43 |
| sambetts | vsaienko: which it does :) | 11:43 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Remove links to github for OpenStack things https://review.openstack.org/321491 | 11:45 |
| jroll | pretty easy docs patch | 11:46 |
| *** hoangcx has quit IRC | 11:48 | |
| *** wajdi has quit IRC | 11:51 | |
| *** wajdi has joined #openstack-ironic | 11:52 | |
| *** Goneri has quit IRC | 11:53 | |
| *** wajdi_ has joined #openstack-ironic | 11:53 | |
| *** phuongnh has quit IRC | 11:56 | |
| *** thrash|g0ne is now known as thrash | 11:56 | |
| *** wajdi has quit IRC | 11:57 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add sanity tests for testing actions with Port https://review.openstack.org/240944 | 11:58 |
| *** sacharya has joined #openstack-ironic | 12:04 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Remove links to github for OpenStack things https://review.openstack.org/321491 | 12:06 |
| *** sacharya has quit IRC | 12:09 | |
| *** trown|outtypewww is now known as trown | 12:10 | |
| openstackgerrit | Mathieu Mitchell proposed openstack/ironic-python-agent: Remove links to github for OpenStack things https://review.openstack.org/321506 | 12:13 |
| mat128 | jroll: equivalent in ipa ^ | 12:13 |
| jroll | <3 | 12:13 |
| *** piet has joined #openstack-ironic | 12:14 | |
| *** dprince has joined #openstack-ironic | 12:19 | |
| jroll | so, this RFE currently depends on a cross-project spec https://bugs.launchpad.net/ironic/+bug/1561100 | 12:24 |
| openstack | Launchpad bug 1561100 in Ironic "[RFE] Centralize config options in Ironic" [Wishlist,In progress] - Assigned to Ramamani Yeleswarapu (ramamani-yeleswarapu) | 12:24 |
| jroll | but that spec has turned into more of "categorize, document better, improve the sample file" | 12:25 |
| jroll | with an optional "centralize config options in code" | 12:25 |
| jroll | most of the code is written (but is blocked on rfe approval and since turned into merge conflicts): https://review.openstack.org/#/q/topic:bug/1561100 | 12:25 |
| jroll | do people have a problem with approving the "centralize" part of this that has been worked on? | 12:26 |
| jroll | dtantsur: lucasagomes: ^ | 12:26 |
| dtantsur | no objections here | 12:26 |
| jroll | cool, should be easy reviewing too | 12:27 |
| jroll | and the oslo config-gen stuff landed yesterday \o/ | 12:27 |
| *** KennethWilke has quit IRC | 12:27 | |
| sambetts | Seems ok to me | 12:27 |
| *** piet has quit IRC | 12:27 | |
| * jroll will wait for lucas opinion but will probably approve that today | 12:27 | |
| lucasagomes | jroll, no problem either, it will cause some other work to need a rebase | 12:28 |
| lucasagomes | but I'm much rather merge it all now at the begining of the cycle | 12:28 |
| lucasagomes | than later | 12:28 |
| jroll | lucasagomes: yeah, that's a good point but shouldn't be a big deal | 12:29 |
| jroll | thanks! | 12:29 |
| lucasagomes | jroll, you going to rebase those patches? | 12:30 |
| openstackgerrit | Gabriel Bezerra proposed openstack/python-oneviewclient: Fixing __getattribute__ def and refactor from_json https://review.openstack.org/321262 | 12:30 |
| jroll | lucasagomes: no, but I was just going to ping the author | 12:30 |
| jroll | ... who is not onlien | 12:30 |
| lucasagomes | ack | 12:30 |
| jroll | anyway, hope they see the bug update | 12:30 |
| jroll | I'd also like to release a new version of ironic soon, maybe once the patches to make grenade work land | 12:33 |
| jroll | hoping for next week (which is newton-1 everywhere else) | 12:33 |
| *** elfosardo_ has joined #openstack-ironic | 12:34 | |
| openstackgerrit | Merged openstack/python-oneviewclient: Fixing __getattribute__ def and refactor from_json https://review.openstack.org/321262 | 12:38 |
| *** elfosardo has quit IRC | 12:38 | |
| *** piet has joined #openstack-ironic | 12:39 | |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Consolidate IPA configuration into a config module https://review.openstack.org/320867 | 12:40 |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Support LLDP data as part of interfaces in inventory https://review.openstack.org/320584 | 12:40 |
| *** rloo has joined #openstack-ironic | 12:40 | |
| *** rbudden has joined #openstack-ironic | 12:42 | |
| *** mjturek1 has joined #openstack-ironic | 12:43 | |
| *** baoli has joined #openstack-ironic | 12:46 | |
| *** parlos has joined #openstack-ironic | 12:47 | |
| *** baoli_ has joined #openstack-ironic | 12:49 | |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Support LLDP data as part of interfaces in inventory https://review.openstack.org/320584 | 12:49 |
| jroll | deva<tab> wha | 12:49 |
| jroll | :/ | 12:49 |
| sambetts | :O | 12:49 |
| jroll | wanted to point out http://specs.openstack.org/openstack/oslo-specs/specs/newton/policy-in-code.html | 12:50 |
| jroll | which I like very much | 12:50 |
| sambetts | ooooo, thats pretty neat looking! defining policy in the same way we do config options | 12:50 |
| jroll | yeah, it also allows ops to run without dropping a policy file in, if they want defaults | 12:52 |
| *** baoli has quit IRC | 12:52 | |
| *** piet has quit IRC | 12:53 | |
| openstackgerrit | Anton Arefiev proposed openstack/ironic-inspector: Fix py3 issue in functional tests https://review.openstack.org/321521 | 12:53 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Fix Bug #1583539 - rpm part https://review.openstack.org/321311 | 12:55 |
| openstack | bug 1583539 in bifrost "Missing dependencies in documentation and scripts for a CentOS 7 install of bifrost" [Undecided,New] https://launchpad.net/bugs/1583539 | 12:55 |
| *** trown is now known as trown|mtg | 12:59 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Gracefully degrade start_iscsi_target for Mitaka ramdisk https://review.openstack.org/319183 | 13:00 |
| sambetts | jroll: I wonder if we should split this down? https://review.openstack.org/#/c/206244 because the the commit message says that its adding portgroups API, but its also exposing pxe_enabled and local_link_connection on normal ports which is not dependant on port groups | 13:01 |
| *** lindycoder has joined #openstack-ironic | 13:01 | |
| jroll | sambetts: maybe, or fix the commit message | 13:02 |
| jroll | either way is fine with me, I lean toward split but I know that's difficult | 13:02 |
| *** lucasagomes is now known as lucas-hungry | 13:03 | |
| * lucas-hungry brb late lunch | 13:03 | |
| sambetts | I'm leaning the same way, I think it'll make for a nicer history, but ;/ | 13:03 |
| sambetts | :/ | 13:04 |
| NobodyCam | good morning Ironicers :) | 13:07 |
| *** Goneri has joined #openstack-ironic | 13:09 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Use correct iscsi portal port in continue_deploy https://review.openstack.org/321145 | 13:11 |
| *** lindycoder has quit IRC | 13:11 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Document that ssh_public_key_path must be set https://review.openstack.org/312782 | 13:11 |
| vdrok | morning NobodyCam | 13:11 |
| sambetts | o/ NobodyCam | 13:11 |
| NobodyCam | morning vdrok sambetts :) | 13:12 |
| vdrok | jroll: so after the patches for grenade are merged, we'll do multitenancy? or partial grenade is required too? | 13:13 |
| NobodyCam | have to reboot... brb | 13:13 |
| jroll | vdrok: just grenade | 13:13 |
| vdrok | ok, good :) | 13:16 |
| *** joprovost has joined #openstack-ironic | 13:16 | |
| *** [1]cdearborn has joined #openstack-ironic | 13:19 | |
| dtantsur | morning NobodyCam | 13:20 |
| *** piet has joined #openstack-ironic | 13:20 | |
| NobodyCam | Good morning dtantsur :) | 13:23 |
| *** ElCoyote_ has quit IRC | 13:23 | |
| NobodyCam | morning jroll :) | 13:25 |
| jroll | hey hey NobodyCam :) | 13:26 |
| *** wajdi_ has quit IRC | 13:26 | |
| NobodyCam | :) | 13:26 |
| *** absubram has quit IRC | 13:27 | |
| *** ElCoyote_ has joined #openstack-ironic | 13:27 | |
| *** joprovost1 has joined #openstack-ironic | 13:27 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add SIGHUP support for ironic-conductor service https://review.openstack.org/321111 | 13:29 |
| *** joprovost has quit IRC | 13:29 | |
| *** joprovost1 is now known as joprovost | 13:29 | |
| *** lucas-hungry is now known as lucasagomes | 13:36 | |
| *** vmud213 has quit IRC | 13:36 | |
| lucasagomes | folks, if you have some time: https://review.openstack.org/#/c/318538/ | 13:36 |
| *** rloo has quit IRC | 13:37 | |
| *** rloo has joined #openstack-ironic | 13:38 | |
| jroll | lucasagomes: no release note? | 13:41 |
| jroll | feels like we should have one, we're dropping the kernel command line interface into that | 13:41 |
| dtantsur | oh, do we? I didn't notice that, then we definitely should | 13:42 |
| dtantsur | also folks, did we already land our notifications support? I see folks discussing formats in a thread called "[tripleo] Zaqar messages standardization" | 13:43 |
