| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-12-08 | |||
| 16:36:42 | sean-k-mooney | stephenfin: it is but alpine does not have cloudinit installed and i need to rework my disk image builder patches | |
| 16:37:10 | sean-k-mooney | we cant use fedora ro ubuntu as they are boot too large and too slow to use without nested virt | |
| 16:37:45 | sean-k-mooney | so long term im planing to try rebuilding my alpine images usign the iso image instead of the mini-fs image im currently using | |
| 16:37:56 | sean-k-mooney | to work around the initramfs issue im hitting | |
| 16:38:12 | sean-k-mooney | if i can get that to work then ill be propsosing using alpine instead of cirros | |
| 16:46:32 | sean-k-mooney | stephenfin: gibi slaweq i have a patch written just running test locally now | |
| 16:47:05 | gibi | sorry I'm bussy upstream | |
| 16:47:07 | gibi | will read back | |
| 16:49:34 | openstackgerrit | sean mooney proposed openstack/nova master: add libvirt_disable_apic workaround https://review.opendev.org/c/openstack/nova/+/766043 | |
| 16:49:35 | openstackgerrit | sean mooney proposed openstack/nova master: [DNM] testing libvirt_disable_apic workaround option https://review.opendev.org/c/openstack/nova/+/766044 | |
| 17:22:01 | gibi | sean-k-mooney, slaweq: thanks for looking into the cirros crash. | |
| 17:22:11 | gibi | sean-k-mooney: could os_command_line glance image propert helps? | |
| 17:27:14 | gibi | sean-k-mooney: hm, that seems to be only work with amazon type images :? | |
| 17:27:16 | gibi | :/ | |
| 17:40:39 | gibi | could we make that image property to work with other images | |
| 17:40:40 | gibi | ? | |
| 18:36:39 | openstackgerrit | Matteo Sposato proposed openstack/nova master: Functional tests remmoved direct post call https://review.opendev.org/c/openstack/nova/+/766068 | |
| 18:39:09 | sean-k-mooney | gibi: os_command_line only works if you use teh kernel seperatly | |
| 18:39:29 | sean-k-mooney | it also work for lxc | |
| 18:40:05 | sean-k-mooney | so to use it to set noapic we would have to use the split kernel and root disk image | |
| 18:41:01 | sean-k-mooney | i belive we could make it work with other image formats but we would still have to use seperate kernel image so that qemu can act as the bootloader and pass the arges instead of grub | |
| 18:45:33 | sean-k-mooney | gibi: that is using https://libvirt.org/formatdomain.html#direct-kernel-boot | |
| 19:00:33 | sean-k-mooney | this is where that is used https://opendev.org/openstack/nova/src/branch/master/nova/virt/libvirt/driver.py#L5285-L5297 | |
| 19:04:21 | sean-k-mooney | gibi: slaweq we could use http://download.cirros-cloud.net/0.5.1/cirros-0.5.1-x86_64-uec.tar.gz | |
| 19:04:48 | sean-k-mooney | that woudl give use a sperate inird and vmlinuz which would allow us to use os_command_line | |
| 19:25:09 | sean-k-mooney | gibi: slaweq https://review.opendev.org/c/openstack/devstack/+/766079 should work although i have not tested it | |
| 19:25:32 | sean-k-mooney | gibi: slaweq that uses os_command_line and the uae cirros images | |
| 19:25:39 | sean-k-mooney | to pass noapic | |
| 21:20:31 | openstackgerrit | Merged openstack/nova master: Enhance simple usage API documentation https://review.opendev.org/c/openstack/nova/+/765871 | |
| 22:22:21 | openstackgerrit | melanie witt proposed openstack/nova master: Disallow CONF.compute.max_disk_devices_to_attach = 0 https://review.opendev.org/c/openstack/nova/+/755360 | |
| #openstack-nova - 2020-12-09 | |||
| 01:38:11 | prometheanfire | any work on the updated mock support? https://review.opendev.org/765680 | |
| 02:30:34 | melwitt | prometheanfire: I resolved the remaining issues in the proposed patch but need someone to rebase it and handle the merge conflicts. I hoped stephenfin could help with that | |
