| *** eghobo has joined #openstack-ironic | 00:03 | |
| *** todd_dsm has joined #openstack-ironic | 00:08 | |
| *** marcoemorais has quit IRC | 00:19 | |
| *** shakamunyi has joined #openstack-ironic | 00:21 | |
| *** marcoemorais has joined #openstack-ironic | 00:21 | |
| *** marcoemorais has quit IRC | 00:27 | |
| *** marcoemorais has joined #openstack-ironic | 00:27 | |
| *** rloo has quit IRC | 00:35 | |
| *** Kui has joined #openstack-ironic | 00:37 | |
| *** todd_dsm has quit IRC | 00:39 | |
| *** todd_dsm has joined #openstack-ironic | 00:40 | |
| *** marcoemorais has quit IRC | 00:43 | |
| *** shakayumi has joined #openstack-ironic | 00:43 | |
| *** shakamunyi has quit IRC | 00:43 | |
| *** rwsu has quit IRC | 00:49 | |
| *** harlowja_away is now known as harlowja | 00:49 | |
| *** chuckC has quit IRC | 00:53 | |
| *** todd_dsm has quit IRC | 00:58 | |
| *** marcoemorais has joined #openstack-ironic | 01:01 | |
| *** ChuckC_ has quit IRC | 01:04 | |
| *** athomas has quit IRC | 01:07 | |
| *** marcoemorais has quit IRC | 01:15 | |
| *** rwsu has joined #openstack-ironic | 01:15 | |
| *** eghobo has quit IRC | 01:16 | |
| *** marcoemorais has joined #openstack-ironic | 01:18 | |
| *** marcoemorais has quit IRC | 01:19 | |
| *** marcoemorais has joined #openstack-ironic | 01:20 | |
| *** marcoemorais has quit IRC | 01:24 | |
| *** pcrews has quit IRC | 01:25 | |
| *** marcoemorais has joined #openstack-ironic | 01:26 | |
| *** pcrews has joined #openstack-ironic | 01:27 | |
| openstackgerrit | Yuiko Takada proposed a change to openstack/python-ironicclient: Add "ironic node-set-power-state" cmd unit test https://review.openstack.org/124267 | 01:30 |
|---|---|---|
| *** mikedillion has joined #openstack-ironic | 01:33 | |
| *** ChuckC_ has joined #openstack-ironic | 01:34 | |
| *** chuckC has joined #openstack-ironic | 01:35 | |
| *** marcoemorais has quit IRC | 01:39 | |
| *** foexle_ has joined #openstack-ironic | 01:39 | |
| *** foexle has quit IRC | 01:43 | |
| *** mikedillion has quit IRC | 01:47 | |
| *** shakayumi has quit IRC | 01:49 | |
| *** dlaube has quit IRC | 01:53 | |
| *** nosnos has joined #openstack-ironic | 01:56 | |
| *** shakamunyi has joined #openstack-ironic | 02:16 | |
| *** shakamunyi has quit IRC | 02:49 | |
| *** ramineni has joined #openstack-ironic | 02:49 | |
| *** shakamunyi has joined #openstack-ironic | 02:49 | |
| *** shakayumi has joined #openstack-ironic | 02:52 | |
| *** shakamunyi has quit IRC | 02:56 | |
| *** pcrews has quit IRC | 02:56 | |
| *** shakayumi has quit IRC | 02:56 | |
| *** alexiz has joined #openstack-ironic | 02:57 | |
| *** pcrews has joined #openstack-ironic | 02:57 | |
| *** pcrews has quit IRC | 03:04 | |
| *** pcrews has joined #openstack-ironic | 03:21 | |
| *** nosnos has quit IRC | 03:22 | |
| *** nosnos has joined #openstack-ironic | 03:23 | |
| *** Poornima has joined #openstack-ironic | 03:24 | |
| *** harlowja is now known as harlowja_away | 03:26 | |
| *** nosnos has quit IRC | 03:27 | |
| *** penick has joined #openstack-ironic | 03:29 | |
| *** penick_ has joined #openstack-ironic | 03:29 | |
| *** harlowja_away is now known as harlowja | 03:32 | |
| *** penick has quit IRC | 03:33 | |
| *** penick_ is now known as penick | 03:33 | |
| *** rameshg87 has joined #openstack-ironic | 03:42 | |
| *** nikunj2512 has joined #openstack-ironic | 03:50 | |
| *** rameshg871 has joined #openstack-ironic | 03:59 | |
| *** pcrews has quit IRC | 03:59 | |
| *** rameshg87 has quit IRC | 04:01 | |
| *** rushiagr_away is now known as rushiagr | 04:01 | |
| *** marcoemorais has joined #openstack-ironic | 04:03 | |
| *** marcoemorais1 has joined #openstack-ironic | 04:05 | |
| *** marcoemorais has quit IRC | 04:08 | |
| *** rameshg871 has quit IRC | 04:09 | |
| *** nosnos has joined #openstack-ironic | 04:14 | |
| *** shikui_ has joined #openstack-ironic | 04:30 | |
| *** jrist has quit IRC | 04:32 | |
| *** Kui has quit IRC | 04:33 | |
| *** jrist has joined #openstack-ironic | 04:35 | |
| *** rushiagr is now known as rushiagr_away | 04:51 | |
| *** pensu has joined #openstack-ironic | 04:56 | |
| *** alexiz has quit IRC | 05:00 | |
| *** loki184 has joined #openstack-ironic | 05:01 | |
| *** rakesh_hs has joined #openstack-ironic | 05:13 | |
| *** rushiagr_away is now known as rushiagr | 05:32 | |
| *** penick has quit IRC | 05:59 | |
| *** harlowja is now known as harlowja_away | 06:05 | |
| *** rwsu has quit IRC | 06:09 | |
| *** k4n0 has joined #openstack-ironic | 06:27 | |
| *** nosnos has quit IRC | 06:39 | |
| *** nosnos has joined #openstack-ironic | 06:40 | |
| *** nosnos has quit IRC | 06:41 | |
| *** nosnos_ has joined #openstack-ironic | 06:41 | |
| GheRivero | morning all | 06:43 |
| *** jcoufal has joined #openstack-ironic | 06:47 | |
| *** jcoufal has quit IRC | 06:47 | |
| openstackgerrit | Anusha Ramineni proposed a change to openstack/ironic: ilo* drivers to use only ilo credentials https://review.openstack.org/124704 | 06:47 |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Imported Translations from Transifex https://review.openstack.org/124953 | 06:47 |
| *** GheRivero has quit IRC | 06:52 | |
| *** GheRivero has joined #openstack-ironic | 06:58 | |
| *** AJaeger has joined #openstack-ironic | 06:59 | |
| AJaeger | devananda, https://review.openstack.org/#/c/124953/ looks fine - the single french file has one string and that is translated... | 07:00 |
| AJaeger | devananda, 124953 is the translation import | 07:00 |
| AJaeger | devananda, but I don't think this is critical to integrate... | 07:00 |
| *** igordcard has joined #openstack-ironic | 07:08 | |
| *** ifarkas has joined #openstack-ironic | 07:18 | |
| *** romcheg has joined #openstack-ironic | 07:20 | |
| *** foexle_ has quit IRC | 07:21 | |
| *** jcoufal has joined #openstack-ironic | 07:22 | |
| *** jcoufal has quit IRC | 07:23 | |
| *** jcoufal has joined #openstack-ironic | 07:23 | |
| *** ndipanov_gone is now known as ndipanov | 07:27 | |
| *** ifarkas has quit IRC | 07:32 | |
| *** jcoufal has quit IRC | 07:32 | |
| *** ifarkas has joined #openstack-ironic | 07:32 | |
| *** ifarkas has quit IRC | 07:33 | |
| *** ifarkas has joined #openstack-ironic | 07:33 | |
| *** jcoufal has joined #openstack-ironic | 07:34 | |
| *** jcoufal has quit IRC | 07:40 | |
| *** jcoufal has joined #openstack-ironic | 07:40 | |
| *** vinbs has joined #openstack-ironic | 07:48 | |
