| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-06-02 | |||
| 11:06:16 | nightmare_unreal | how can one reboot the host then ? | |
| 11:06:32 | nightmare_unreal | OS commands | |
| 11:06:35 | lyarwood | nightmare_unreal: yes | |
| 11:06:53 | nightmare_unreal | okay :) Thanks | |
| 11:35:06 | openstackgerrit | Alexandre Arents proposed openstack/nova stable/stein: Reserve DISK_GB resource for the image cache https://review.opendev.org/732529 | |
| 11:53:37 | openstackgerrit | Alexandre Arents proposed openstack/nova stable/stein: Add admin doc information about image cache resource accounting https://review.opendev.org/732542 | |
| 11:58:20 | openstackgerrit | John Garbutt proposed openstack/nova master: Update quota apis with keystone limits and usage https://review.opendev.org/713499 | |
| 11:58:29 | openstackgerrit | John Garbutt proposed openstack/nova master: Add reno for unified limits https://review.opendev.org/715271 | |
| 13:06:01 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/train: libvirt: Don't delete disks on shared storage during evacuate https://review.opendev.org/731273 | |
| 13:06:01 | openstack | bug 1550919 in OpenStack Compute (nova) "[Libvirt]Evacuate fail may cause disk image be deleted" [Medium,Fix released] https://launchpad.net/bugs/1550919 - Assigned to Lee Yarwood (lyarwood) | |
| 13:06:01 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/train: Add functional test for bug 1550919 https://review.opendev.org/731272 | |
| 13:12:58 | johnthetubaguy | lyarwood, stephenfin: looking at this one: https://review.opendev.org/#/c/680648/4 I am thinking this works a little bit, it means the action log works properly :) | |
| 13:16:27 | lyarwood | johnthetubaguy: that's true | |
| 13:16:45 | johnthetubaguy | I am not quite sure if that is what they meant in the commit message though | |
| 13:16:55 | johnthetubaguy | made me wonder if we break that elsewhere actually | |
| 13:17:43 | johnthetubaguy | its really useful when that works, with the current horizon UI | |
| 13:20:26 | stephenfin | johnthetubaguy: You might be doing this already, but could I get you to review the rest of that series from jaypipes too? https://review.opendev.org/#/c/615618/ I'm happy to rework whatever you spot (I see one -1 so far) | |
| 13:20:41 | stephenfin | whoops, two | |
| 13:20:59 | johnthetubaguy | stephenfin: I kinda stopped as I see a few sets of things there, some I like, some I hate | |
| 13:21:25 | johnthetubaguy | basically I think we might need a spec for the quota classes stuff, as it impacts the API if we do it properly | |
| 13:21:51 | johnthetubaguy | I am all for removing quota class from the context though, seems like a good tidy up | |
| 13:22:10 | johnthetubaguy | and the unit tests using the string "default" for the quota class makes really good sense | |
| 13:23:20 | stephenfin | johnthetubaguy: Ah, okay, I'd thought it was pretty slam dunk but clearly not | |
| 13:24:08 | stephenfin | So maybe I'd be better off just abandoning the -1 ones rather than rework them | |
| 13:24:11 | johnthetubaguy | stephenfin: its a bit self centered, by patches need the code he drops, and get more complicated with his chnages | |
| 13:24:27 | johnthetubaguy | s/by/my/ | |
| 13:24:31 | johnthetubaguy | re unified limits | |
| 13:24:34 | stephenfin | yup :) I was hoping you'd review this for just that reason | |
| 13:24:51 | stephenfin | (I think I stuck you on the review, in fact, but maybe not) | |
| 13:24:53 | stephenfin | *s | |
| 13:24:58 | johnthetubaguy | its something worth tidy up though | |
| 13:25:17 | johnthetubaguy | that happens so much, I don't see those mails too often I am afraid | |
| 13:25:37 | johnthetubaguy | ... I should say the filter doesn't involve your name, its not a personal thing :) | |
| 13:26:24 | johnthetubaguy | stephenfin: I am torn on the abandon thing, because I agree that mess needs cleaning up (I think unified limits does that quite a bit, FWIW) | |
| 13:26:47 | stephenfin | heh :) | |