| lucasagomes | jroll, oh true | 13:44 |
| lucasagomes | dtantsur, jroll will add a release note, 1 sec | 13:44 |
| jroll | dtantsur: we did not, that begins here https://review.openstack.org/#/c/298461/ | 13:45 |
| dtantsur | aha.. I wonder if we want consistency with other projects on that | 13:46 |
| dtantsur | did we borrow our format from Nova or just invented? | 13:46 |
| jroll | dtantsur: we did borrow from nova | 13:46 |
| dtantsur | ah cool | 13:46 |
| jroll | rather, borrowed from nova's latest version of it :) | 13:46 |
| jroll | it has versions and such | 13:46 |
| jroll | dtantsur: if they're talking about zaqar though I'm not sure it's the same as oslo.notification or whatever | 13:47 |
| dtantsur | jroll, I think the idea was to push it to oslo, then zaqar can pick it up and push further to e.g. web sockets | 13:48 |
| dtantsur | but I'm not sure I recall it correctly | 13:48 |
| jroll | dtantsur: yeah, I've no clue | 13:48 |
| rloo | hi and morning jroll, dtantsur, lucasagomes, NobodyCam, sambetts, vdrok, TheJulia and everyone else | 13:48 |
| sambetts | Hi rloo | 13:48 |
| lucasagomes | rloo, morning | 13:48 |
| vdrok | morning rloo ! | 13:48 |
| dtantsur | morning rloo | 13:48 |
| jroll | morning rloo :) | 13:48 |
| rloo | does anyone know if Chris Dearborn is here? I don't know his nick. | 13:49 |
| jroll | [1]cdearborn: ^ | 13:49 |
| rloo | thx jroll | 13:49 |
| jroll | np | 13:49 |
| rloo | [1]cdearborn: if you have a few minutes, am trying to understand your concerns with https://review.openstack.org/321246. maybe i didn't word it properly. | 13:50 |
| jroll | hahaha this is awesome, from irmc CI | 13:51 |
| jroll | 22:18:59 | Let inspection magic show begin. | | 13:51 |
| *** PollyZ has joined #openstack-ironic | 13:51 | |
| rloo | jroll: but does the magic work? :) | 13:51 |
| jroll | it does! | 13:52 |
| rloo | jroll: cool. hopefully there are fireworks at the end too! | 13:52 |
| jroll | hehe | 13:52 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-python-agent: Get root device hints from the node object https://review.openstack.org/318538 | 13:52 |
| *** lindycoder has joined #openstack-ironic | 13:52 | |
| *** ametts has joined #openstack-ironic | 13:52 | |
| lucasagomes | jroll, so, re deprecation. We are the ones generting the node specific pxe configuration files right? So they hints in the kernel cmdline are the same as in the nodes | 13:53 |
| *** parlos has quit IRC | 13:53 | |
| jroll | lucasagomes: in theory, yeah | 13:53 |
| lucasagomes | just stringfyed in the case of cmdline | 13:53 |
| jroll | I don't think we need to deprecate it | 13:53 |
| jroll | but, static dhcp is a thing | 13:53 |
| jroll | so maybe? | 13:53 |
| lucasagomes | jroll, good point, yeah maybe | 13:53 |
| dtantsur | jroll, even with statis DHCP IPA can access the node, right? | 13:53 |
| * lucasagomes wonders | 13:53 | |
| dtantsur | so I don't see point in any deprecations here.. it's been supported since inception of IPA | 13:54 |
| jroll | dtantsur: well, someone could configure hints in the cmd line but not node.properties | 13:54 |
| lucasagomes | dtantsur, yup, but I think jroll means adding the hitns in the PXE config for static DHCP directly and not in the node | 13:54 |
| jroll | yeah | 13:54 |
| lucasagomes | tho that's a bit dogde, since, ironic won't even be able to validate such hints | 13:54 |
| dtantsur | mmmmm, doesn't sound like a sane solution | 13:54 |
| dtantsur | like "always sda" or what? | 13:54 |
| jroll | I'm not saying it's sane, I'm saying it's possible | 13:54 |
| jroll | idk, sda, model number, etc | 13:55 |
| jroll | like I said I think it's fine to just drop it | 13:55 |
| jroll | we don't really advertise/document that, nor encourage it | 13:55 |
| jroll | just something to think about | 13:55 |
| lucasagomes | jroll, ++ yeah | 13:55 |
| dtantsur | yeah, so it justifies a release note, but probably nothing else | 13:55 |
| milan | jroll, rloo morning guys! :) | 13:56 |
| jroll | hi milan | 13:56 |
| milan | jroll, thanks for the ultra-fast review yesterday on the tempest thing ;) | 13:56 |
| milan | NobodyCam, o/ :) | 13:56 |
| jroll | milan: not a problem, I was hoping it was actually testing microversions though :) | 13:56 |
| rloo | hi milan. When are you going to work on ironic proper? :) | 13:56 |
| dtantsur | lol | 13:57 |
| milan | rloo :D | 13:57 |
| jroll | heh | 13:57 |
| dtantsur | we can start with asking when milan is going to work :D | 13:57 |
| * dtantsur is kidding ofc | 13:57 | |
| milan | dtantsur, thx :P | 13:57 |
| lucasagomes | hah | 13:57 |
| dtantsur | I'm evil cause I'm in the office without tea ;) | 13:57 |
| milan | yeah, reminds me tomorrow I'm dodging the work ;) | 13:58 |
| dtantsur | yeah-yeah | 13:58 |
| * rloo gives dtantsur a big pot of green tea | 13:58 | |
| milan | aaaah, right :) | 13:58 |
| milan | tea | 13:58 |
| dtantsur | thanks rloo :) | 13:58 |
| lucasagomes | dtantsur, you should pay kofi-kofi a visit then (tho idk if they do have tea) | 13:58 |
| NobodyCam | morning rloo milan and lucasagomes :) | 13:58 |
| dtantsur | lucasagomes, it's hard to find any sane tea in Czech outside of tea-houses | 13:58 |
| lucasagomes | NobodyCam, hi there | 13:58 |
| dtantsur | which still makes me sad | 13:59 |
| dtantsur | well, not as hard as to find good sushi :) | 13:59 |
| rloo | dtantsur: really? just beer is easy to get I guess | 13:59 |
| dtantsur | rloo, beer, pork, surely | 13:59 |
| dtantsur | italian food is not rare as well | 13:59 |
| openstackgerrit | Sam Betts proposed openstack/ironic-inspector: Add generic LLDP processing hook https://review.openstack.org/321082 | 14:00 |
| milan | sambetts, hey, we've got the tempest test case ready: could use a +2 ;) https://review.openstack.org/#/c/293926/ | 14:00 |
| lucasagomes | dtantsur, same here on the sushi | 14:00 |
| openstackgerrit | Merged openstack/python-ironicclient: Use name randomizer from tempest https://review.openstack.org/307904 | 14:00 |
| dtantsur | lucasagomes, wow, that's weird. you must have some fish, you're not that far from the sea | 14:01 |
| sambetts | dtantsur: If you want I can always post you a big box of tea ;) | 14:01 |
| lucasagomes | dtantsur, yeah, we do have all the raw material. Problem is to assembly it into a nice sushi | 14:01 |
| dtantsur | thanks, it's possible to buy some here :) | 14:01 |
| *** joprovost has quit IRC | 14:02 | |
| dtantsur | lucasagomes, wow, what slackers :D | 14:02 |
| lucasagomes | lol | 14:02 |
| *** mgoddard has quit IRC | 14:03 | |
| sambetts | :-P | 14:03 |
| *** sacharya has joined #openstack-ironic | 14:05 | |
| *** rbudden_ has joined #openstack-ironic | 14:06 | |
| *** trown|mtg is now known as trown | 14:08 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Remove links to github for OpenStack things https://review.openstack.org/321506 | 14:08 |
| *** rbudden has quit IRC | 14:09 | |
| *** rbudden_ is now known as rbudden | 14:09 | |
| *** sacharya has quit IRC | 14:10 | |
| *** wuhg has quit IRC | 14:10 | |
| dtantsur | jroll, could you please take a look at stable patch https://review.openstack.org/320862 ? | 14:14 |
| mgould | dtantsur: it's been a long time since I was in a Czech tea-house, but I remember them having hundreds of hot drinks, but almost no actual tea | 14:17 |
| mgould | Camellia sinensis or GTFO | 14:18 |
| vsaienko | jroll, I heard that partial grenade job doesn't work like expected | 14:19 |
| *** sabeen1 has quit IRC | 14:20 | |
| *** [1]cdearborn has quit IRC | 14:21 | |
| *** cdearborn has joined #openstack-ironic | 14:21 | |
| jroll | vsaienko: elaborate? | 14:21 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Tests for testing port-create command https://review.openstack.org/291802 | 14:21 |
| *** sabeen has joined #openstack-ironic | 14:22 | |
| vsaienko | yeah, I'm trying to figure out what partial upgrade should do | 14:22 |
| jroll | dtantsur: approved | 14:23 |
| jroll | vsaienko: well, it should upgrade only ironic-conductor and not ironic-api, I think | 14:23 |
| jroll | vsaienko: partial is meant to test rolling upgrades (or rather the middle state where things are different version) | 14:24 |
| jroll | vsaienko: see the rolling upgrade process here: https://review.openstack.org/#/c/299245/ | 14:24 |
| sambetts | jroll, vsaienko: things like only all of ironic and not nova too right? aren't there a lot of permutations of partial upgrade? | 14:25 |
| jroll | vsaienko: bonus points (eventually) would be to have two conductors running and only upgrade one, but that's going to be extremely difficult because networking, I think | 14:25 |
| jroll | sambetts: people keep trying to inject that, we want to start with "partial upgrade of ironic" | 14:25 |
| sambetts | isn't our ironic virt driver part of ironic? | 14:25 |