| *** athomas has joined #openstack-ironic | 07:55 | |
| *** jistr has joined #openstack-ironic | 07:55 | |
| *** MattMan has quit IRC | 07:56 | |
| *** MattMan has joined #openstack-ironic | 07:56 | |
| *** andreykurilin_ has joined #openstack-ironic | 07:57 | |
| *** lsmola has joined #openstack-ironic | 07:59 | |
| *** loki184 has quit IRC | 08:02 | |
| *** derekh has joined #openstack-ironic | 08:03 | |
| *** dtantsur|afk is now known as dtantsur | 08:04 | |
| dtantsur | Morning Ironic | 08:04 |
| GheRivero | morning all | 08:05 |
| GheRivero | anybody knows what's going on with check-tripleo-ironic-undercloud-precise-nonha? | 08:06 |
| *** pensu has quit IRC | 08:14 | |
| *** pensu has joined #openstack-ironic | 08:15 | |
| *** loki184 has joined #openstack-ironic | 08:23 | |
| *** lucasagomes has joined #openstack-ironic | 08:23 | |
| romcheg | Morning everyone! | 08:28 |
| *** ramineni1 has joined #openstack-ironic | 08:32 | |
| *** ramineni has quit IRC | 08:34 | |
| *** ramineni1 has quit IRC | 08:38 | |
| *** andreykurilin_ has quit IRC | 08:44 | |
| *** viktors|afk is now known as viktors | 08:44 | |
| *** marcoemorais1 has quit IRC | 08:47 | |
| *** foexle has joined #openstack-ironic | 08:49 | |
| dtantsur | romcheg, GheRivero, morning | 08:54 |
| dtantsur | GheRivero, no idea, was not paying attention recently | 08:54 |
| derekh | Hi all, the ironic jobs has been failing since last night bug 1375641 , the novabm jobs are ok, any ides ? | 09:17 |
| dtantsur | derekh, wow. anything interesting in ironic logs? | 09:25 |
| dtantsur | WARNING wsme.api [-] Client-side error: Node 3ced5637-d168-4d89-8074-f8002154e47c can not be updated while a state transition is in progress. | 09:26 |
| dtantsur | hmmm... | 09:26 |
| dtantsur | and completely nothing in conductor logs | 09:27 |
| dtantsur | lucasagomes, feel like joining our party ^^^? | 09:27 |
| * lucasagomes reads | 09:28 | |
| lucasagomes | 1 sec just finishing a comment in a review | 09:28 |
| *** yuanying has quit IRC | 09:28 | |
| openstackgerrit | Derek Higgins proposed a change to openstack/ironic: Revert "Conductor changes target_power_state before starting work" https://review.openstack.org/124990 | 09:29 |
| *** yuanying has joined #openstack-ironic | 09:29 | |
| derekh | dtantsur: just the log you pasted | 09:29 |
| dtantsur | derekh, I also suspect ^^^ is related | 09:29 |
| derekh | dtantsur: lucasagomes ^^ I'm trying that revert, it failed tripleo ci with the same error and merged around the same time the problem started | 09:30 |
| lucasagomes | derekh, oh! ack... lemme take a look at the logs too, but yeah let's revert to unbroke u guys | 09:31 |
| derekh | lucasagomes: Ok, we'll know in just under an hour, its not just us that are broke, ironic is too ;-) | 09:32 |
| lucasagomes | yeah indeed | 09:33 |
| lucasagomes | :( | 09:33 |
| dtantsur | updated bug as affecting Ironic | 09:33 |
| *** yuanying has quit IRC | 09:34 | |
| * dtantsur does not feel comfortable that we introduce so many serious changes just before rc1 :( | 09:34 | |
| derekh | The tripleo ci is fairly reliable these day, paying attention to the results should help | 09:35 |
| dtantsur | derekh, did we decide on making it voting some day? | 09:37 |
| dtantsur | that would help a lot | 09:37 |
| openstackgerrit | A change was merged to openstack/ironic: Add HashRingManager to wrap hash ring singleton https://review.openstack.org/120951 | 09:38 |
| derekh | dtantsur: ok, so this is where it gets complicated, technically it is voting but in a different queue to the other ci jobs (and not in gate) so it easily be ignored | 09:39 |
| derekh | dtantsur: but somehow enforcing the vote would be better | 09:39 |
| dtantsur | right | 09:39 |
| lucasagomes | yeah, if we could get it enforcing the vote in all projects that tripleo touches, otherwise if we do only in Ironic we will have the risk of get broken by things not related to ironic at all | 09:44 |
| dtantsur | lucasagomes, I think I detected error in the power state patch | 09:45 |
| dtantsur | node['target_power_state'] = states.NOSTATE should be https://review.openstack.org/#/c/124225/4/ironic/conductor/utils.py on line 87 | 09:45 |
| *** ramineni has joined #openstack-ironic | 09:45 | |
| lucasagomes | dtantsur, oh, right, I'm looking into it now | 09:46 |
| dtantsur | derekh, lucasagomes lemme propose alternative patch and see of it fixes the problem | 09:46 |
| lucasagomes | dtantsur, oh! makes sense | 09:47 |
| lucasagomes | because it already has that target state but then it does nothing (just returns) without cleaning that up | 09:47 |
| lucasagomes | tho the manager should unset that :/ lemme see | 09:48 |
| openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic: Always reset target_power_state in node_power_action https://review.openstack.org/124994 | 09:50 |
| dtantsur | lucasagomes, derekh ^^^ | 09:51 |
| dtantsur | now let's wait for the job to pass/not to pass | 09:52 |
| lucasagomes | yeah, it looks like that bug was there b4 | 09:53 |
| lucasagomes | odd | 09:53 |
| dtantsur | lucasagomes, no, previously target_power_state was set after that code | 10:00 |
| dtantsur | line 102 https://review.openstack.org/#/c/124225/4/ironic/conductor/utils.py | 10:00 |
| dtantsur | so on early return target_power_state was still NOSTATE | 10:01 |
| dtantsur | after devananda's patch it's no longer the first place to set target_power_state | 10:01 |
| lucasagomes | dtantsur, +111 | 10:01 |
| lucasagomes | maybe we should put a big ass try...finally there to unset that | 10:03 |
| lucasagomes | gosh | 10:03 |
| *** yuanying has joined #openstack-ironic | 10:08 | |
| dtantsur | lucasagomes, let's check our guesses and unbreak the gate. then we can think how to solve it the best way | 10:11 |
| lucasagomes | yup | 10:11 |
| lucasagomes | yeah, I think it's grand your patch might fix it | 10:11 |
| lucasagomes | but odd enough we have no unittests caughting that new behavior etc | 10:11 |
| openstackgerrit | A change was merged to openstack/ironic: Stop running check_uptodate in the pep8 testenv https://review.openstack.org/124919 | 10:13 |
| *** yuanying has quit IRC | 10:16 | |
| *** jcoufal has quit IRC | 10:23 | |
| *** jcoufal has joined #openstack-ironic | 10:24 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: For convention rename the first classmethod parameter cls https://review.openstack.org/125000 | 10:25 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: For convention rename the first classmethod parameter to cls https://review.openstack.org/125000 | 10:28 |
