| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-01-16 | |||
| 20:39:19 | melwitt | ok, should have read further before typing. once moved in-tree they had to be renamed to not collide with naming, now this is the next step https://docs.openstack.org/infra/manual/zuulv3.html#reworking-legacy-jobs-to-be-v3-native | |
| 20:39:58 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: Clarify fitting hugepages log message https://review.opendev.org/702981 | |
| 20:42:45 | melwitt | looks like there's some work in-flight for the grenade jobs https://review.opendev.org/#/q/topic:grenade_zuulv3+(status:open+OR+status:merged) | |
| 20:51:50 | amuller | hi folks, looking at nova.conf... [glance] -> api_servers was deprecated in https://github.com/openstack/nova/commit/cffbc2e431e90f7b6a67cf4a82349ee1d8de202a | |
| 20:51:55 | amuller | I was wondering what option replaced it? | |
| 20:52:14 | efried | amuller: Use standard ksa opts to get at glance | |
| 20:52:32 | amuller | efried: do you remember the section and conf key off the top of your head? | |
| 20:52:39 | efried | we don't support rotating IPs (poor man's load balancer hack) there anymore | |
| 20:53:20 | amuller | [keystone_authtoken] ? | |
| 20:53:30 | efried | amuller: If you simply remove api_servers, it should just pick up what's in your service catalog. If you really need to specify something different, you can use [glance]endpoint_override. | |
| 20:53:40 | efried | But I would suggest if you find yourself needing that ^ then your service catalog is busted :) | |
| 20:53:52 | efried | I guess some edge deployments might want to override, I think I heard that from someone. | |
| 20:55:33 | dansmith | from me | |
| 20:55:35 | amuller | efried++ | |
| 20:55:37 | amuller | thank you | |
| 20:55:57 | efried | no problem, good luck. | |
| 20:56:32 | amuller | I see that TripleO Stein still sets [glance] api_servers | |
| 20:56:38 | amuller | though the option was deprecated in Queens =p | |
| 20:56:59 | efried | if only someone from tripleo was paying attention | |
| 20:59:32 | efried | nova meeting now ish in #openstack-meeting | |
| 21:23:52 | openstackgerrit | melanie witt proposed openstack/nova master: DNM: try to get some debug info for bug 1844929 https://review.opendev.org/701478 | |
| 21:23:52 | openstack | bug 1844929 in OpenStack Compute (nova) "grenade jobs failing due to "Timed out waiting for response from cell" in scheduler" [High,Confirmed] https://launchpad.net/bugs/1844929 | |
| 21:50:21 | openstackgerrit | Dustin Cowles proposed openstack/nova master: Provider Config File: Function to further validate and retrieve configs https://review.opendev.org/676029 | |
| 21:50:21 | openstackgerrit | Dustin Cowles proposed openstack/nova master: Provider Config File: Functions to merge provider configs to provider tree https://review.opendev.org/676522 | |
| 21:50:22 | openstackgerrit | Dustin Cowles proposed openstack/nova master: Provider Config File: Enable loading and merging of provider configs https://review.opendev.org/693460 | |
| 21:58:49 | dustinc | efried: it is ready for re-review now :) | |
| 21:58:56 | efried | ack | |
| 22:00:41 | donnyd | sean-k-mooney: FN is back in operation | |
| 22:10:42 | efried | o/ donnyd! | |
| 22:49:31 | efried | dustinc: if you write one more meaty test case, I'll be +2 on that top patch. | |
| 22:50:01 | dustinc | efried: happy to, did you post the case in gerrit? | |
| 22:50:02 | efried | and I convinced gibi to put the series in his queue for tomorrow morning, so you could conceivably come back next week to a completed bp :) | |
| 22:50:07 | efried | dustinc: yeah | |
| 22:50:20 | dustinc | ok, maybe I will drop my other task to get that done.. | |
| 23:52:55 | openstackgerrit | Merged openstack/nova stable/rocky: Join migration_context and flavor in Migration.instance https://review.opendev.org/696572 | |
| #openstack-nova - 2020-01-17 | |||
| 00:30:17 | openstackgerrit | melanie witt proposed openstack/nova master: DNM: try to get some debug info for bug 1844929 https://review.opendev.org/701478 | |
| 00:30:17 | openstack | bug 1844929 in OpenStack Compute (nova) "grenade jobs failing due to "Timed out waiting for response from cell" in scheduler" [High,Confirmed] https://launchpad.net/bugs/1844929 | |
| 01:00:03 | brinzhang | stephenfin: thanks, that's true, we should update the updated_at field before aggregate or host save() | |
| 01:09:24 | openstackgerrit | Brin Zhang proposed openstack/nova master: Populate 'updated_at' when aggregate is updated https://review.opendev.org/580271 | |
| 01:10:36 | brinzhang | stephenfin, efried: done | |