| jroll | sambetts: "partial upgrade of openstack" is a completely different thing that nobody tests today afaik | 14:25 |
| jroll | no, that's part of nova :) | 14:25 |
| jroll | I see what you mean, we should get there, one thing at a time | 14:26 |
| sambetts | ++ | 14:26 |
| vsaienko | jroll, this documentation I was looking for, need to read it carefully, thanks! | 14:26 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Tests for testing chassis-create command https://review.openstack.org/293634 | 14:26 |
| jroll | vsaienko: no problem :) | 14:26 |
| sambetts | jroll, vsaienko: we could maybe 2 two conductors in a dual slave CI job, one conductor on each slave | 14:27 |
| *** yuikotakada has joined #openstack-ironic | 14:27 | |
| vsaienko | jroll, two conductors means devstack multinode | 14:27 |
| sambetts | maybe do* | 14:27 |
| *** KennethWilke has joined #openstack-ironic | 14:28 | |
| *** yuikotakada has quit IRC | 14:28 | |
| dtantsur | mgould, heh, you can get a lot of things there, but tea is among them | 14:28 |
| jroll | sambetts: vsaienko: right, the networking would be crazy though :) | 14:28 |
| vsaienko | oh, it is a huge scope of work... Need some time to check what can we do for now | 14:28 |
| *** yuikotak_ has joined #openstack-ironic | 14:29 | |
| sambetts | jroll, vsaienko: yeah :/ we really need to rework our entire devstack plugin to make things like that possible with the VMs | 14:29 |
| vsaienko | sambetts: we definitely need to switch to Flat network... | 14:31 |
| sambetts | vsaienko: the work to do that would get us half way there for multinode I think | 14:32 |
| sambetts | I'm not sure how the networking works between the slaves though :/ | 14:32 |
| sambetts | might have to fake a real flat network using tunnel or something which would suck | 14:33 |
| *** vishwanathj has joined #openstack-ironic | 14:33 | |
| mat128 | sambetts: why would the multiple conductors (one running v1, one running v2) have to run on different slaves? | 14:36 |
| mat128 | never tried it, but I can't see why it would not be possible to run multiple processes with different paths for logs | 14:36 |
| sambetts | mat128: only because of how we fudge the networking for faking BM in devstack/testing | 14:37 |
| jroll | well, we probably could run both on the same node, actually | 14:38 |
| sambetts | we could probably fudge it even harder and make it work on one node but *shudders* | 14:38 |
| jroll | we don't do anything to the conductor to make networking work, we just muck with the interfaces themselves I believe | 14:38 |
| jroll | the only thing is the conductors would need different hostnames for registering with the db | 14:39 |
| jroll | which is nbd | 14:39 |
| * jroll tries it in devstack quickly | 14:39 | |
| mat128 | :) | 14:39 |
| vsaienko | jroll, what is the criteria to land multitenancy patches. Is full upgrade grenade ok? Or we still blocked by partial upgrades? | 14:40 |
| jroll | vsaienko: full is all we need, we still have work to do for rolling upgrades | 14:40 |
| vsaienko | jroll, understood | 14:41 |
| jroll | | 2016-05-26 14:41:21 | 2016-05-26 14:42:51 | 2 | conductor1 | ["fake", "agent_ipmitool", "pxe_ssh", "agent_ssh", "pxe_ipmitool"] | 1 | | 14:42 |
| jroll | | 2016-05-26 14:41:46 | 2016-05-26 14:42:46 | 3 | conductor02 | ["fake", "agent_ipmitool", "pxe_ssh", "agent_ssh", "pxe_ipmitool"] | 1 | | 14:42 |
| jroll | this totes works | 14:42 |
| jroll | looks like they split up the hash ring fine too | 14:43 |
| sambetts | can it actually talk to the VMs though? They need to run on different ports right? | 14:43 |
| jroll | hm, maybe not | 14:43 |
| jroll | sambetts: eh? conductor doesn't listen at all | 14:43 |
| sambetts | Oh yeah :'( | 14:44 |
| sambetts | so much layer break | 14:44 |
| jroll | ? | 14:44 |
| sambetts | IMO ironic nodes shouldn't talk to the Ironic API | 14:45 |
| jroll | hm, is there any way to look at the hash ring? | 14:45 |
| jroll | sambetts: well... that's life until someone has a solid proposal otherwise | 14:46 |
| sambetts | jroll: yeah :) I've left my idea all over this -> https://review.openstack.org/#/c/306418/ | 14:47 |
| jroll | I just realized how terrible the ssh driver is with many nodes heh | 14:48 |
| openstackgerrit | Merged openstack/virtualbmc: Use constraints for all the things https://review.openstack.org/318682 | 14:48 |
| vsaienko | jroll use libvirt from ironic-staging-drivers ;-) | 14:49 |
| mat128 | jroll, sambetts: https://review.openstack.org/#/c/212206/ | 14:49 |
| mat128 | a lot of discussions and "why do we need this" but the main idea is right | 14:49 |
| jroll | to check power state for a single node, it gets a list of nodes, and loops over them running dumpxml (over ssh!) until it finds the one it's looking for, grabs the mac address, and then gets power status | 14:49 |
| jroll | crazy with 64 nodes | 14:49 |
| vsaienko | jroll we have tested libvirt on scale with 4k nodes, it works well | 14:49 |
| jroll | vsaienko: I don't actually need performance, it was just interesting | 14:50 |
| jroll | mat128: yeah, I'm familiar with that, but I don't know who's going to step up and actually do that work | 14:50 |
| jroll | dtantsur: sambetts: I totally forgot about promoting that endpoint, sigh | 14:51 |
| * jroll needs to get organized better | 14:51 | |
| lucasagomes | sambetts, I wonder if the trusts mechanism in ks v3 could be used for IPA talking to the ironic api | 14:52 |
| vsaienko | going to upload new patchset for multitenancy patches, is there anyone who forgot to left comments? | 14:53 |
| dtantsur | lucasagomes, well, as usual: if we don't use PXE ;) | 14:53 |
| *** joprovost has joined #openstack-ironic | 14:53 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Remove support for the old bash ramdisk https://review.openstack.org/319227 | 14:53 |
| dtantsur | jroll, see what I got ^^^ :) | 14:53 |
| dtantsur | (that's not a review request, just boasting) | 14:53 |
| lucasagomes | dtantsur, heh why is that? | 14:53 |
| lucasagomes | it will talk to the api independent of the boot mode right? | 14:54 |
| dtantsur | lucasagomes, any secret you put on tftp can be used by anyone | 14:54 |
| jroll | dtantsur: \o/ | 14:54 |
| dtantsur | so unless we have a secure method to put whatever trust we have on a ramdisk, there's little we can do | 14:54 |
| lucasagomes | dtantsur, yeah :-/ | 14:55 |
| *** rloo has quit IRC | 14:55 | |
| *** rloo has joined #openstack-ironic | 14:56 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add test to verify ironic multitenancy https://review.openstack.org/269157 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add Link-Local-Connection info to ironic port https://review.openstack.org/256365 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Added operator documentation for ironic portgroups https://review.openstack.org/228496 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add portgroups to support LAG interfaces - API https://review.openstack.org/206244 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Create common neutron module https://review.openstack.org/317390 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add network_interface node field DB, OBJ https://review.openstack.org/317391 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Refactor ironic enroll-node code https://review.openstack.org/256364 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add configure_provision_network function https://review.openstack.org/256367 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Update Ironic VM network connection https://review.openstack.org/256366 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add 'neutron' network driver https://review.openstack.org/317393 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Update the deploy drivers with network flipping logic https://review.openstack.org/213262 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add network_interface API https://review.openstack.org/317392 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add network_drivers https://review.openstack.org/285852 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add Ironic/Neutron integration documentation https://review.openstack.org/258596 | 14:56 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Allow to use NETWORK_DRIVERS in devstack https://review.openstack.org/293520 | 14:56 |
| * dtantsur is impressed every time it happens | 14:56 | |
| sambetts | lucasagomes: tbh even if we had authenticated access from IPA to Ironic API I wouldn't want it to talk to it | 14:57 |
| dtantsur | right. this spec has some "but" though | 14:57 |
| dtantsur | for example it does not support ironic standalone. it does not support inspection. it still relies on a secure channel between ironic and IPA | 14:58 |