| *** sabah has joined #openstack-ironic | 10:31 | |
| lucasagomes | the revert seems to have worked, let's wait for the alternative patch | 10:33 |
| *** pelix has joined #openstack-ironic | 10:34 | |
| derekh | lucasagomes: dtantsur the fix worked aswell https://review.openstack.org/#/c/124994/ | 10:43 |
| lucasagomes | derekh, right, I think we should go with the fix then | 10:44 |
| derekh | lucasagomes: sounds ok to me | 10:44 |
| *** Poornima has quit IRC | 10:45 | |
| lucasagomes | derekh, ack! thanks | 10:45 |
| dtantsur | derekh, cool! | 10:49 |
| dtantsur | any cores around to approve the change? | 10:49 |
| lucasagomes | yuriyz, Haomeng|2 maybe | 10:49 |
| yuriyz | morning Ironic | 10:50 |
| dtantsur | yuriyz, morning! we need your help approving https://review.openstack.org/#/c/124994/ to unbreak tripleo jobs | 10:51 |
| dtantsur | thanks | 10:53 |
| lucasagomes | morning yuriyz | 10:55 |
| *** romcheg has quit IRC | 10:56 | |
| *** lucasagomes is now known as lucas-hungry | 11:01 | |
| *** loki184 has quit IRC | 11:02 | |
| *** romcheg has joined #openstack-ironic | 11:07 | |
| *** ramineni has quit IRC | 11:08 | |
| *** yuanying has joined #openstack-ironic | 11:13 | |
| *** yuanying has quit IRC | 11:21 | |
| *** nikunj2512 has quit IRC | 11:23 | |
| *** mikal has quit IRC | 11:23 | |
| *** igordcard has quit IRC | 11:28 | |
| *** jistr is now known as jistr|english | 11:31 | |
| *** vdrok_ has joined #openstack-ironic | 11:31 | |
| *** mrda-vacation has quit IRC | 11:33 | |
| openstackgerrit | A change was merged to openstack/ironic: Always reset target_power_state in node_power_action https://review.openstack.org/124994 | 11:34 |
| *** mikal has joined #openstack-ironic | 11:35 | |
| dtantsur | lucas-hungry, derekh ^^^ | 11:37 |
| *** vinbs has quit IRC | 11:37 | |
| derekh | dtantsur: great, thanks | 11:38 |
| *** dhellmann has quit IRC | 11:48 | |
| *** dhellmann has joined #openstack-ironic | 11:50 | |
| *** lazy_prince2 has joined #openstack-ironic | 11:55 | |
| *** romcheg has quit IRC | 12:02 | |
| dtantsur | short walk, brb | 12:04 |
| openstackgerrit | A change was merged to openstack/ironic: For convention rename the first classmethod parameter to cls https://review.openstack.org/125000 | 12:10 |
| *** igordcard has joined #openstack-ironic | 12:11 | |
| openstackgerrit | A change was merged to openstack/ironic: Force the SSH commands to use their default language https://review.openstack.org/124061 | 12:16 |
| *** yuanying has joined #openstack-ironic | 12:18 | |
| *** dprince has joined #openstack-ironic | 12:18 | |
| *** romcheg has joined #openstack-ironic | 12:26 | |
| *** yuanying has quit IRC | 12:27 | |
| *** foexle has quit IRC | 12:30 | |
| *** ifarkas has quit IRC | 12:33 | |
| *** ifarkas has joined #openstack-ironic | 12:35 | |
| lucas-hungry | dtantsur, w00t | 12:39 |
| *** lucas-hungry is now known as lucasagomes | 12:39 | |
| *** rushiagr is now known as rushiagr_away | 12:51 | |
| *** jistr|english is now known as jistr | 12:56 | |
| *** k4n0 has quit IRC | 13:01 | |
| *** shikui_ has quit IRC | 13:03 | |
| *** krtaylor has quit IRC | 13:06 | |
| *** krtaylor has joined #openstack-ironic | 13:07 | |
| *** lazy_prince2 has quit IRC | 13:09 | |
| *** rloo has joined #openstack-ironic | 13:13 | |
| *** ramineni has joined #openstack-ironic | 13:13 | |
| *** pensu has quit IRC | 13:15 | |
| *** igordcard has quit IRC | 13:16 | |
| *** igordcard has joined #openstack-ironic | 13:17 | |
| *** yuanying has joined #openstack-ironic | 13:23 | |
| *** linggao has joined #openstack-ironic | 13:24 | |
| *** rushiagr_away is now known as rushiagr | 13:24 | |
| *** ramineni has quit IRC | 13:27 | |
| *** pensu has joined #openstack-ironic | 13:30 | |
| *** yuanying has quit IRC | 13:32 | |
| *** ramineni has joined #openstack-ironic | 13:35 | |
| *** rushiagr is now known as rushiagr_away | 13:36 | |
| *** sabah has quit IRC | 13:37 | |
| jroll | morning ironic :) | 13:39 |
| dtantsur | jroll, morning | 13:41 |
| lucasagomes | jroll, morning | 13:45 |
| *** jasondotstar has joined #openstack-ironic | 13:46 | |
| rloo | morning jroll, afternoon dtantsur & lucasagomes | 13:47 |
| *** DCWilliams_VA has joined #openstack-ironic | 13:47 | |
| dtantsur | rloo, hi | 13:48 |
| lucasagomes | rloo, morning :) | 13:49 |
| GheRivero | morning @ironic | 13:51 |
| romcheg | Morning rloo, jroll! | 13:51 |
| *** DCWilliams_VA has left #openstack-ironic | 13:53 | |
| rloo | afternoon GheRivero, romcheg! | 13:53 |
| *** athomas has quit IRC | 13:59 | |
| *** nosnos_ has quit IRC | 13:59 | |
| *** nosnos has joined #openstack-ironic | 13:59 | |
| *** nosnos has quit IRC | 14:04 | |
| *** ramineni1 has joined #openstack-ironic | 14:04 | |
| *** athomas has joined #openstack-ironic | 14:05 | |
| *** ramineni has quit IRC | 14:07 | |
| *** DCWilliams_VA has joined #openstack-ironic | 14:14 | |
| *** chuckC has quit IRC | 14:18 | |
| *** rushiagr_away is now known as rushiagr | 14:19 | |
| *** DCWilliams_VA has quit IRC | 14:19 | |
| MattMan | Hi, is there a best practice means of getting nova's baremetal* CLI to use Ironic instead of the built in nova-baremetal ? trying this on Icehouse. | 14:19 |
| lucasagomes | MattMan, we have a proxy in J for that | 14:20 |
| lucasagomes | but only for partial of the nova bm commands | 14:20 |
| lucasagomes | MattMan, https://review.openstack.org/#/c/120433 | 14:20 |
| devananda | g'mornig, all | 14:21 |
| MattMan | lucasagomes: thanks... I'll give that patch a shot. | 14:21 |
| lucasagomes | MattMan, cool, no worries | 14:23 |
| lucasagomes | devananda, morning | 14:23 |
| *** DCWilliams_VA has joined #openstack-ironic | 14:23 | |
| NobodyCam | good morning ironic | 14:24 |
| openstackgerrit | A change was merged to openstack/ironic: Refactor PXE clean up tests https://review.openstack.org/122137 | 14:26 |
| devananda | lucasagomes: hi! on https://review.openstack.org/#/c/124493/12/ironic/conductor/manager.py -- I can add tests for "is_active_on", but I wrote this so that it wouldn't affect existing behavior of register / unregister | 14:26 |
| *** ChuckC_ has quit IRC | 14:27 | |
| lucasagomes | devananda, right | 14:27 |