| 13:27:04 | stephenfin | hmm, perhaps you could incorporate them into your series, if you've bandwidth? | |
| 13:27:25 | johnthetubaguy | it could be a good follow on | |
| 13:27:54 | stephenfin | okay, so get unified limits in and _then_ look at this again? | |
| 13:28:33 | stephenfin | (for reference, the reason I even spotted these is that I'm trying to merge the long tail of really-old-but-likely-still-valid patches we have) | |
| 13:29:59 | openstackgerrit | Kevin Zhao proposed openstack/nova master: [WIP] CI: add tempest-integrated-compute-aarch64 job https://review.opendev.org/714439 | |
| 14:20:46 | josephillips | hi im having a issue with nova when i perform a resize nova try to perform a request to a old url of placement | |
| 14:21:01 | josephillips | even all the endpoint are correctly | |
| 14:21:42 | openstack | Launchpad bug 1881704 in OpenStack Compute (nova) "Resize don't work on Train when placement is in another Address" [Undecided,New] | |
| 14:21:42 | josephillips | https://bugs.launchpad.net/nova/+bug/1881704 | |
| 14:28:34 | openstackgerrit | Merged openstack/nova master: docs: Remove subtitles, metadata from man pages https://review.opendev.org/729916 | |
| 14:39:48 | rouk | hmph, did the block-device-mapping metadata api break in train? for some reason volumes dont show up, but the root disk does. | |
| 14:40:26 | sean-k-mooney | not that im aware of | |
| 14:41:02 | sean-k-mooney | rouk: are you refering to the device role tagging metadata or just volume metadata in general | |
| 14:41:05 | rouk | any debugging steps i can follow? i have a stein and a train cluster, i add a volume in stein, shows up, add a volume in train, no dice. | |
| 14:41:50 | rouk | sean-k-mooney: latest/meta-data/block-device-mapping/ doesnt get volumes, but does show root disk, basically. | |
| 14:42:24 | rouk | code looks simple in nova/api/metadata/base.py... not sure how it could break unless instance.mappings is changed in stein? | |
| 14:43:02 | openstackgerrit | Merged openstack/nova master: hacking: Modify checks for translated logs https://review.opendev.org/728967 | |
| 14:43:09 | openstackgerrit | Merged openstack/nova master: Add py38 package metadata https://review.opendev.org/724990 | |
| 14:43:48 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/stein: libvirt: Don't delete disks on shared storage during evacuate https://review.opendev.org/732696 | |
| 14:43:48 | openstack | bug 1550919 in OpenStack Compute (nova) "[Libvirt]Evacuate fail may cause disk image be deleted" [Medium,Fix released] https://launchpad.net/bugs/1550919 - Assigned to Lee Yarwood (lyarwood) | |
| 14:43:48 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/stein: Add functional test for bug 1550919 https://review.opendev.org/732695 | |
| 14:45:53 | rouk | sean-k-mooney: sorry, changed in train, the metadata api code looks unchanged... so itd have to be a change of mappings format? any ideas? | |
| 14:46:10 | sean-k-mooney | rouk: are you using trusted_vfs by any chance | |
| 14:46:29 | sean-k-mooney | actully never mind | |
| 14:46:36 | rouk | im going to say... no? never heard of it. | |
| 14:46:51 | sean-k-mooney | i was debating if the addtion of a continue would affect it but it will not | |
| 14:47:22 | sean-k-mooney | rouk: its an sriov related feature i was just looking at git blame | |
| 14:47:57 | rouk | ive traced as far as https://github.com/openstack/nova/blob/master/nova/api/metadata/base.py#L284 so far | |
| 14:48:07 | rouk | which looks unchanged | |
| 14:49:14 | sean-k-mooney | right so from a debuging point of view we are expecting it to get too https://github.com/openstack/nova/blob/d9cd0e1f1ca1175a538d25149deb804bde493079/nova/api/metadata/base.py#L417-L423 | |
| 14:49:44 | sean-k-mooney | i assume you are not seeing | |
| 14:49:46 | sean-k-mooney | LOG.debug('Metadata for device of unknown type %s has not ' | |
| 14:49:48 | sean-k-mooney | 'been included in the ' | |
| 14:49:50 | sean-k-mooney | 'output', device.__class__.__name__) | |