| 02:30:48 | melwitt | I don't have as much of the context on the rest of the patch | |
| 02:35:35 | prometheanfire | melwitt: thanks for the update :D | |
| 02:37:22 | melwitt | np | |
| 03:42:15 | openstackgerrit | Brin Zhang proposed openstack/nova master: Replaces tenant_id with project_id from List/Update Servers APIs https://review.opendev.org/c/openstack/nova/+/764292 | |
| 03:42:16 | openstackgerrit | Brin Zhang proposed openstack/nova master: Replace all_tenants with all_projects in List Server APIs https://review.opendev.org/c/openstack/nova/+/765311 | |
| 03:42:16 | openstackgerrit | Brin Zhang proposed openstack/nova master: Replace tenants* with projects* of policies https://review.opendev.org/c/openstack/nova/+/765315 | |
| 07:25:06 | LarsErikP | melwitt: I noticed whis was merged a few hours ago \o/ can it be backported to ussuri? | |
| 08:06:00 | openstackgerrit | Mamduh proposed openstack/os-vif stable/ussuri: Refactor code of linux_net to more cleaner and increase performace https://review.opendev.org/c/openstack/os-vif/+/765419 | |
| 08:12:42 | gibi | LarsErikP: sure, it can be backported. But please propose a backport to stable/victoria first then stable/ussuri | |
| 08:15:46 | LarsErikP | melwitt: Uh.. Don't know what happended there. Forgot the link: https://review.opendev.org/c/openstack/nova/+/759348/ | |
| 08:16:51 | LarsErikP | gibi: Uhm.. not sure howto do that. Maybe melwitt could do it, as she fixed this originally? =) | |
| 08:18:11 | openstackgerrit | Mamduh proposed openstack/os-vif stable/ussuri: Fix - os-vif fails to get the correct UpLink Representor https://review.opendev.org/c/openstack/os-vif/+/765967 | |
| 08:20:30 | openstackgerrit | Lars Erik Pedersen proposed openstack/nova stable/victoria: Omit resource inventories from placement update if zero https://review.opendev.org/c/openstack/nova/+/766177 | |
| 08:20:53 | LarsErikP | gibi: I found the button ^ :P | |
| 08:23:13 | LarsErikP | should I add reviewers? and.. there is merge conflicts in ussuri. not feeling comfortable to deal with that :S | |
| 08:25:53 | openstackgerrit | Mamduh proposed openstack/os-vif stable/train: Refactor code of linux_net to more cleaner and increase performace https://review.opendev.org/c/openstack/os-vif/+/765912 | |
| 08:30:29 | gibi | LarsErikP:added some stable cores to the review. Thanks for proposing the backport | |
| 08:31:37 | gibi | LarsErikP: as for the ussuri backport, lets merge the victoria one first then I think melwitt can resolv the merge conflict in ussuri or I can take it if needed | |
| 08:32:13 | LarsErikP | gibi: good plan :-) thanks so much! | |
| 08:32:37 | gibi | LarsErikP: :) | |
| 08:44:37 | bauzas | good morning Nova | |
| 08:55:18 | gibi | bauzas: O/ | |
| 09:30:54 | openstackgerrit | Adrian Chiris proposed openstack/os-vif stable/ussuri: Fix - os-vif fails to get the correct UpLink Representor https://review.opendev.org/c/openstack/os-vif/+/765967 | |
| 09:46:15 | lyarwood | elod: morning, have you had anytime to look at the pip failures in stable before I start digging in? | |
| 09:47:59 | lyarwood | elod: https://review.opendev.org/q/Ia2007bc63ef09931ea0197cef29d6a5614ed821a - I was checking in on this series and noticed that everything prior to victoria is failing with pip 20.2.4 | |
| 09:48:02 | elod | lyarwood: good morning :) I'm looking at several issues now, which pip failures do you mean? :) | |
| 09:48:13 | lyarwood | elod: https://zuul.opendev.org/t/openstack/build/cb6247d4b3644045ab6d83a064e812c6 for example | |
| 09:48:33 | lyarwood | actually ussuri looks okay sorry | |
| 09:49:07 | lyarwood | ERROR: Package 'bandit' requires a different Python: 2.7.17 not in '>=3.5' | |