| lucasagomes | devananda, so, for the glance thing, we need a admin token to be passed within the context to glance | 14:27 |
| *** yuanying has joined #openstack-ironic | 14:28 | |
| lucasagomes | devananda, and about the tests, I think we should add some specially for that new update_existing behavior | 14:29 |
| devananda | lucasagomes: comments posted. from an API POV, the behavior of register/unregister didn't change | 14:31 |
| lucasagomes | devananda, cheers, ack | 14:31 |
| devananda | lucasagomes: but I agree that there should be a unit test now for the online column's behavior | 14:31 |
| lucasagomes | devananda, yeah | 14:32 |
| lucasagomes | devananda, for the auth_token thing, should we get a admin auth_token with ks before submitting the request to glance? :/ | 14:32 |
| lucasagomes | idk where I would get a auth-token from cause it's not a request, it's a periodic task | 14:33 |
| *** jcoufal has quit IRC | 14:33 | |
| devananda | right, i'll poke at that in a minute. how are you testing it? | 14:33 |
| lucasagomes | devananda, I've a devstack setup with a machine deployed | 14:34 |
| lucasagomes | 2 conductors with diff configs "host" and the tftpboot dir | 14:35 |
| lucasagomes | I kill the conductor that is currently managing that node and wait the periodic task to run on the other conductor | 14:35 |
| devananda | gotcha | 14:35 |
| lucasagomes | I think the auth token is the only missing bit to get something kinda working | 14:36 |
| lucasagomes | I mean working | 14:36 |
| devananda | lucasagomes: so I think we need to set up the keystone_authtoken config option properly for this | 14:36 |
| *** krtaylor has quit IRC | 14:36 | |
| lucasagomes | devananda, like we have in nova for the ironic cli? | 14:36 |
| devananda | I dont know if we've been configuring glance access for that token, since it didn't need it | 14:36 |
| devananda | not exactly, no | 14:36 |
| devananda | like we have in ironic for swift and neutron access | 14:36 |
| *** yuanying has quit IRC | 14:36 | |
| devananda | look at ironic/common/swift.py | 14:36 |
| lucasagomes | right, lemme take a look | 14:37 |
| *** krtaylor has joined #openstack-ironic | 14:37 | |
| *** jcoufal has joined #openstack-ironic | 14:37 | |
| devananda | lucasagomes: I think devstack is setting up that account correctly. devstack/lib/ironic:create_ironic_accounts() | 14:40 |
| lucasagomes | devananda, yeah... it's an admin | 14:41 |
| devananda | lucasagomes: you should see in your /etc/irnic/ironic.conf settings for [keystone_authtoken] admin_user, admin_password, etc. so that should be enough -- we just need to elevate to that | 14:41 |
| lucasagomes | I think we just need to check within our code, before passing the context to the glance cli | 14:41 |
| devananda | yep | 14:41 |
| lucasagomes | if the token contains an auth-token | 14:41 |
| lucasagomes | in case the strategy is keystone ofc | 14:41 |
| lucasagomes | and if not, get one | 14:42 |
| *** pcrews has joined #openstack-ironic | 14:42 | |
| lucasagomes | devananda, yeah, aight, looking into that now | 14:42 |
| *** krtaylor has quit IRC | 14:42 | |
| *** openstackgerrit has quit IRC | 14:42 | |
| lucasagomes | makes me wonder how it never was needed, maybe because the images were public | 14:43 |
| *** krtaylor has joined #openstack-ironic | 14:44 | |
| devananda | the normal path uses the nova context | 14:46 |
| *** krtaylor has quit IRC | 14:47 | |
| *** ramineni1 has quit IRC | 14:49 | |
| NobodyCam | devananda: see pm | 14:51 |
| *** DCWilliams_VA has quit IRC | 14:52 | |
| devananda | adam_g: https://review.openstack.org/#/c/124162/ merged into nova now, too. | 14:55 |
| devananda | adam_g: so we ought to see more stability in the parallel tempest runs | 14:55 |
| *** ramineni has joined #openstack-ironic | 14:55 | |
| *** DCWilliams_VA has joined #openstack-ironic | 14:56 | |
| *** ramineni1 has joined #openstack-ironic | 14:56 | |
| devananda | lucasagomes: re the config option in https://review.openstack.org/#/c/124610/ - i pinged jroll yesterday, but i'm not sure we reached a conclusion. he seemed to want the config option | 14:56 |
| *** krtaylor has joined #openstack-ironic | 14:57 | |
| devananda | lucasagomes: namely because they had to tune the sync power state task. | 14:57 |
| devananda | jroll: any further thoughts on that? | 14:57 |
| lucasagomes | right let's discuss it... I don't think it may harm to offer a way to ppl tune ir up or down, or even disable the periodic task | 14:57 |
| lucasagomes | it doesn't affect the agent, if a cluster has only that driver running | 14:58 |
| lucasagomes | why should we have that periodic task | 14:58 |
| devananda | true | 14:59 |
| *** ramineni has quit IRC | 15:00 | |
| *** todd_dsm has joined #openstack-ironic | 15:00 | |
| *** rakesh_hs has quit IRC | 15:00 | |
| *** rwsu has joined #openstack-ironic | 15:16 | |
| *** ChuckC_ has joined #openstack-ironic | 15:22 | |
| *** openstackgerrit has joined #openstack-ironic | 15:24 | |
| *** chuckC has joined #openstack-ironic | 15:25 | |
| rloo | hi devananda, do you have a minute to explain node.is_active_on? (wrt https://review.openstack.org/#/c/124493) | 15:27 |
| rloo | devananda: it is set to the conductor id when deploying, and unset when tearing down. | 15:27 |
| *** jcoufal has quit IRC | 15:27 | |
| devananda | rloo: yup | 15:27 |
| rloo | devananda: if deploy fails, would/should it be unset? | 15:27 |
| rloo | devananda: and if it is unset when tear down, how does that help with the comment 'clear... so that no other conductor will recreate the deploy env...'? | 15:28 |
| rloo | devananda: if it isn't set, won't another conductor recreate the deploy env? or maybe I'm confused. | 15:28 |
| devananda | if a node doesn't have an active state, no conductor will rebuild it to active state | 15:29 |
| devananda | if a ndoe has an active state, but the conductor it was last active on is dead | 15:29 |
| devananda | then another conductor should rebuild its state | 15:29 |
| devananda | so when is_active_on is set to a not-active conductor | 15:30 |
| devananda | what ever cnoductor that node currently maps to should take_over() | 15:30 |
| rloo | devananda: ahh. got it. the take_over scenario. thx. | 15:31 |
| devananda | perhaps the column name should be changed? last_active_on? activity_managed_by? state_cached_on? | 15:31 |
| *** arif-ali has joined #openstack-ironic | 15:31 | |
