| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-06-20 | |||
| 19:01:15 | Nick_A | what are some potential reasons why metadata would have keys but not password (either in openstack/latest/password or openstack/latest/meta_data.json) when both are used to create the instance? | |
| 19:05:48 | efried | dansmith: Since request_filter.py is fresh in your head, how would you feel about moving the logic in the @trace decorator just into the main loop? | |
| 19:06:03 | efried | so we don't have to remember to decorate every method | |
| 19:06:59 | dansmith | efried: meh, seems like not much to remember and obvious if you're copying, but if you think it's important then okay | |
| 19:07:57 | efried | dansmith: When I was looking at sean-k-mooney1's patch and considering how best it could be done more simply, the fact that the trace was a decorator was slightly in the way of one of the better ways of organizing. Otherwise as you say not a big deal. | |
| 19:09:53 | dansmith | efried: I'm likely biased, but I totes don't see that much organization is needed at this point and prefer to wait for more things to be added so that clear patterns develop | |
| 19:09:54 | sean-k-mooney1 | efried: im not sure if dansmith has seen my patch yet | |
| 19:10:21 | efried | sean-k-mooney1: He did, swhat prompted me to bug him about it :) | |
| 19:10:36 | sean-k-mooney1 | ah i was doing bug assignment | |
| 19:10:44 | sean-k-mooney1 | let me check the patch | |
| 19:11:02 | dansmith | prepare to be overwhelmed with reading material | |
| 19:11:52 | sean-k-mooney1 | haha ok so you agree with erric :) | |
| 19:13:00 | sean-k-mooney1 | am are ye ok with me resturcting the code so ist more like the normal filter-schduler filers instead? | |
| 19:13:24 | dansmith | not really, no | |
| 19:13:46 | sean-k-mooney1 | ok... i can just revert to functions | |
| 19:14:06 | sean-k-mooney1 | and keep it all in one file but i think that is going to get annoying in the medium to long term | |
| 19:14:25 | dansmith | when it gets annoying we can split it up | |
| 19:15:00 | sean-k-mooney1 | well it was already annying me a littel :) but not enought to fight this battel to much | |
| 19:16:40 | sean-k-mooney1 | efried: im going to finish writing the 3 transform (as function not fucntors) and then ill start working on the dirver chagnes to report the tratis and move the filters to the end of the patch series | |
| 19:18:07 | sean-k-mooney1 | i might do the extraction fo the common helper code as a singel patch too and then build on that | |
| 19:18:25 | sean-k-mooney1 | that sound ok with people? | |
| 19:19:11 | dansmith | sean-k-mooney1: the three transforms need only be a single function..hopefully that's what you're planning right? | |
| 19:19:46 | sean-k-mooney1 | i was going to have one function for each to make the testing simplere but i can compess them | |
| 19:20:13 | dansmith | that is way (way) overkill, IMHO and just serves to get us to "annoying" sooner for no benefit :) | |
| 19:20:19 | sean-k-mooney1 | i would prefer to just add another fucntion to the dict everytime we add a new image metadata trait we support | |
| 19:20:38 | dansmith | but they're identical in behavior right? | |
| 19:21:03 | sean-k-mooney1 | i think i can generalise it actully | |
| 19:21:24 | sean-k-mooney1 | the main different is the field they match on and the prefix for the trait | |
| 19:21:36 | dansmith | I don't understand how it's not already generalized, but.. okay | |
| 19:21:38 | sean-k-mooney1 | that was on my todo list for the next versions anyway | |
| 19:23:58 | sean-k-mooney1 | which fucntio were you refering to by the way i plan to have only one filter fucntion transform_image_metadata but ideally i would also have one transfrom function that is paramaterised on teh traits prefix to use and the image metadata value to check | |
| 19:24:36 | dansmith | all you need is one function that inside it has a list of the properties to traitify | |
| 19:24:42 | sean-k-mooney1 | my StorageTrasform object is currently only parmaterised on the key to check but i wanted to refactor to also thake the prefix | |
| 19:25:11 | sean-k-mooney1 | ya ill make it simpelr in the next version | |
| 19:33:12 | openstackgerrit | Eric Fried proposed openstack/nova master: Inline trace_request_filter https://review.opendev.org/666663 | |
| 19:33:15 | efried | dansmith, sean-k-mooney1: ^ | |
| 19:33:35 | dansmith | damn, three away from double devil | |
| 19:33:57 | efried | I should go claim those next few numbers... | |
| 19:35:17 | sean-k-mooney1 | efried: i like that we will actully be using the return value for something outside of the tests | |
| 19:35:37 | efried | sean-k-mooney1: I didn't change anything from what was there before... | |
| 19:35:41 | sean-k-mooney1 | well i guess it was used in teh decorator before | |
| 19:35:45 | efried | yeah | |
| 19:36:19 | sean-k-mooney1 | im updating the docs string in my patch with the input paramaters so ill add a return parmaater to call that out too | |
| 19:36:27 | openstackgerrit | Eric Fried proposed openstack/nova master: DNM: Claim https://review.opendev.org/666666 | |
| 19:36:32 | efried | booya | |