| 09:49:10 | elod | yes, this bandit is what I just started to look at | |
| 09:49:35 | elod | I guess we've got a new bandit version with incorrect setup.cfg :S | |
| 09:49:51 | elod | but haven't checked yet | |
| 09:51:04 | elod | at least there was a release, for sure: https://pypi.org/project/bandit/#history | |
| 09:51:14 | lyarwood | oh it's just a new version that drops py2 support | |
| 09:51:35 | lyarwood | okay we can cap this easily on stable where we are still using py2 | |
| 09:53:10 | elod | I'll look for the bandit changes and let's see if I can put this new version to disallow-list in global requirements | |
| 09:54:25 | lyarwood | https://github.com/PyCQA/bandit/releases/tag/1.6.3 it dropped py2 support | |
| 09:55:30 | lyarwood | elod: https://review.opendev.org/c/openstack/requirements/+/766170 | |
| 09:55:39 | lyarwood | elod: I *think* that's enough right? | |
| 09:56:52 | elod | thanks! setup.cfg looks ok ( https://github.com/PyCQA/bandit/blob/1.6.3/setup.cfg ) so I think this patch with the cap should be OK | |
| 09:57:16 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/train: libvirt: Skip encryption metadata lookups if secret already exists on host https://review.opendev.org/c/openstack/nova/+/765771 | |
| 09:57:26 | lyarwood | kk testing above | |
| 09:59:38 | elod | maybe we need to add that to global-requirements.txt | |
| 10:00:26 | lyarwood | hmm it's only a test-requirment in nova | |
| 10:04:53 | elod | lyarwood: it seems the patch won't work: https://opendev.org/openstack/requirements/src/branch/stable/train/blacklist.txt#L5 | |
| 10:05:32 | elod | so if I understand correctly we need to add it to every branch in every test-requirements.txt :/ | |
| 10:05:50 | elod | * the bandit cap | |
| 10:06:24 | lyarwood | elod: in nova right? | |
| 10:06:28 | lyarwood | fun | |
| 10:06:30 | elod | yes | |
| 10:06:33 | lyarwood | okay I'll do that now | |
| 10:07:39 | elod | (and "fortunately" I see lots of bandit failures all along other openstack repos :/ so it will be a nice amount of bandit patch if I'm not mistaken...) | |
| 10:08:26 | stephenfin | lyarwood: elod: We should probably move bandit and other linters out of test-requirements.txt and into tox.ini since they're not subject to upper-constraints | |
| 10:08:56 | stephenfin | https://github.com/openstack/python-openstackclient/blob/master/tox.ini#L31-L34 | |
| 10:09:04 | stephenfin | from https://github.com/openstack/python-openstackclient/commit/20769cd7b27d51da84a324a17922427eba5c6eac | |
| 10:09:34 | lyarwood | stephenfin: we can start doing that on master | |
| 10:09:44 | lyarwood | stephenfin: I'm not sure we want to change that on stable however right? | |
| 10:10:26 | stephenfin | I wouldn't see a reason not to, assuming your issue is derived from the new pip 20.3 resolver, rather than simply uncapped requirements | |
| 10:10:47 | stephenfin | If it's the latter, obviously just cap them and be done with it, sure | |
| 10:11:16 | lyarwood | it's the latter sorry, I assumed it was pip to begin with | |
| 10:11:22 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/train: Cap bandit at 1.6.2 when using py2 https://review.opendev.org/c/openstack/nova/+/766171 | |
| 10:11:28 | stephenfin | ah, then yes, cap all the way | |
| 10:11:29 | lyarwood | but it just wasn't capped | |
| 10:13:01 | lyarwood | okay lets try this again | |
| 10:13:06 | elod | hmmm, stephenfin I don't see the benefit of movint from test-req to tox.ini. we use capping in test-req, too | |
| 10:13:18 | stephenfin | elod: not for linters, you don't | |
| 10:13:31 | stephenfin | https://github.com/openstack/requirements/blob/master/blacklist.txt | |
| 10:13:38 | lyarwood | as we just found out :) | |