| rloo | devananda: well, it is 'currently-active' or what it thinks it is currently-active-on? | 15:31 |
| rloo | devananda: yeah, so I guess I don't like the 'last' part. | 15:32 |
| devananda | it's not what the node thinks -- it is the id of the last recorded conductor to have prepared the environment | 15:32 |
| *** coolsvap|afk is now known as coolsvap | 15:32 | |
| *** DCWilliams_VA has quit IRC | 15:33 | |
| *** yuanying has joined #openstack-ironic | 15:33 | |
| *** dtantsur is now known as dtantsur|afk | 15:34 | |
| rloo | devananda: if this is only meant to track what conductor has local state for a node, maybe node.local_state_on? | 15:35 |
| rloo | node.conductor_with_local_state? | 15:36 |
| *** krtaylor has quit IRC | 15:37 | |
| *** krtaylor has joined #openstack-ironic | 15:38 | |
| *** todd_dsm has quit IRC | 15:41 | |
| *** yuanying has quit IRC | 15:42 | |
| *** hemna has joined #openstack-ironic | 15:44 | |
| *** ndipanov is now known as ndipanov_gone | 15:51 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add periodic task to rebuild conductor local state https://review.openstack.org/124610 | 15:57 |
| lucasagomes | devananda, ^ | 15:57 |
| lucasagomes | devananda, one thing, diff than other clients, the glance client doesn't seem to accept the username, password ... | 15:58 |
| lucasagomes | https://github.com/openstack/python-glanceclient/blob/master/glanceclient/v1/client.py#L25-L29 | 15:58 |
| lucasagomes | like we do for neutron https://github.com/openstack/python-neutronclient/blob/master/neutronclient/client.py#L87-L102 | 15:58 |
| lucasagomes | so I get the auth_token with keystone and populate it in the context | 15:58 |
| lucasagomes | seems to be working, it regenerates all the configs, cache the images etc | 16:00 |
| jroll | devananda, lucasagomes, on the periodic task... someone running the agent would still need that task in case a conductor crashes, no? the request still gets forwarded to a conductor, if it forwards to a conductor that isn't there, there will be problems | 16:01 |
| devananda | jroll: no | 16:01 |
| *** ifarkas has quit IRC | 16:02 | |
| jroll | because you're now basically storing the hash ring in the db aiui | 16:02 |
| jroll | maybe not | 16:02 |
| lucasagomes | jroll, that part is already done | 16:02 |
| lucasagomes | jroll, if the conductor dies, the api will redirect the request to another conductor that has that driver and is active | 16:02 |
| jroll | ok, so we're going to continue to calculate the ring on every request then? | 16:03 |
| lucasagomes | what we are doing is more the take over, to prepare the env for the new conductor to be able to manage the node it took over | 16:03 |
| jroll | right | 16:03 |
| lucasagomes | jroll, for now I think so, do we cache the hash_ring now | 16:03 |
| devananda | jroll: i dont want to, but to preserve as much functionality as possible at THIS point in the cycle, yes | 16:03 |
| lucasagomes | tho* | 16:03 |
| devananda | we do not cache the hash ring on the API side | 16:03 |
| jroll | ok, then would be cool to disable the periodic task | 16:03 |
| devananda | and when receiving RPC requests, conductors arenot rejecting things that dont map there | 16:03 |
| devananda | since they aren't checking the mapping at that ppoint | 16:04 |
| jroll | I would also think it would be nice to make the interval configurable, if you're doing 100+ nodes per conductor the power loop can get very slow | 16:04 |
| devananda | when a conductor shuts down, it deregisters, so the API services will see that right away | 16:04 |
| devananda | when it crashes, there'll be a delay before API services see that it's offline | 16:04 |
| devananda | ... /me confirms that | 16:04 |
| jroll | yeah, agree | 16:04 |
| jroll | there's a heartbeat timeout | 16:05 |
| lucasagomes | yeah seems correct | 16:05 |
| devananda | yep | 16:05 |
| devananda | get_active_driver_dict considers both the heartbeat and online fields | 16:05 |
| lucasagomes | devananda, I can confirm that the images are being cached and the configs are generated correctly | 16:05 |
| lucasagomes | now I'm a bit stuck with the neutron bit | 16:05 |
| devananda | nice! | 16:05 |
| devananda | ok, stuck on what? | 16:05 |
| lucasagomes | it does call update neutron, but it still pointing to the old tftpdir! | 16:06 |
| devananda | oh | 16:06 |
| jroll | odd | 16:06 |
| lucasagomes | all I changed in my conf for the second conductor was host, tftp_master_path = /opt/stack/data/ironic/tftpboot2/master_images tftp_root = /opt/stack/data/ironic/tftpboot2 | 16:06 |
| jroll | can you tell if that's what neutron is responding to dhcp requests with | 16:06 |
| jroll | ? | 16:06 |
| lucasagomes | you know if there's any other bit I need to change? or if there's a problem because both conductors are running on the same ip | 16:07 |
| lucasagomes | so neutron just ignore the update | 16:07 |
| jroll | idk, I would check neutron logs | 16:07 |
| jroll | and just to be sure, check that conductor two has restarted since you changed that conf :P | 16:07 |
| devananda | jroll: i think he's looking for the periodic task to update it | 16:08 |
| devananda | lucasagomes: neutron does have some issues with >1 thing sharing an IP | 16:08 |
| devananda | but IDK if that's the problem here | 16:08 |
| jroll | devananda: sounds like the task ran and called to neutron | 16:08 |
| devananda | lucasagomes: is conductor2 creating the new config files? | 16:09 |
| devananda | lucasagomes: nm, yhou said that :) | 16:09 |
| lucasagomes | yeah it has a new config file | 16:10 |
| lucasagomes | hmm I may need another vm to test it | 16:10 |
| lucasagomes | by looking at the logs it seems all good | 16:10 |
| devananda | so yea, it won't change in tjhis case | 16:10 |
| lucasagomes | will paste it for ye | 16:10 |
| devananda | lucasagomes: look at the options being sent to neutron | 16:10 |
| devananda | ironic/dhcp/neutron.py:138 | 16:10 |
| devananda | boot file name (pxelinux.0) and the IP | 16:11 |
| devananda | all the routing beyond that is done by MAC address and symlinks on the conductor | 16:11 |
| devananda | so to test this, we'd need conductors with different IPs | 16:11 |
| NobodyCam | brb | 16:12 |
| lucasagomes | http://paste.openstack.org/show/117187/ | 16:12 |
| lucasagomes | the error with the handler there is just a log interpolation, need to fix that too | 16:12 |
| lucasagomes | devananda, yeah :( right I will need a new vm for that | 16:12 |