| sambetts | lucasagomes: Ironic API IMO should be public API, and the conductor should be plumbed into the provisioning network, the provisioning network should be able to be completely isolated from the outside world and the control plane | 14:58 |
| dtantsur | it's possible now, if you're using pxe_* drivers, not agent_* | 14:59 |
| sambetts | dtantsur: we still need to lookup/heartbeat so we need control plane access right? | 15:00 |
| sambetts | because its through the API | 15:00 |
| dtantsur | sambetts, well, access to ironic API, not the whole control plane | 15:00 |
| lucasagomes | sambetts, yeah maybe, I think the future versions of the API will accomodate a lot more of the use cases that we have now that wasn't the case in the past | 15:00 |
| dtantsur | or even: access to two endpoints of ironic API | 15:00 |
| lucasagomes | ironic was designed so the nova API would serve as an public api for it | 15:01 |
| lucasagomes | the current one I mean* | 15:01 |
| *** mgoddard has joined #openstack-ironic | 15:02 | |
| *** dmk0202 has quit IRC | 15:04 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: Add boot information into the inventory https://review.openstack.org/320448 | 15:05 |
| *** joprovost1 has joined #openstack-ironic | 15:05 | |
| *** joprovost has quit IRC | 15:06 | |
| *** joprovost1 is now known as joprovost | 15:06 | |
| sambetts | right, but each conductor is in charge of a particular set of nodes right, we do that to scale up ironic, right now if IPA wants to heartbeat it has to hit the Ironic API which then has to forward the request to the right conductor, instead of the communication being directly from conductor in charge of that node to the node and back. So we basically have to scale up the API at the same time as | 15:08 |
| sambetts | the conductors because even though we can scale them separately the amount of traffic hitting the Ironic API scales linearly with the number of nodes registered into Ironic | 15:08 |
| sambetts | also I might put my server running my conductor service nice and close the the set of nodes it is managing, and the API might be somewhere a long way away from that machine, so talking to API to talk to my conductor seems like the long way around | 15:11 |
| *** mgoddard has quit IRC | 15:13 | |
| *** jaybeale has joined #openstack-ironic | 15:13 | |
| *** mgoddard has joined #openstack-ironic | 15:13 | |
| *** piet has quit IRC | 15:13 | |
| *** mgoddard has quit IRC | 15:14 | |
| *** mgoddard has joined #openstack-ironic | 15:14 | |
| *** itamarl has quit IRC | 15:15 | |
| *** mtanino has joined #openstack-ironic | 15:16 | |
| *** lennyb has quit IRC | 15:17 | |
| vdrok | jroll: I think if run two conductors on the same node, api won't know to which to pass the messages, get_topic_for will be switching between them randomly | 15:18 |
| jroll | vdrok: not if you set [DEFAULT]/host differently | 15:19 |
| *** rcernin has quit IRC | 15:19 | |
| vdrok | heh, didn't know about that :) | 15:19 |
| *** penick has joined #openstack-ironic | 15:22 | |
| lucasagomes | jroll, ++ | 15:23 |
| *** sabeen1 has joined #openstack-ironic | 15:26 | |
| lucasagomes | sambetts, right, and proposed solution is to have a separated service that handles the communication between the IPA and the conductor? Or having IPA to talk to the conductor directly ? | 15:27 |
| *** sabeen has quit IRC | 15:28 | |
| sambetts | lucasagomes: I've thought about both and in the end I think I settled on, I would suggest separate processes, one or more per conductor, with one foot inside a neutron network and one foot outside | 15:31 |
| *** tesseract has quit IRC | 15:34 | |
| *** sacharya has joined #openstack-ironic | 15:34 | |
| lucasagomes | sambetts, right, doesn't it make it harder to scale? Now you have to scale up this internal API as the same time as the conductor + manage the external API as well | 15:34 |
| lucasagomes | at the same* | 15:35 |
| *** piet has joined #openstack-ironic | 15:36 | |
| *** hamzy has quit IRC | 15:37 | |
| sambetts | lucasagomes: the external API can scale based on the number of users/nova's you have, and everything else scales with the conductors, if I double my nodes I add a second conductor and that starts the internal API processes, and there is no need to touch the external API | 15:37 |
| sambetts | neutron starts the internal API processes, like it does is the metadata proxy service | 15:38 |
| sambetts | or ^ | 15:38 |
| * JayF == lucasagomes | 15:38 | |
| JayF | we have long running agents so we already see "the flood" of heartbeats on our api cluster | 15:39 |
| JayF | and it'd be 1000x more complex if we had to scale out 2 clusters instead of one. | 15:39 |
| *** rama_y has joined #openstack-ironic | 15:39 | |
| JayF | However, I could see that being a valuable *choice* for a deployer to make if they wanted, limiting the "agent api cluster" to only being able to do the two api calls the agent needs by policy, and nothing else whatseover | 15:39 |
| JayF | and with it being limited to agents, you could lock it down to the needed paths even moreso | 15:39 |
| *** yuikotak_ has quit IRC | 15:40 | |
| * JayF hopes he digested enough of that context before jumping in | 15:40 | |
| jroll | couldn't you just do that by standing up two sets of api clusters, one for users with one policy, one for agents with different policy? | 15:40 |
| sambetts | JayF: what I'm saying is that doing it my way reduces the need to scale the API cluster, because you get your scaling by adding more conductors | 15:40 |
| *** joprovost1 has joined #openstack-ironic | 15:40 | |
| JayF | sambetts: then I'm missing something; you want to expose an api on conductors? | 15:41 |
| sambetts | JayF: implementation details aside, basically, yes | 15:41 |
| JayF | Yikes. | 15:41 |
| JayF | That would change the failure model around conductors considerably, wouldn't it? | 15:42 |
| sambetts | the conductor already talks directly to the ramdisk, why shouldn't the ramdisk talk to the conductor | 15:42 |
| JayF | Like honestly, I don't understand if you go that route, why not eliminate a dedicated api layer altogether/ | 15:42 |
| sambetts | ? | 15:43 |
| *** joprovost has quit IRC | 15:43 | |
| JayF | I'm not sure how the agent is any different than any other api client | 15:43 |
| JayF | like I'm extending your idea to it's logical conclusion: why have a dedicated api service separate from conductors at all? | 15:43 |
| sambetts | because you need to have a way to talk to the whole cluster, e.g. node list etc | 15:44 |
| *** joprovost1 has quit IRC | 15:45 | |
| JayF | oh, so you'd actually link the agent and the conductor | 15:45 |
| JayF | you wouldn't like, load balance on top of the conductor | 15:45 |
| JayF | that is pretty much an impossible configuration with static dhcp :( | 15:45 |
| *** joprovost has joined #openstack-ironic | 15:45 | |
| JayF | like ipa-api-url= on the agent would be the direct ip of the conductor doing teh deploy | 15:45 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Use correct iscsi portal port in continue_deploy https://review.openstack.org/321145 | 15:45 |
| sambetts | JayF: yes | 15:45 |
| JayF | Doesn't that mean the failure of a single conductor is more likely to fail a deployment/ | 15:46 |
| JayF | Right now, if a deploy is in DEPLOYWAIT and the conductor managing it dies, the deploy can be picked up by another conductor | 15:46 |
| JayF | in this case, you'd have to completely restart/reconfigure the agent to recover from failure | 15:46 |
| mat128 | sambetts, JayF maybe ir-cond can do the heartbeat to agents running? | 15:47 |
| mat128 | not really thought about the idea througly | 15:47 |
| mat128 | throughly* | 15:47 |
| *** joprovost has quit IRC | 15:50 | |
| *** mtanino has quit IRC | 15:51 | |
| *** joprovost has joined #openstack-ironic | 15:51 | |
| sambetts | JayF: I also thought about putting a separate http API service between the conductor and the node, which run completely unauthenticated serving the heartbeat and lookup endpoints and that was it | 15:52 |
| *** mtanino has joined #openstack-ironic | 15:52 | |
| JayF | To me; there's no need for this to be a different service. You can choose to do that today with the existing API service. | 15:52 |
| JayF | What would this "agent api service" do differently than a normal ironic-api instance? | 15:52 |
| sambetts | completely unauthenticated, hooked directly into the neutron namespace, and all conductor to ramdisk communication would be proxied through it | 15:55 |
| JayF | I still think you could implement this today using apache configuration + existing ironic-api services | 15:56 |
| *** jaybeale has quit IRC | 15:57 | |
| mat128 | JayF, sambetts: yes you can, we've been running like since the beginning | 15:57 |
| mat128 | reverse proxy that only allows specific paths | 15:57 |