| 19:36:39 | efried | now I just need to find a use for it | |
| 19:37:12 | sean-k-mooney1 | efried: with great power comes epic trolling :P | |
| 19:37:17 | efried | I can't wait | |
| 19:37:42 | mriedem | ooo | |
| 19:38:02 | mriedem | i can almost smell the goats blood | |
| 19:38:07 | efried | I'll be taking offers. Bitcoin and souls only. | |
| 19:39:49 | mriedem | i'd offer up taylor swift's soul so i don't have to hear that me-e-e-e-e song anymore | |
| 19:40:19 | openstackgerrit | Eric Fried proposed openstack/nova master: DNM: Claim https://review.opendev.org/666666 | |
| 19:43:48 | efried | kashyap: tbc, the spec in jroll's thread is for a *physical* tpm passthrough, so he doesn't have to wait for the "not-on-disk" feature of *virtual* tpm to become available. | |
| 19:46:31 | openstackgerrit | Matt Riedemann proposed openstack/python-novaclient master: WIP: Add host and hypervisor_hostname to create servers https://review.opendev.org/647671 | |
| 19:53:11 | openstackgerrit | Matt Riedemann proposed openstack/python-novaclient master: WIP: Add host and hypervisor_hostname to create servers https://review.opendev.org/647671 | |
| 19:57:45 | lbragstad | could add some black sabbath links to your docs | |
| 20:04:16 | Nick_A | why would admin password be stored in config drive meta_data.json but not shown in either meta_data.json or password when curling metadata from the guest? | |
| 20:39:45 | openstackgerrit | Adam Spiers proposed openstack/nova master: Use fake flavor instead of empty dict in test https://review.opendev.org/662555 | |
| 20:39:45 | openstackgerrit | Adam Spiers proposed openstack/nova master: Add extra spec parameter and image property for memory encryption https://review.opendev.org/664420 | |
| 20:39:46 | openstackgerrit | Adam Spiers proposed openstack/nova master: Extract SEV-specific bits on host detection https://review.opendev.org/636334 | |
| 20:39:46 | openstackgerrit | Adam Spiers proposed openstack/nova master: Pass extra_specs to flavor in vif tests https://review.opendev.org/662556 | |
| 20:39:47 | openstackgerrit | Adam Spiers proposed openstack/nova master: Apply SEV-specific guest config when SEV is required https://review.opendev.org/644565 | |
| 20:39:47 | openstackgerrit | Adam Spiers proposed openstack/nova master: Add |
|
| 20:39:48 | openstackgerrit | Adam Spiers proposed openstack/nova master: Enable booting of libvirt guests with AMD SEV memory encryption https://review.opendev.org/666616 | |
| 20:51:26 | aspiers | efried: first pass at series restructuring done | |
| 20:51:41 | aspiers | I'm sure you'll find things you still don't like :) | |
| 20:51:43 | efried | aspiers: cool. Sorry to be a PITA about this | |
| 20:51:48 | aspiers | np | |
| 20:52:07 | efried | it's the 'master switch' that's the most important thing. | |
| 20:52:36 | aspiers | I may not entirely disagree with your commit boundary preferences, but I agree 100% with the master switch thing, and 1000% with a pedantic approach which achieves high quality results ;-) | |
| 20:52:44 | aspiers | s/disagree/agree/ | |
| 20:53:06 | aspiers | OK, that's enough (actually too much) work for today | |
| 20:53:14 | aspiers | 10pm and still haven't eaten | |
| 20:53:30 | efried | there must be a magic triangle of pedantry/quality/productivity | |
| 20:53:32 | aspiers | haha | |
| 20:53:44 | aspiers | screw productivity, it's all about pedantry/quality ;-) | |
| 20:54:02 | aspiers | BTW tomorrow may be my last day working on SEV for ~3 weeks since next week is SUSE hackweek, and then I'm off to the USA for 2 weeks holiday | |
| 20:54:13 | efried | ight. | |
| 20:54:24 | aspiers | I think Boris is back around now, so *maybe* he can help progress things while I'm gone | |
| 20:54:34 | efried | aspiers: In your absence, if small things come up, permission to address them without you? | |
| 20:54:39 | aspiers | sure! | |
| 20:54:59 | efried | so like you might (*might*) come back to find this merged | |
| 20:55:03 | aspiers | haha | |
| 20:55:15 | aspiers | well that would be a lovely welcome back, but I'm not expecting it | |
| 20:55:18 | efried | yes, laughter appropriate | |
| 20:55:43 | aspiers | BTW it's looking like I'm being more officially allocated to upstream work from now on | |
| 20:55:55 | aspiers | so this is very good news (for me at least, maybe not for you) | |
| 20:56:42 | efried | nice | |
| 20:56:47 | efried | it is good news for me | |
| 20:56:51 | aspiers | :) | |
| 20:56:55 | efried | been getting kinda breezy around here | |
| 20:57:00 | aspiers | more excuses to be a pedant ;-p | |
| 20:58:00 | aspiers | would be great if I got time to hack on that requested_resources stuff, for example | |
| 20:58:16 | aspiers | at the moment I'm slightly feeling the heat on the SEV stuff, as we have some internal deadlines | |
| 20:58:47 | aspiers | not for forcing stuff down upstream's throat, but for reaching something that is functionally near complete downstream | |
| 21:00:00 | aspiers | OK, dinner time. Hopefully zuul CI won't shit all over my latest patchsets :) | |
| 21:00:04 | aspiers | Cya tomorrow o/ | |
| 21:00:47 | mriedem | having a meeting today? | |
| 21:01:00 | efried | oh, crap, I am completely unprepared. | |