| lucasagomes | urgh | 16:12 |
| lucasagomes | will take a while to me to confirm, today I've to leave a bit early (I'm moving houses so I need to go to the new place do some stuff) | 16:13 |
| devananda | do you have neutron's log from that time? | 16:13 |
| devananda | ack, nm | 16:13 |
| lucasagomes | devananda, yup may have | 16:13 |
| lucasagomes | 1 sec | 16:13 |
| lucasagomes | neutron log http://paste.openstack.org/show/117189/ | 16:15 |
| jroll | maybe look at /opt/stack/data/neutron/dhcp/e72cb063-425f-4afa-9f4e-8856199b9e5a/host | 16:16 |
| jroll | or everything in that uuid dir, even | 16:16 |
| lucasagomes | stack@vm:~/devstack$ cat /opt/stack/data/neutron/dhcp/e72cb063-425f-4afa-9f4e-8856199b9e5a/host | 16:17 |
| lucasagomes | fa:16:3e:a5:23:7b,host-10-1-0-3.openstacklocal,10.1.0.3 | 16:17 |
| lucasagomes | fa:16:3e:83:ab:40,host-10-1-0-1.openstacklocal,10.1.0.1 | 16:17 |
| lucasagomes | 52:54:00:48:5e:6b,host-10-1-0-4.openstacklocal,10.1.0.4,set:574c6e0a-b9af-42d2-af29-bf88bddea2d6 | 16:17 |
| lucasagomes | the dhcp options | 16:17 |
| lucasagomes | stack@vm:~/devstack$ cat /opt/stack/data/neutron/dhcp/e72cb063-425f-4afa-9f4e-8856199b9e5a/opts | 16:17 |
| lucasagomes | tag:tag0,option:router,10.1.0.1 | 16:17 |
| lucasagomes | tag:574c6e0a-b9af-42d2-af29-bf88bddea2d6,option:tftp-server,192.168.122.156 | 16:17 |
| lucasagomes | tag:574c6e0a-b9af-42d2-af29-bf88bddea2d6,option:bootfile-name,pxelinux.0 | 16:17 |
| lucasagomes | tag:574c6e0a-b9af-42d2-af29-bf88bddea2d6,option:server-ip-address,192.168.122.156 | 16:17 |
| lucasagomes | tag:tag0,option:dns-server,10.1.0.3stack@vm:~/devstack$ | 16:17 |
| *** marcoemorais has joined #openstack-ironic | 16:17 | |
| lucasagomes | maybe we just need to change the tftpd process | 16:18 |
| lucasagomes | but, a real test would need 2 diff nodes | 16:18 |
| lucasagomes | yeah that's the tftpd | 16:18 |
| lucasagomes | 7361 ? Ss 0:00 in.tftpd -v -v -v -v -v --map-file /opt/stack/data/ironic/tftpboot/map-file /opt/stack/data/ironic/tftpboot | 16:18 |
| jroll | hmm yeah, maybe each conductor needs a tftpd | 16:18 |
| lucasagomes | yeah | 16:18 |
| *** r-daneel has joined #openstack-ironic | 16:19 | |
| lucasagomes | they are running on the same node :/ | 16:19 |
| devananda | running >1 conductor on the same host isn't supported, heh | 16:19 |
| lucasagomes | yup | 16:19 |
| lucasagomes | but I couldn't find any way quicker to test :P | 16:19 |
| devananda | lucasagomes: it looks like your test has gone as far as it can with a single VM | 16:19 |
| devananda | cheers :) | 16:19 |
| lucasagomes | I can confirm that all the images and configs have been generated correctly | 16:19 |
| jroll | yeah | 16:19 |
| lucasagomes | yeah | 16:19 |
| jroll | lucasagomes: boot a node with ironic and run the conductor there | 16:19 |
| jroll | :P | 16:19 |
| devananda | heh | 16:20 |
| lucasagomes | hah will do | 16:20 |
| *** r-daneel has quit IRC | 16:20 | |
| lucasagomes | aight if nobody confirms it today, tomorrow I will setup the right env to make it sure it works | 16:20 |
| lucasagomes | (or try to) | 16:21 |
| Shrews | devananda: hrm, is_active_on should be set immediately during deploy, but i'm not seeing that in my test. curious | 16:21 |
| devananda | lucasagomes: cheers, thanks again. I'm going to give it a shot today, as I'll be distracted by meetings tomorrow | 16:22 |
| lucasagomes | devananda, right, no problem! lemme know the outcome (comment on the patch or something) | 16:22 |
| *** viktors is now known as viktors|afk | 16:23 | |
| *** ramineni1 has quit IRC | 16:25 | |
| *** r-daneel has joined #openstack-ironic | 16:28 | |
| *** jistr has quit IRC | 16:28 | |
| *** marcoemorais has quit IRC | 16:31 | |
| *** marcoemorais has joined #openstack-ironic | 16:31 | |
| *** lazy_prince is now known as killer_prince | 16:35 | |
| Shrews | devananda: ah, hrm. i think we need to update objects/node.py for the new column | 16:38 |
| devananda | Shrews: ah, yup | 16:38 |
| *** yuanying has joined #openstack-ironic | 16:38 | |
| Shrews | gotta run out for a bit. bbs | 16:40 |
| lucasagomes | Shrews, oh I put that on the following patch :/ we may need to move it to the previous one | 16:41 |
| *** marcoemorais has quit IRC | 16:42 | |
| lucasagomes | aight I will call it a day, have a great night everyone | 16:46 |
| *** lucasagomes is now known as lucas-afk | 16:46 | |
| NobodyCam | have a good night lucas-afk | 16:47 |
| *** yuanying has quit IRC | 16:47 | |
| *** athomas has quit IRC | 16:52 | |
| devananda | GheRivero: ping | 16:57 |
| devananda | vdrok: I was looking at 124400 and wondering why all the DB fixture code is in tests/base, not tests/db/base - but it looks just like legacy from when we forked from Nova | 16:58 |
| *** derekh has quit IRC | 16:58 | |
| devananda | vdrok: what do you think about separating the "needs admin context" from "needs a db fixture" tests? I think these are really seaprate things | 16:59 |
| devananda | also, running unit tests against mysql locally, I see only two INSERTs into one of our tables,a nd I dont see any ALTERs at all | 17:00 |
| *** krtaylor has quit IRC | 17:01 | |
| *** marcoemorais has joined #openstack-ironic | 17:05 | |
| openstackgerrit | A change was merged to openstack/ironic: Imported Translations from Transifex https://review.openstack.org/124953 | 17:07 |
| openstackgerrit | A change was merged to openstack/ironic: Remove useless variable in migration https://review.openstack.org/125089 | 17:16 |
| *** harlowja_away is now known as harlowja | 17:18 | |
| *** todd_dsm has joined #openstack-ironic | 17:28 | |
| *** coolsvap is now known as coolsvap|afk | 17:35 | |
| *** hemna has quit IRC | 17:39 | |
| *** yuanying has joined #openstack-ironic | 17:43 | |
| *** rushiagr is now known as rushiagr_away | 17:48 | |
| *** rushiagr_away is now known as rushiagr | 17:51 | |
| *** yuanying has quit IRC | 17:52 | |
| *** lsmola has quit IRC | 17:54 | |
| *** pelix has quit IRC | 17:55 | |
| *** todd_dsm has quit IRC | 17:59 | |
| *** todd_dsm has joined #openstack-ironic | 18:06 | |
| *** todd_dsm has quit IRC | 18:14 | |
| adam_g | anyone know why websockify is listed in ironics requirements.txt? | 18:24 |
| NobodyCam | adam_g: because openstack wanted it there: https://github.com/openstack/ironic/commit/9fe674e2dd31ac1fa94315d9f8b7cbca93c642f5 | 18:25 |