| mat128 | reduces attack surface a little bit | 15:57 |
| *** dprince has quit IRC | 15:57 | |
| JayF | we implement similar restrictions, but through our switches and networks. Using static DHCP makes that path a bit easier too, I think | 15:58 |
| *** dprince has joined #openstack-ironic | 15:58 | |
| mat128 | static DHCP as in static leases, or purely "static" DHCP/PXE | 15:59 |
| sambetts | JayF: I'm talking basically using this service to NAT all conductor communication into neutron networks so that it allows 1 conductor to control nodes in mutliple isolated neutron networks | 15:59 |
| mat128 | we do the latter, don,t see any advantage to the former | 15:59 |
| JayF | mat128: we have all our dhcp/pxe booting setup statically. Probably not a decision we would repeat if building out today. | 16:00 |
| JayF | mat128: but it reduced a lot of risk for us when Ironic was younger :) | 16:00 |
| lucasagomes | aarefiev, around? I see sylvain comment here https://review.openstack.org/#/c/316031 saying to wait for the api_model, which is now merged | 16:00 |
| aarefiev | lucasagomes: yeah, I'll ping him one more time | 16:01 |
| mat128 | JayF: the dynamic PXE stuff wasn't even in place back then, I remember talking about it with Lucas in Vancouver | 16:01 |
| mat128 | so you had no choice :) | 16:02 |
| lucasagomes | aarefiev, fair enough, so nothing to update in that patch? | 16:02 |
| JayF | mat128: actually I don't think that's true | 16:02 |
| *** mgoddard_ has joined #openstack-ironic | 16:02 | |
| *** mgoddard has quit IRC | 16:02 | |
| *** wajdi has joined #openstack-ironic | 16:02 | |
| mat128 | really? | 16:02 |
| JayF | mat128: An intern for OnMetal is the person who ripped out the dhcp code and made it a pluggable interface | 16:02 |
| JayF | mat128: so we could write a noop driver and plug it in :) | 16:02 |
| JayF | I think? | 16:02 |
| jroll | mat128: I think you mean "stateless ipxe" | 16:02 |
| mat128 | jroll is correct | 16:02 |
| jroll | where ipxe.txt is served from the api | 16:02 |
| mat128 | my bad, there was always code for dhcp | 16:02 |
| mat128 | patched out downstream here ;) | 16:03 |
| jroll | heh | 16:03 |
| openstackgerrit | Merged openstack/ironic: minor changes to security documentation https://review.openstack.org/320566 | 16:03 |
| mat128 | until the noop driver landed | 16:03 |
| aarefiev | lucasagomes: looks like no | 16:03 |
| lucasagomes | aarefiev, cool I will see if I get him to look at the patch again too | 16:03 |
| lucasagomes | aarefiev, looking fwd to get that fixed | 16:03 |
| JayF | mat128: we're getting perilously close to running master. Once metrics and the rescue stuff we have outstanding for review get in, networking is one of the last pieces. | 16:03 |
| aarefiev | me too :) | 16:03 |
| *** wajdi_ has joined #openstack-ironic | 16:04 | |
| dtantsur | heading home, see you tomorrow | 16:05 |
| sambetts | o/ dtantsur | 16:05 |
| lucasagomes | dtantsur, g'night | 16:05 |
| *** jaybeale has joined #openstack-ironic | 16:06 | |
| vdrok | night dtantsur | 16:06 |
| *** wajdi has quit IRC | 16:07 | |
| *** trown is now known as trown|lunch | 16:07 | |
| *** piet has quit IRC | 16:08 | |
| *** PollyZ_ has joined #openstack-ironic | 16:09 | |
| *** PollyZ has quit IRC | 16:10 | |
| *** PollyZ_ is now known as PollyZ | 16:10 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/virtualbmc: Set IPMI session timeout for the virtual BMCs https://review.openstack.org/316022 | 16:11 |
| *** rbudden has quit IRC | 16:11 | |
| *** joprovost has quit IRC | 16:14 | |
| *** ChrisAusten has joined #openstack-ironic | 16:15 | |
| * sambetts -> home | 16:15 | |
| sambetts | Night all! o/ Enjoy the long weekend, I'm out of office tomorrow | 16:16 |
| *** sambetts is now known as sambetts|afk | 16:16 | |
| *** ifarkas has quit IRC | 16:16 | |
| lucasagomes | sambetts|afk, g'night | 16:17 |
| lucasagomes | vdrok, interesting I didn't know we couldn't register new users in the wiki page | 16:18 |
| * lucasagomes check ML about | 16:18 | |
| vdrok | night sambetts|afk | 16:18 |
| vdrok | lucasagomes: yep, there was a thread about this recently | 16:18 |
| lucasagomes | vdrok, fair enough | 16:19 |
| lucasagomes | yeah maybe we should temporarily move the agenda to the ironic whiteboard and add a note to the wiki redirecting people to it | 16:19 |
| lucasagomes | jroll, ^ not sure if you are aware | 16:19 |
| *** _vsaienko has joined #openstack-ironic | 16:20 | |
| *** _vs has joined #openstack-ironic | 16:20 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add tests for testing node-delete command https://review.openstack.org/251458 | 16:21 |
| vdrok | lucasagomes: jroll here is the thread if you haven't found yet - http://lists.openstack.org/pipermail/openstack-dev/2016-May/094408.html | 16:21 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add tests for testing node-delete command https://review.openstack.org/251458 | 16:24 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add tests for testing node-delete command https://review.openstack.org/251458 | 16:25 |
| lucasagomes | vdrok, ty :-) damn spammers | 16:31 |
| *** _vsaienko has quit IRC | 16:32 | |
| *** _vs has quit IRC | 16:32 | |
| *** kromanenko_ has quit IRC | 16:34 | |
| *** kromanenko__ has quit IRC | 16:34 | |
| *** Romanenko_K has quit IRC | 16:34 | |
| *** wajdi_ has quit IRC | 16:35 | |
| *** dprince has quit IRC | 16:35 | |
| *** dprince has joined #openstack-ironic | 16:35 | |
| lucasagomes | calling it a day too | 16:41 |
| lucasagomes | have a great evening all! See you all tomorrow :-) | 16:41 |
| *** lucasagomes is now known as lucas-afk | 16:41 | |
| jroll | lucas-afk: vdrok: I'd prefer keep the agenda on the wiki, but have a note to drop a note in email/irc if people can't update it | 16:43 |
| jroll | or something | 16:43 |
| jroll | idk | 16:43 |
| JayF | If you ask in -infra, they'll enable new signups long enuogh for you to signup | 16:43 |
| *** PollyZ has quit IRC | 16:49 | |
| *** PollyZ has joined #openstack-ironic | 16:51 | |
| *** mgoddard_ has quit IRC | 16:53 | |
| *** mgoddard_ has joined #openstack-ironic | 16:53 | |
| *** ChubYann has joined #openstack-ironic | 16:53 | |
| *** absubram has joined #openstack-ironic | 16:55 | |
| *** derekh has quit IRC | 16:56 | |
| *** absubram has quit IRC | 16:56 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-inspector: Updated from global requirements https://review.openstack.org/321679 | 16:58 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-lib: Updated from global requirements https://review.openstack.org/321680 | 16:58 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/321681 | 16:58 |
| *** absubram has joined #openstack-ironic | 16:59 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-ironic-inspector-client: Updated from global requirements https://review.openstack.org/321720 | 17:04 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-ironicclient: Updated from global requirements https://review.openstack.org/321721 | 17:04 |
| *** baoli_ has quit IRC | 17:06 | |
| *** baoli has joined #openstack-ironic | 17:06 | |
| *** ijw has joined #openstack-ironic | 17:11 | |
| *** ijw has quit IRC | 17:11 | |
| *** fragatina has joined #openstack-ironic | 17:15 | |
| *** moshele has quit IRC | 17:16 | |
| *** ijw has joined #openstack-ironic | 17:16 | |
| *** mtanino has quit IRC | 17:17 | |
| *** fragatina has quit IRC | 17:20 | |
| *** trown|lunch is now known as trown | 17:20 | |
| *** wajdi has joined #openstack-ironic | 17:27 | |
| *** dprince has quit IRC | 17:28 | |
| *** ijw has quit IRC | 17:28 | |
| *** wajdi_ has joined #openstack-ironic | 17:28 | |
| *** baoli_ has joined #openstack-ironic | 17:29 | |
| *** mgould is now known as mgould|afk | 17:29 | |
| *** joprovost has joined #openstack-ironic | 17:30 | |
| *** wajdi has quit IRC | 17:31 | |
| *** baoli has quit IRC | 17:32 | |
| rloo | jroll: wrt meeting agenda. why not put it in the etherpad, just above the subteam status reports? | 17:33 |
| rloo | jroll: would mean one less *thing* to do each week (no need to update the wiki/meetings page) | 17:34 |
| jroll | rloo: dunno, I think the whiteboard is too cluttered already, and I appreciate the emails I get when the agenda is updated | 17:34 |
| jroll | no strong reasons though | 17:34 |
| rloo | jroll: this whole wiki thing is a pain. i don't want people to feel like they are second-class citizens. | 17:35 |
| jroll | rloo: also direct links to the wiki is important | 17:36 |
| jroll | rloo: sure, I agree, just don't see the need to change personally | 17:36 |
| jroll | rloo: this wiki thing has been a thing for quite some time, I haven't heard about it being a problem wrt our agenda | 17:37 |
| rloo | jroll: maybe cuz the only ones updating our agenda are the 'older' folks? or cuz the newer ones that cannot, don't say anything? | 17:37 |