| 14:50:26 | rouk | i havnt flipped the debug level yet | |
| 14:50:54 | rouk | let me get the cluster nova-api flipped to debug | |
| 14:52:01 | sean-k-mooney | cool lets see if its hiting either the "Metadata for device of unknown" path or the "'Metadata for device with unknown bus" log messages | |
| 14:53:45 | rouk | hmm, just getting using cached metadata, any sure fire flush for it? | |
| 14:54:55 | rouk | in the metadata i only get ami and root entries, which point to sda, even though ive added sdb and sdc | |
| 15:03:19 | rouk | no debug messages about unknown types when i see a full query go though | |
| 15:05:41 | rouk | are you sure this device metadata appears in the api under block-device-mapping? | |
| 15:07:19 | rouk | get_ec2_metadata never seems to call that block | |
| 15:09:42 | rouk | "devices" in meta_data.json in both openstack versions in my tests are empty, this code seems unused for this? | |
| 15:11:15 | josephillips | question when i perform a resize who is resposable of | |
| 15:11:23 | josephillips | get the url of placement | |
| 15:11:32 | josephillips | nova scheduler or nova-compute? | |
| 15:11:33 | josephillips | ? | |
| 15:12:00 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/rocky: libvirt: Don't delete disks on shared storage during evacuate https://review.opendev.org/732708 | |
| 15:12:00 | openstack | bug 1550919 in OpenStack Compute (nova) "[Libvirt]Evacuate fail may cause disk image be deleted" [Medium,Fix released] https://launchpad.net/bugs/1550919 - Assigned to Lee Yarwood (lyarwood) | |
| 15:12:00 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/rocky: Add functional test for bug 1550919 https://review.opendev.org/732707 | |
| 15:27:00 | rouk | sean-k-mooney: the rows are in the db correctly... and the code looks like it should be showing them, unless its filtering somehow. | |
| 15:28:08 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/queens: libvirt: Don't delete disks on shared storage during evacuate https://review.opendev.org/732717 | |
| 15:28:08 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/queens: libvirt: Fix misleading debug msg "Instance is running" https://review.opendev.org/732716 | |
| 15:28:08 | openstack | bug 1550919 in OpenStack Compute (nova) "[Libvirt]Evacuate fail may cause disk image be deleted" [Medium,Fix released] https://launchpad.net/bugs/1550919 - Assigned to Lee Yarwood (lyarwood) | |
| 15:28:08 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/queens: Add functional test for bug 1550919 https://review.opendev.org/732715 | |
| 15:29:50 | openstack | bug 1869050 in OpenStack Compute (nova) queens "migration of anti-affinity server fails due to stale scheduler instance info" [Low,In progress] https://launchpad.net/bugs/1869050 - Assigned to Balazs Gibizer (balazs-gibizer) | |
| 15:29:50 | openstackgerrit | Balazs Gibizer proposed openstack/nova stable/queens: Reproduce bug 1869050 https://review.opendev.org/731563 | |
| 15:31:04 | openstackgerrit | Balazs Gibizer proposed openstack/nova stable/queens: Update scheduler instance info at confirm resize https://review.opendev.org/731564 | |
| 15:47:15 | bauzas | gibi: fwiw, I'm working on the routed-networks spec, hopefully be done by tomorrow so we could discuss it before the x-project PTG session with neutron | |
| 15:48:03 | gibi | bauzas: cool. thanks for the headsup | |
| 15:55:25 | openstackgerrit | Stephen Finucane proposed openstack/nova master: libvirt: Remove MIN_*_INTERFACE_MTU, MIN_*_TX_QUEUE_SIZE https://review.opendev.org/730866 | |
| 15:55:25 | openstackgerrit | Stephen Finucane proposed openstack/nova master: libvirt: Remove MIN_LIBVIRT_ETHERNET_SCRIPT_PATH_NONE https://review.opendev.org/730865 | |
| 15:55:26 | openstackgerrit | Stephen Finucane proposed openstack/nova master: libvirt: Remove unused host parameters https://review.opendev.org/732725 | |
| 16:02:10 | josephillips | any idea of how to solve this | |
| 16:02:11 | josephillips | has allocations against this compute host but is not found in the database. | |