| Shrews | devananda: ok with me moving this https://review.openstack.org/#/c/124610/7/ironic/objects/node.py to 124493 for you? | 18:26 |
| NobodyCam | brb | 18:29 |
| adam_g | NobodyCam, hmm. silly | 18:31 |
| *** vdrok_ has quit IRC | 18:31 | |
| devananda | NobodyCam: openstack bot only updates requrements we have. it doesn'ta dd new ones | 18:32 |
| devananda | so someone wanted it there at some point | 18:32 |
| devananda | Shrews: yup, go for it | 18:32 |
| Shrews | devananda: looking closer, it's more code than that actually. not sure if we should move things around at this point | 18:33 |
| devananda | Shrews: i'm kinda side tracked by things right now, so poke away. lucas-afk also wanted some more unit tests on 124493 | 18:33 |
| devananda | hmm | 18:34 |
| *** openstackgerrit has quit IRC | 18:47 | |
| *** yuanying has joined #openstack-ironic | 18:48 | |
| *** openstackgerrit has joined #openstack-ironic | 18:49 | |
| *** yuanying has quit IRC | 18:56 | |
| *** andreykurilin_ has joined #openstack-ironic | 19:02 | |
| *** rushiagr is now known as rushiagr_away | 19:02 | |
| *** chuckC has quit IRC | 19:03 | |
| *** ChuckC_ is now known as ChuckC | 19:03 | |
| *** krtaylor has joined #openstack-ironic | 19:03 | |
| *** chuckC_ has joined #openstack-ironic | 19:03 | |
| *** blogan has joined #openstack-ironic | 19:03 | |
| blogan | does ironic use paste deploy? | 19:04 |
| *** annegentle has quit IRC | 19:04 | |
| *** hemna has joined #openstack-ironic | 19:05 | |
| Shrews | blogan: no. pecan | 19:06 |
| blogan | Shrews: Im pretty sure paste deploy can be used with pecan, though it might require some fenagling | 19:07 |
| Shrews | blogan: oh. if that's true, i've seen no indication of its use | 19:08 |
| blogan | Shrews: I'm just wondering if there was a big reason why ironic isn't using it | 19:08 |
| Shrews | blogan: we use what all other openstack projects use, so your question isn't ironic specific | 19:09 |
| jroll | Shrews: some projects use paste :P | 19:09 |
| jroll | e.g. neutron | 19:09 |
| blogan | Shrews: well from the 3 or 4 openstack projects I've been looking at closely, they all use paste deploy | 19:09 |
| Shrews | jroll: what all other "good" openstack projects use :-P | 19:09 |
| blogan | lol | 19:09 |
| blogan | i am totally fine with not using it | 19:09 |
| jroll | how dare you make fun of neutron | 19:09 |
| jroll | :P | 19:10 |
| * Shrews pastes jroll's tongue to the floor | 19:10 | |
| jroll | yeah, idk why ironic doesn't use it, but I'm happy with that | 19:10 |
| blogan | so does it seem like the trend is to not use paste? | 19:10 |
| jroll | don't really see what benefits it would bring us | 19:10 |
| blogan | just like the trend seems to be going towards pecan and wsme | 19:10 |
| jroll | indeed | 19:11 |
| jroll | but that's for actually running an api server, looks like paste deploy is just for finding/configuring them | 19:11 |
| blogan | so to get middleware, it looks like ironic is using the wrap_app keyword in pecan's make_app method, is that correct? | 19:11 |
| jroll | something like that | 19:12 |
| blogan | i guess I see a reason to use paste when you want to add a configurable pipeline, and swap in middleware, but I should just hold off on using it until we actually need something like that | 19:13 |
| blogan | thanks for the info jroll and Shrews | 19:13 |
| jroll | np | 19:14 |
| jroll | I feel like you can add middleware somewhat easily | 19:14 |
| * jroll looks | 19:14 | |
| jroll | maybe not | 19:14 |
| blogan | i'd need to research pecan's wrap_app functionality | 19:16 |
| devananda | blogan: can you describe why we would use pastedeploy as oposed to what we have today? | 19:23 |
| devananda | blogan: fwiw, we are using pecan/wsme right now, and afaik using paste deploy hasn't been suggested before | 19:24 |
| blogan | devananda: i'm not sure what you have today isn't able to do everything paste deploy can't, which is why I'm in here. I'm also not a paste deploy expert or proponent, just gathering information as to why its not being used in ironic when it is in many other projects | 19:24 |
| devananda | a quick read of the docs suggests it's a framework for loading wsgi apps | 19:25 |
| devananda | so one might use pastedeploy to load ironic's wsgi app - -but irnic doesn't need to use it, per se. is that right? | 19:26 |
| blogan | yeah but also being able to chain together middleware, afaik | 19:26 |
| blogan | oh ironic doesn't need to use it to have a wsgi app, i think paste deploy might add some more easier control for an operator if they need extra middleware | 19:27 |
| devananda | gotcha | 19:28 |
| devananda | so yea, we provide two methods of serving the ironic wsgi app today | 19:28 |
| blogan | im not saying one way is the right way though | 19:28 |
| devananda | including as an apache mod_wsgi app | 19:29 |
| blogan | yeah and I think we'll do a similar thing | 19:31 |
| blogan | i'm using ironic as a reference for a project I'm working on becasue we're using pecan and wsme | 19:31 |
| *** dprince has quit IRC | 19:34 | |
| *** yuanying has joined #openstack-ironic | 19:53 | |
| *** yuanying has quit IRC | 20:03 | |
| *** shikui_ has joined #openstack-ironic | 20:15 | |
| *** Guest43089 has joined #openstack-ironic | 20:20 | |
| *** Guest43089 is now known as annegentle_ | 20:20 | |
| openstackgerrit | David Shrewsbury proposed a change to openstack/ironic: Add "active" tracking to nodes and conductors https://review.openstack.org/124493 | 20:20 |
| *** mordred is now known as jeblair | 20:21 | |
| *** jeblair is now known as mordred | 20:21 | |
| Shrews | Does anyone know how to force git-review to download a change from gerritt (with the -d option) into a *differently* named branch? | 20:24 |
| Shrews | mordred: know that by any chance? ^^^ | 20:25 |
| rloo | Shrews: I couldn't figure that out either. | 20:25 |
| Shrews | i'm not sure it is possible looking at the git-review help | 20:26 |
| Shrews | which makes me a sad panda | 20:26 |
| JayF | Shrews: you can cherry-pick commits out of it | 20:27 |
| Shrews | JayF: i want to be able to 'git review' my changes back, though | 20:27 |
| Shrews | i guess i could use twisted, git magic to make that happen, but don't want to | 20:28 |
| Shrews | hrm, wonder if i can rename an existing branch | 20:28 |
| rloo | Shrews: you can rename the existing branch | 20:28 |
| Shrews | FIRST! | 20:28 |
| jroll | Shrews: you can git review -t topicname | 20:28 |