| JayF | I'm honestly a bit with rloo here | 17:37 |
| JayF | unless we get infra's permissino to advertise their willingness to enable signups as needed directly on the page | 17:38 |
| *** e0ne has quit IRC | 17:39 | |
| jroll | rloo: I think those folks should speak up if they have a problem with it, personally | 17:39 |
| jroll | but I really don't care, whatever works for the team | 17:39 |
| jroll | I don't believe etherpad is the right place for it | 17:39 |
| rloo | oh interesting. so there is going to be a 'minimal' fee to go to the Project Teams gathering. | 17:39 |
| jroll | but I don't have an alternative | 17:39 |
| JayF | rloo: in exchange for it being less sponsor laden. I think that's a good tradeoff. | 17:40 |
| rloo | jroll: well, i'm fine waiting to see if anyone complains. | 17:40 |
| rloo | JayF: depends on what 'minimal fee' is. I don't know why I'm mentioning it, the fee will always be minimal compared to flight/hotel :) | 17:40 |
| *** electrofelix has quit IRC | 17:41 | |
| JayF | rloo: I lost my taste for sponsors at the summit when at the same summit we got integrated, MAAS got a keynote. | 17:44 |
| JayF | rloo: doesn't feel community to be able to buy and sell influence at community events | 17:44 |
| *** vishwanathj has quit IRC | 17:44 | |
| jroll | JayF: you should've seen austin where intel showed how they went to a corner and rewrote nova in golang including vms, containers, and bare metal | 17:45 |
| mat128 | JayF: I remember that MAAS keynote and one question from the public: "But is this using Ironic?" "Nope" | 17:45 |
| *** david-lyle has quit IRC | 17:47 | |
| *** rama_y has quit IRC | 17:47 | |
| *** david-lyle has joined #openstack-ironic | 17:47 | |
| *** devananda has joined #openstack-ironic | 17:50 | |
| *** ijw has joined #openstack-ironic | 17:51 | |
| *** athomas has quit IRC | 17:53 | |
| *** Nisha_away has joined #openstack-ironic | 18:03 | |
| *** rama_y has joined #openstack-ironic | 18:18 | |
| *** baoli_ has quit IRC | 18:21 | |
| *** baoli has joined #openstack-ironic | 18:21 | |
| *** jaybeale has quit IRC | 18:23 | |
| *** wajdi_ is now known as wajdi | 18:34 | |
| *** _vsaienko has joined #openstack-ironic | 18:40 | |
| *** _vs has joined #openstack-ironic | 18:40 | |
| *** lucas-afk has quit IRC | 18:48 | |
| *** daemontool has quit IRC | 18:51 | |
| *** lucasagomes has joined #openstack-ironic | 18:54 | |
| *** baoli has quit IRC | 19:01 | |
| *** baoli has joined #openstack-ironic | 19:01 | |
| *** PollyZ_ has joined #openstack-ironic | 19:04 | |
| *** PollyZ has quit IRC | 19:05 | |
| *** PollyZ_ is now known as PollyZ | 19:05 | |
| *** e0ne has joined #openstack-ironic | 19:13 | |
| *** fragatina has joined #openstack-ironic | 19:16 | |
| *** fragatina has quit IRC | 19:17 | |
| *** fragatina has joined #openstack-ironic | 19:17 | |
| zer0c00l | This is probably a stupid question: Are the rpcapi calls from ironic-api to ironic-condutor async | 19:21 |
| *** PollyZ has quit IRC | 19:21 | |
| jroll | zer0c00l: call is sync, cast is async | 19:21 |
| rloo | zer0c00l: look in conductor/rpcapi.py to see if a call or cast is being done. depends on the actual call | 19:22 |
| *** baoli has quit IRC | 19:23 | |
| zer0c00l | pecan.request.rpcapi.do_node_deploy( | 19:23 |
| zer0c00l | rloo: okay | 19:23 |
| zer0c00l | jroll: thanks | 19:23 |
| jroll | right, look at that method in rpcapi | 19:23 |
| jroll | often it will be a call, which means sync | 19:24 |
| jroll | but then if you look at the server side (conductor.manager)... | 19:24 |
| jroll | it will spawn a thread and return immediately | 19:24 |
| *** Waltz has joined #openstack-ironic | 19:24 | |
| rloo | so it kind of depends on what you meant by sync or async :) | 19:25 |
| zer0c00l | i see | 19:25 |
| jroll | heh | 19:25 |
| jroll | the rpc call is usually sync but the actual action is usually async | 19:25 |
| zer0c00l | okay :) | 19:25 |
| *** zzzeek has quit IRC | 19:31 | |
| openstackgerrit | Xavier proposed openstack/python-oneviewclient: Adding dynamic allocation operations https://review.openstack.org/258699 | 19:32 |
| *** zzzeek has joined #openstack-ironic | 19:39 | |
| *** baoli has joined #openstack-ironic | 19:41 | |
| *** Nisha_away has quit IRC | 19:44 | |
| *** [1]cdearborn has joined #openstack-ironic | 19:47 | |
| *** e0ne has quit IRC | 19:49 | |
| *** e0ne has joined #openstack-ironic | 19:55 | |
| *** absubram has quit IRC | 19:55 | |
| TheJulia | so.... this is going to feel like egg on the face, but has anyone validated the json in the raid docs? | 19:55 |
| *** sdake_ has joined #openstack-ironic | 19:58 | |
| *** PollyZ has joined #openstack-ironic | 20:00 | |
| *** sdake has quit IRC | 20:00 | |
| *** cdearborn has quit IRC | 20:02 | |
| rloo | TheJulia: I thought the HPE folks did. Ramesh et al. | 20:03 |
| rloo | TheJulia: OH. the documentation. dunno. | 20:04 |
| TheJulia | heh, yeah, I think they were thinking dictonaries, it should be double quoted :( | 20:04 |
| TheJulia | since json | 20:04 |
| rloo | TheJulia: quite possible. | 20:04 |
| rloo | TheJulia: I mean, it is quite possible that the docs are wrong. | 20:04 |
| TheJulia | I just sifted through the history of the original review | 20:04 |
| TheJulia | it looks that way :\ | 20:04 |
| * TheJulia opens a bug | 20:05 | |
| rloo | TheJulia: as long as the code is right. Updating doc is easier and we don't have to worry about breaking people's code. | 20:05 |
| rloo | thx for noticing and doing something about it TheJulia ! | 20:05 |
| *** rbudden has joined #openstack-ironic | 20:07 | |
| *** ayoung has quit IRC | 20:07 | |
| TheJulia | rloo: oh no problem, it was an internal user that went "Hey, wait a second... and I tossed it into a linter and went... yup, something is really wrong here" :( | 20:08 |
| rloo | TheJulia: a nice low-hanging-fruit for someone to fix... ;) | 20:09 |
| TheJulia | I'll actually do it while I'm sitting here right now, I've been interrupted like 500 times today :( | 20:10 |
| *** absubram has joined #openstack-ironic | 20:11 | |
| *** jaybeale has joined #openstack-ironic | 20:12 | |
| rloo | Hey TheJulia, I want to bug you. (jk) | 20:12 |
| TheJulia | lol | 20:13 |
| *** PollyZ has quit IRC | 20:13 | |
| TheJulia | only took me like 3 days to get through just my work email... I really should have just deleted it all | 20:14 |
| *** PollyZ has joined #openstack-ironic | 20:14 | |
| rloo | TheJulia: it just means that you are important and we all missed you! | 20:16 |
| *** rbudden has quit IRC | 20:19 | |
| TheJulia | :) | 20:19 |
| TheJulia | https://github.com/openstack/ironic/blame/master/doc/source/deploy/raid.rst#L171 vs line 177 seems weird to me..... | 20:22 |
| openstackgerrit | Mario Villaplana proposed openstack/ironic: Add notification base classes and docs https://review.openstack.org/298461 | 20:23 |
| *** e0ne has quit IRC | 20:23 | |
| *** e0ne has joined #openstack-ironic | 20:27 | |
| *** e0ne has quit IRC | 20:29 | |
| *** e0ne has joined #openstack-ironic | 20:33 | |
| JayF | rloo: that project-config patch is on my list for today, about to start working on it, but a bump is always appreciated :) | 20:35 |
| rloo | JayF: ok. bump = +/- 1? | 20:35 |
| JayF | I'm using it like you would for old school forum software, where posting any comment "bumps" something to the top of the thread | 20:36 |
| JayF | so you re-1ing so I see it I consider a 'bump' | 20:36 |
| rloo | JayF: ok. I wasn't sure if you didn't notice the comments initially cuz I had 0'd it, so I -1'd it. | 20:37 |
| JayF | no I saw them, and have half-finished responding to comments, just had a few other things in my hair + missed a majority of last week sick | 20:37 |
| rloo | JayF: Ok good. Sorry you were sick last week. I am trying (sort of) to get patches turned around faster, so just wanted to make sure you knew. | 20:38 |
| JayF | I appreciate it | 20:39 |
| JayF | if you have room on your review radar, the rescue spec (which is a priority, albeit mid-to-low) is getting dusty with no reviews at all on them | 20:39 |
| JayF | s/them/it/ | 20:40 |
| rloo | JayF: that ain't going to happen soon. I've got driver composition/volume that I want to look at and I think i might start looking at multitenant patches soon (depending on grenade) | 20:40 |
| rloo | JayF: is the rescue stuff fairly straightforward? code written already? | 20:41 |
| mariojv | hey, random question, not sure if this is the best channel for it. what's a good way to manage multiple patchsets that depend on each other with gerrit? | 20:41 |
| mariojv | i have the patch for ironic notifications, and the patch for adding a notif - if i keep them on the same branch, then run git review, will it try to update both patches? | 20:41 |