| jroll | instead of using the branch name for topic | 20:29 |
| rloo | Shrews: git branch -m | 20:29 |
| jroll | like... when you push your changes to gerrit, that is | 20:29 |
| rloo | Shrews may be first, but he was wondering, whereas I knew that it could be renamed ;) | 20:29 |
| openstackgerrit | David Shrewsbury proposed a change to openstack/ironic: Add periodic task to rebuild conductor local state https://review.openstack.org/124610 | 20:30 |
| *** jasondotstar has quit IRC | 20:30 | |
| Shrews | w00t | 20:31 |
| rloo | Shrews, I was just reviewing 124510. what did you do? :D | 20:31 |
| Shrews | rloo: just left a comment explaining | 20:31 |
| Shrews | rloo: basically, 124493 was incomplete b/c it needed code from 124610 to actually work | 20:32 |
| Shrews | is_active_on was not being persisted to the database | 20:32 |
| rloo | Shrews: yeah, that makes sense. Although I had a comment in that file. Will move it ;) | 20:32 |
| *** blogan has left #openstack-ironic | 20:32 | |
| Shrews | rloo: i left a comment on 124493 that all comments still need to be addressed | 20:33 |
| rloo | thx Shrews. Why didn't you address them too? :D | 20:33 |
| *** pensu has quit IRC | 20:33 | |
| Shrews | lucas-afk: took your advice and moved the node object code around ^^^^ | 20:33 |
| Shrews | rloo: i am lazy and incompetent | 20:34 |
| rloo | Shrews: not too incompetent. You addressed the comment I was going to add (about VERSION) ;) | 20:34 |
| Shrews | mostly incompetent, then :) | 20:35 |
| rloo | ha ha. Hey, Shrews, what's the difference between node.reservation and node.is_active_on. wrt being 'internal'. | 20:36 |
| Shrews | umm, not sure. i think internal attributes are ones we don't expose to the API | 20:37 |
| Shrews | that's what lucas' comment says at least | 20:38 |
| rloo | Shrews: right, but why don't we want to expose it, but we expose reservation? | 20:38 |
| Shrews | the value is not signifcant to a user. it's a db row id number | 20:39 |
| Shrews | perhaps if it were a uuid of the conductor.... | 20:40 |
| Shrews | as for reservation, i'm not sure what that one is, tbh | 20:40 |
| rloo | Shrews: oh oh. So why aren't we using uuid, wouldn't that be more useful. | 20:40 |
| Shrews | rloo: we don't want to use that for a foreign key. if we want to expose the uuid, we have to change our queries to do a join between nodes and conductors tables | 20:41 |
| rloo | Shrews: oh. ugh. db stuff. ok thx. | 20:42 |
| * Shrews throws out db jargon to scare rloo | 20:42 | |
| rloo | but it could be useful for debugging, to have the uuid. | 20:42 |
| rloo | Shrews, you need to throw harder; it didn't make it across the border. | 20:43 |
| Shrews | probably a future feature to expose that | 20:43 |
| rloo | yeah, need to leave stuff for the rest of us to do ;) | 20:43 |
| jroll | reservation is nice to have exposed | 20:49 |
| jroll | that's the conductor hostname that currently has a lock | 20:49 |
| *** linggao has quit IRC | 20:50 | |
| *** enterprisedc has quit IRC | 20:56 | |
| *** yuanying has joined #openstack-ironic | 20:58 | |
| rloo | jroll: I was just wondering whether it might be useful to expose is_active_on too. But that can be a future debate/change. | 20:59 |
| devananda | i thought briefly about using the hostname for is_active_on | 21:02 |
| *** marcoemorais has quit IRC | 21:03 | |
| devananda | Shrews: you want cherry-pick | 21:03 |
| *** marcoemorais has joined #openstack-ironic | 21:03 | |
| devananda | SHAS=$(for p in $PATCH_SERIES; do git review -d $p; echo $(git log | head -1 | awk '{print $2}'); done); git checkout master && (for p in $SHAS; do git cherry-pick $p; done) | 21:05 |
| devananda | something like that | 21:05 |
| devananda | reconstituting a patch tree where there are new branches is fun :) | 21:05 |
| * devananda uses words wrong | 21:05 | |
| devananda | reconstituting a patch series where some patches in the middle have new revisions is fun :) | 21:06 |
| jroll | devananda: more simply, git checkout master && for p in $PATCH_SERIES; do git review -x $p; done | 21:06 |
| jroll | :) | 21:07 |
| devananda | -x ? | 21:07 |
| jroll | cherry picks straight from gerrit | 21:07 |
| *** yuanying has quit IRC | 21:07 | |
| devananda | oooh that's new | 21:07 |
| devananda | thanks | 21:07 |
| jroll | I've been using it for months, not sure how new | 21:07 |
| devananda | huh | 21:09 |
| JayF | Our original automation for applying unmerged patches downstream used -x | 21:10 |
| JayF | so it has to have been working for at least 4 months or more, right jroll | 21:10 |
| jroll | "automation" is an interesting word | 21:10 |
| jroll | yes | 21:10 |
| *** hemna_ has joined #openstack-ironic | 21:10 | |
| JayF | it would've been fully automatic | 21:10 |
| JayF | if only jroll were a robot | 21:10 |
| jroll | actually, that one liner I posted *was* the automation | 21:10 |
| JayF | haha | 21:10 |
| jroll | just s/$PATCH_SERIES/$(cat ironic-patches) | 21:10 |
| *** hemna has quit IRC | 21:11 | |
| devananda | looks like about 8 months | 21:11 |
| devananda | https://github.com/openstack-infra/git-review/commit/e1e9d2ccfa65e610d8930b5d2f7f27cc7eebd491 | 21:12 |
| devananda | so yea, it's "new" :) | 21:12 |
| *** krtaylor has quit IRC | 21:13 | |
| Shrews | devananda: uhhh, your way is complex. i just renamed the current branch :-P | 21:29 |
| *** dhellmann is now known as dhellmann_ | 21:30 | |
| *** andreykurilin_ has quit IRC | 21:42 | |
| *** hemna_ has quit IRC | 21:50 | |
| *** hemna_ has joined #openstack-ironic | 22:00 | |
| *** yuanying has joined #openstack-ironic | 22:03 | |
| *** krtaylor has joined #openstack-ironic | 22:09 | |
| *** yuanying has quit IRC | 22:12 | |
| *** romcheg1 has joined #openstack-ironic | 22:24 | |
| *** ryanpetrello has quit IRC | 22:41 | |
| *** ryanpetrello has joined #openstack-ironic | 22:47 | |
| *** romcheg1 has quit IRC | 22:49 | |
| *** yuanying has joined #openstack-ironic | 22:58 | |
| *** achanda has joined #openstack-ironic | 23:13 | |
| *** lucas-afk has quit IRC | 23:25 | |
| *** igordcard has quit IRC | 23:26 | |
| *** hemna__ has joined #openstack-ironic | 23:35 | |
| *** hemna_ has quit IRC | 23:36 | |
| *** yuanying has quit IRC | 23:46 | |
| *** yuanying has joined #openstack-ironic | 23:46 | |
| *** yuanying has quit IRC | 23:51 | |
| *** chuckC_ has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!