| mariojv | it seems like it would be annoying to keep rebasing multiple times for each commit in a series | 20:42 |
| JayF | rloo: interface is already merged, my spec is for an agent driver implementation on it. Don't overload yourself on my behalf though, those other things I think are higher prio | 20:42 |
| JayF | mariojv: yeah, so think of it this way | 20:42 |
| JayF | mariojv: you build a set of commits on master, git review pushes *each commit* as a dependent patchset | 20:42 |
| JayF | but lets say you don't edit the first patch, you edit the second one; it will only push the second one if the first one has no changes when rebased | 20:42 |
| JayF | for notifications, you absolutely want to do them as dependant patchsets | 20:42 |
| JayF | because you'll not going to have anything merge before the foundational patch | 20:43 |
| mariojv | ok, great. will it auto-add depends-on too, or is that manual? | 20:43 |
| JayF | this is the original depends-on :) | 20:43 |
| mariojv | awesome. thanks! | 20:43 |
| JayF | you only need depends-on in the commit message for cross-project or if the changes are unrelated otherwise | 20:43 |
| *** rbudden has joined #openstack-ironic | 20:48 | |
| JayF | https://review.openstack.org/#/c/321153 is a simple spec update for metrics if someone has a moment to review+land. 1x+2 and 2x+1 already (including jroll, but he has mental-effort in this change so didn't vote as a core) | 20:48 |
| *** yuikotak_ has joined #openstack-ironic | 20:57 | |
| *** yuikotak_ has quit IRC | 20:58 | |
| *** yuikotak_ has joined #openstack-ironic | 20:58 | |
| *** yuikotak_ is now known as yuikotakadamori_ | 20:59 | |
| *** Goneri has quit IRC | 21:03 | |
| *** fragatina has quit IRC | 21:09 | |
| *** lindycoder has quit IRC | 21:11 | |
| *** trown is now known as trown|outtypewww | 21:11 | |
| *** rbudden has quit IRC | 21:16 | |
| *** _vsaienko has quit IRC | 21:17 | |
| *** _vs has quit IRC | 21:17 | |
| *** aNupoisc has joined #openstack-ironic | 21:19 | |
| *** causten_ has joined #openstack-ironic | 21:19 | |
| jroll | JayF: https://bugs.launchpad.net/ironic/+bug/1586173 | 21:21 |
| openstack | Launchpad bug 1586173 in Ironic "Tempest tests don't verify CLEANING state" [Undecided,New] | 21:21 |
| JayF | ty jroll | 21:22 |
| jroll | np | 21:22 |
| *** ChrisAusten has quit IRC | 21:23 | |
| *** causten_ has quit IRC | 21:24 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic: Correct RAID documentation JSON https://review.openstack.org/321862 | 21:24 |
| *** fragatina has joined #openstack-ironic | 21:29 | |
| openstackgerrit | Mario Villaplana proposed openstack/ironic: Add power state change notifications https://review.openstack.org/321865 | 21:30 |
| *** jrist has quit IRC | 21:32 | |
| *** Egyptian[Home] has joined #openstack-ironic | 21:34 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic-specs: Support rolling upgrades https://review.openstack.org/299245 | 21:34 |
| *** yuikotakadamori_ has quit IRC | 21:37 | |
| *** jrist has joined #openstack-ironic | 21:45 | |
| JayF | crosspost from #openstack-infra: | 21:48 |
| JayF | Is there a reason for constructs like https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/devstack-gate.yaml#L690 we export the value to a local variable instead of just referencing {ipxe-enabled} in the if directly? | 21:48 |
| JayF | jroll: ^ do you have any idea? | 21:48 |
| *** rbradfor is now known as rbradfor_not_fou | 21:51 | |
| *** rbradfor_not_fou is now known as rbradf_not_found | 21:52 | |
| *** sdake has joined #openstack-ironic | 21:53 | |
| *** e0ne has quit IRC | 21:53 | |
| JayF | I see folks not using the intermediate var in that file, so I won't | 21:54 |
| *** sdake_ has quit IRC | 21:54 | |
| *** PollyZ_ has joined #openstack-ironic | 21:55 | |
| *** jaybeale has quit IRC | 21:57 | |
| *** wajdi has quit IRC | 21:58 | |
| *** ametts has quit IRC | 21:58 | |
| *** PollyZ has quit IRC | 21:59 | |
| *** PollyZ_ has quit IRC | 22:00 | |
| *** PollyZ has joined #openstack-ironic | 22:00 | |
| *** amotoki has joined #openstack-ironic | 22:02 | |
| *** Waltz has quit IRC | 22:02 | |
| jroll | JayF: no, likely just silliness | 22:08 |
| * jroll not here right now | 22:08 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic: Add 'How to get a decision on something' to FAQ https://review.openstack.org/321246 | 22:12 |
| *** ayoung has joined #openstack-ironic | 22:12 | |
| *** rbudden has joined #openstack-ironic | 22:12 | |
| aNupoisc | cinerama: are you there? | 22:13 |
| *** Egyptian[Home] has quit IRC | 22:13 | |
| cinerama | hi aNupoisc | 22:13 |
| aNupoisc | hi. actually yesterday i tried couple of things but didn't worked | 22:13 |
| aNupoisc | cinerama: i am just getting stuck at same keystone point | 22:14 |
| aNupoisc | can you just check if any of these values needs to be changed | 22:14 |
| aNupoisc | http://pastebin.intel.com/view/7c76eaae#L14 | 22:14 |
| cinerama | aNupoisc, ok so you had a clean VM to start with, and you ran bifrost's install steps as specified in the docs? | 22:15 |
| cinerama | also that might be an internal pastebin, the link doesn't work? | 22:15 |
| *** [1]cdearborn has quit IRC | 22:15 | |
| aNupoisc | cinerama: yes i did that. Oh i will pastebin a newone | 22:16 |
| aNupoisc | cinerama: http://pastebin.com/ma2HaqNd | 22:16 |
| aNupoisc | cinerama: it is from ironic.conf | 22:17 |
| cinerama | aNupoisc can you please remind me what error you're seeing? and can you check the /etc/ironic-inspector/inspector.conf auth_strategy too? | 22:20 |
| aNupoisc | cinerama: yes so on running enroll-dynamic I am seeing No auth found from Keystone,iddleware.auth_token | 22:22 |
| aNupoisc | cinerama: i have set auth_strategy to noauth | 22:22 |
| *** Goneri has joined #openstack-ironic | 22:23 | |
| TheJulia | aNupoisc: how did you install everything? | 22:23 |
| aNupoisc | i can't see ironic-inspector | 22:23 |
| aNupoisc | I am following http://docs.openstack.org/developer/bifrost/readme.html#installation | 22:24 |
| aNupoisc | TheJulia | 22:26 |
| * TheJulia should make another post-it note... rev documentation | 22:26 | |
| aNupoisc | cinerama: on cloning bifrost.git I am applying a patch which my team had used earlier to run bifrost but that patch is not an issue | 22:26 |
| aNupoisc | cinerama: shall i pastebin the gitt diff? | 22:26 |
| *** jaybeale has joined #openstack-ironic | 22:27 | |
| TheJulia | aNupoisc: if you can get us the ansible playbook output with -vvvv where the you get the stacktrace from keystonemiddleware.auth_token, that will be most helpful | 22:28 |
| aNupoisc | TheJulia: yes sure | 22:28 |
| aNupoisc | TheJulia: give a minute | 22:28 |
| cinerama | i am kind of curious about your diff too, perhaps it works in a different environment | 22:29 |
| *** ijw has quit IRC | 22:29 | |
| cinerama | also, when i was helping you the other day you gave me the impression you were working off an unaltered clone | 22:29 |
| TheJulia | if it changes the role... that might be the cause | 22:29 |
| *** rbudden has quit IRC | 22:31 | |
| TheJulia | cinerama: was it a recent clone/install? | 22:32 |
| cinerama | aNupoisc, ^^^ | 22:32 |
| aNupoisc | cinerama: TheJulia: http://pastebin.com/zus9iZQq this is -vvvv output | 22:33 |
| aNupoisc | http://pastebin.com/F5iMJehA this is ironic-api.log putput | 22:34 |
| aNupoisc | cinerama: yes i tried with unaltered repo and altered repo after patch too | 22:34 |
| TheJulia | Ahh! context! | 22:37 |
| TheJulia | something is up with your ironic config and it is trying to force authentication | 22:38 |
| *** PollyZ has quit IRC | 22:38 | |
| cinerama | uh huh | 22:38 |
| *** PollyZ has joined #openstack-ironic | 22:39 | |
| *** ijw has joined #openstack-ironic | 22:39 | |
| TheJulia | the fact that it is pointing to a specific port/point makes me think that bifrost was installed on top of maybe something like a devstack install? | 22:39 |
| cinerama | so i had aNupoisc check out the ironic conf, was using keystone but we changed it to noauth. still having trouble after that | 22:39 |
| cinerama | yes i asked about that too | 22:39 |
| TheJulia | we, at present (unless someone merged it and I've not noticed while I was out... which is entirely possible) only do line level string replacements, so we don't squash the entire ironic.conf file | 22:40 |
| cinerama | that's right i think | 22:40 |
| cinerama | the patch to get away from using the sample conf did not merge yet | 22:40 |
| TheJulia | tl;dr to allow the user to have a little more flexibility. There is a patch to go back to being template driven, which will eventually land | 22:40 |
| TheJulia | Yeah, motivation for sample config context is strong though ;) | 22:41 |
| cinerama | did not pass CI in a couple spots, meant to poke it but #things happened | 22:41 |
| aNupoisc | TheJulia: so yes I am creating a bifrost VM and then bifrost is installed in that VM | 22:41 |
| TheJulia | c'est la vie | 22:41 |
| cinerama | there *is* a relevant patch that landed sometime recently though IIRC hang on | 22:43 |
| *** Goneri has quit IRC | 22:43 | |
| cinerama | aNupoisc, if you do a "git log" in your bifrost, what's the last revision you see (aside from your in-house patches)? | 22:43 |
| aNupoisc | cinerama: this is the git diff http://pastebin.com/5Lg5xypG | 22:44 |
| TheJulia | also, if you `grep "auth_strategy" /etc/ironic/ironic.conf` everything should be set to noauth | 22:44 |
| aNupoisc | cinerama: TheJulia: haha having so many pastes | 22:44 |
| cinerama | aNupoisc, just the date of the last change in the git log, or the subject would be helpful | 22:45 |
| aNupoisc | cinerama: commit dfcfcc22a843653c372d1ba8cf7fea757d09b7b | 22:45 |
| aNupoisc | TheJulia: yes everything is set to noauth | 22:46 |
| aNupoisc | is it possible to screen share? | 22:46 |
| TheJulia | when was this install performed? | 22:46 |
| aNupoisc | cinerama TheJulia | 22:46 |
| cinerama | aNupoisc, that's from back in january | 22:46 |
| cinerama | TheJulia, remember the oslo config changes to sample config | 22:46 |
| gmmaha | cinerama: TheJulia: i work with aNupoisc and this was the setup i had back in Jan and it worked then.. Pretty much tearing it down and re-creating it again.. | 22:47 |
| TheJulia | oh hey gmmaha! | 22:47 |
| gmmaha | howdy TheJulia | 22:47 |
| cinerama | hi gmmaha | 22:47 |
| gmmaha | hey cinerama | 22:47 |
| aNupoisc | yes | 22:47 |
| gmmaha | i am stuck elsewhere and unable to help aNupoisc debug this.. hence us bothering you guys a little.. | 22:47 |
| TheJulia | so, if you guys ought to pull down the latest state of master and apply your patches on top | 22:48 |
| gmmaha | yeah.. should ahve done that a while back.. been procrastinating :| | 22:48 |
| TheJulia | or at least stable/mitaka, You may also just want to move your /etc/ironic/ironic.conf file out of the way | 22:48 |
| aNupoisc | TheJulia: yeah may be i can try that | 22:48 |
| TheJulia | there have been a number of config changes recently and we've had to fix a few of our regular expressions, hence the templating change noted in earlier discussion that we're contemplating | 22:49 |
| cinerama | i don't see anything troublesome in your diff but you may be having difficulty if you're using a older/stale bifrost with newer ironic | 22:50 |
| gmmaha | cinerama: aaaah.. never thought about that.. hmmmmm | 22:50 |
| cinerama | just because of changes to ironic | 22:50 |
| TheJulia | cinerama: better put than I | 22:50 |
| gmmaha | cinerama: true that.. | 22:50 |
| gmmaha | TheJulia: cinerama: thanks.. we will try nd update the patch to stable/mitaka and see how it plays out | 22:51 |
| cinerama | now we do have a stable branch but if you're not using stable ironic it won't help you | 22:51 |
| TheJulia | the hope is to avoid such things, but they are inevitable sometimes :) | 22:51 |
| aNupoisc | gmmaha cinerama TheJulia yes i am tearing down and will clone stable/mitika | 22:51 |
| gmmaha | TheJulia: :) | 22:52 |
| TheJulia | aNupoisc: we have a specific way of forcing it to clone stable/mitaka for everything, one moment | 22:52 |
| gmmaha | cinerama: true that.. we will try and jump onto the stable rlease and see how it plays out! | 22:52 |
| cinerama | well you don't necessarily have to use stable (unless it's in your other requirements), just if you're going bleeding edge on one, you need to for both | 22:53 |
| TheJulia | https://github.com/openstack/bifrost/blob/master/playbooks/roles/bifrost-prep-for-install/defaults/main.yml#L15 | 22:54 |
| TheJulia | that will make your life much happier | 22:54 |
| TheJulia | if you update those to stable/mitaka | 22:54 |
| TheJulia | We pull the clone because there are cases where you don't get things like... the sample config :) | 22:54 |
| gmmaha | TheJulia: aah.. yes, best to match them both.. saves us quite the pain | 22:55 |
| TheJulia | This is where you can tell, I _really_ like having the sample config contet | 22:55 |
| TheJulia | context | 22:55 |
| gmmaha | cinerama: yes, you are right.. best to match both and save us all some trouble :) | 22:56 |
| gmmaha | TheJulia: +1 on the config context | 22:57 |
| aNupoisc | cinerama TheJulia ironicclient_git_branch well set | 22:57 |
| TheJulia | I can't tell you the number of times I've opened a managed config file with 0 context and have facepalmed and groaned over the years | 22:57 |
| TheJulia | aNupoisc: you'll want to do line 16 as well | 22:58 |
| gmmaha | TheJulia: :) | 22:58 |
| TheJulia | line 17/18... I would be less worried about.... 17 actually might not have a stable branch now that I think about it since it is an infra library | 22:58 |
| aNupoisc | TheJulia: ah yeah :) | 22:58 |
| TheJulia | 18... may also not, checking | 22:59 |
| TheJulia | nope, 18 does not | 22:59 |
| aNupoisc | TheJulia: haha okay 15,16 seems very imp in this case | 23:02 |
| TheJulia | indeed :) | 23:02 |
| TheJulia | aNupoisc: you could alternatively set tags, but I don't think they are really worth worrying about too much :) | 23:03 |
| TheJulia | at least, for the items other than 15,16 | 23:03 |
| TheJulia | Anyway, sounds like we're on a happier path, just make sure you mv your old /etc/ironic/ironic.conf to /etc/ironic/ironic.conf.old. If we get the same error, then something else is in play here, but I suspect we may have something somewhere that is not playing well config wise :( | 23:05 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [inspector] https://review.openstack.org/301955 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [iscsi] https://review.openstack.org/309615 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [pxe] https://review.openstack.org/309612 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [seamicro] https://review.openstack.org/303113 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [agent] https://review.openstack.org/309603 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [ssh] https://review.openstack.org/303649 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [deploy] https://review.openstack.org/309206 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [ipmi] https://review.openstack.org/301967 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [dhcp] https://review.openstack.org/301830 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [swift] https://review.openstack.org/303653 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [console] https://review.openstack.org/295946 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [virtualbox] https://review.openstack.org/303655 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [glance] https://review.openstack.org/304829 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [DEFAULT] https://review.openstack.org/309070 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [api] https://review.openstack.org/309186 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [database] https://review.openstack.org/301815 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [irmc] https://review.openstack.org/302476 | 23:06 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [iboot] https://review.openstack.org/301856 | 23:06 |
| aNupoisc | TheJulia: yes i will keep name changes in mind. Hope version change solve the problem :) | 23:06 |
| * TheJulia blinks | 23:06 | |
| *** ayoung has quit IRC | 23:07 | |
| TheJulia | Yeah stuff like ^^^^ :( | 23:07 |
| *** absubram has quit IRC | 23:07 | |
| *** ayoung has joined #openstack-ironic | 23:07 | |
| TheJulia | aNupoisc: please let us know if you have any issues. I don't know how long cinerama will be around this evening, but it is like two hours after I should have called it a day, so have a wonderful evening! :) | 23:08 |
| cinerama | i will be around for a bit longer | 23:10 |
| *** ChubYann has quit IRC | 23:18 | |
| aNupoisc | TheJulia cinerama: yes of course :) even i will be here a bit long | 23:23 |
| aNupoisc | have nice evening both of youu | 23:23 |
| *** jaybeale has quit IRC | 23:24 | |
| *** PollyZ has quit IRC | 23:27 | |
| *** ChubYann has joined #openstack-ironic | 23:30 | |
| *** ayoung has quit IRC | 23:34 | |
| *** thrash is now known as thrash|g0ne | 23:38 | |
| aNupoisc | cinerama TheJulia so i have cloned a stable/mitika but unable to apply the patch over it | 23:51 |
| aNupoisc | so taking a bit to solve the issue | 23:51 |
| aNupoisc | so if the file that is being changed by the patch, if that file has changed then will that make a problem | 23:52 |
| *** wajdi_ has joined #openstack-ironic | 23:53 | |
| *** Egyptian[Home] has joined #openstack-ironic | 23:58 | |
| *** rama_y has quit IRC | 23:58 | |
| *** ChrisAusten has joined #openstack-ironic | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!