| 01:13:37 | openstackgerrit | Brin Zhang proposed openstack/nova master: FUP Populate the 'updated_at' when aggregate metadata is updated https://review.opendev.org/702790 | |
| 03:55:12 | openstackgerrit | Dustin Cowles proposed openstack/nova master: Provider Config File: Functions to merge provider configs to provider tree https://review.opendev.org/676522 | |
| 03:55:12 | openstackgerrit | Dustin Cowles proposed openstack/nova master: Provider Config File: Enable loading and merging of provider configs https://review.opendev.org/693460 | |
| 04:02:59 | dustinc | efried, gibi: ^ series ready for re-review, thanks!! | |
| 05:51:26 | openstackgerrit | Brin Zhang proposed openstack/nova master: Follow-up: Add delete_on_termination to volume-attach API https://review.opendev.org/680655 | |
| 08:44:54 | kiseok7 | Hello all, I would like to add a code on upstream; an option in [cinder] section to allow to attach only volumes in the specified AZ. how I can process it? it need a blue print? | |
| 08:46:52 | lyarwood | kiseok7: doesn't [cinder]/cross_az_attach=False provide this? | |
| 08:47:09 | lyarwood | kiseok7: or did you want use a specific cinder AZ | |
| 08:47:20 | lyarwood | want to* | |
| 08:56:28 | kiseok7 | lyarwood: yes, it provides a part of. with [cinder]/cross_az_attach=False, nova creates instance with a volume in same AZ. but in my case, I need to attach another volume to the instance in different AZ. | |
| 09:02:19 | lyarwood | kiseok7: kk, I thought the AZ stuff in cinder was just left over from when it was nova-volume in our tree tbh, it might be worth writing this up on the mailing list for a wider discussion. | |
| 09:04:24 | kiseok7 | lyarwood: thank for help . I will try it. | |
| 09:21:15 | bauzas | good morning Nova | |
| 10:03:36 | brinzhang | bauzas: good morning | |
| 10:04:01 | bauzas | brinzhang: morning ! | |
| 10:04:02 | brinzhang | nova has the nova-live-migration zuul fail, can you try to check? | |
| 10:04:57 | brinzhang | bauzas: that impact so many patch :( | |
| 10:05:20 | bauzas | brinzhang: orly? | |
| 10:05:31 | bauzas | I can look at the gate | |
| 10:05:58 | brinzhang | Yeah, cool~ | |
| 11:33:44 | admin0 | hi all .. if all 3 controllers crashed ( no db anymore, wiped out) and all I have is the compute node now ( with the disk file) how do I boot it up ? i can't seem to find the libvirt xml anywhere | |
| 11:37:22 | yoctozepto | admin0: libvirtd alive? could ask it via virsh | |
| 11:37:50 | admin0 | virsh list --all == blank | |
| 11:37:51 | yoctozepto | oh, you probably mean that node died too | |
| 11:38:41 | admin0 | i have to showcase how openstack saves data even when the 3 controllers are dead .so instealled openstack,booted up some instances, nuked the 3 controllers and rebooted the compute node | |
| 11:38:51 | admin0 | so all i see now is the disk files in nova | |
| 11:39:03 | admin0 | but not the .xml definitions | |
| 11:39:08 | yoctozepto | well, the definitions are in the database... | |
| 11:39:21 | yoctozepto | if you want to show just that data is where it is | |
| 11:39:43 | yoctozepto | could boot any other vm and attach this image | |
| 11:39:52 | sean-k-mooney | admin0: well nova does not exactly save data on the compute nodes but in the live case there will be libvirt domain xmls | |
| 11:39:54 | admin0 | but for the test-case, all 3 controllers and db is nuked and unavailable | |
| 11:40:15 | yoctozepto | admin0: yeah, but you said you have the disk files | |
| 11:40:21 | admin0 | i do have them | |
| 11:40:24 | yoctozepto | could boot any vm in there | |
| 11:40:28 | yoctozepto | on that compute host | |
| 11:40:36 | yoctozepto | if you really want to show it is there for some reason | |
| 11:40:39 | sean-k-mooney | so when the contler are gone you can still use virsh to start the instaces | |
| 11:40:52 | admin0 | the only way i see is install virt-manager on my machne, add the compute node, and then create a new instance pointing to this disk | |
| 11:40:53 | sean-k-mooney | by the way you proably need to use sudo virsh list --all | |
| 11:41:09 | yoctozepto | ^ and this | |
| 11:41:12 | sean-k-mooney | if you dont use sudo you wont see the system slice | |
| 11:41:33 | yoctozepto | sean-k-mooney: admin0 mentioned downing the compute | |
| 11:41:36 | admin0 | i am logged in as root, did sudo virsh list --all .. see nothing | |
| 11:41:43 | yoctozepto | so it's likely it's reallly not there | |
| 11:41:46 | admin0 | yes .. the compute node was also rebooted | |
| 11:41:50 | yoctozepto | yeah | |
| 11:41:59 | yoctozepto | then either restore controllers | |
| 11:42:07 | sean-k-mooney | ok in that case have you confirmed the libvirtd systemd service is running | |
| 11:42:19 | admin0 | so test case is "oops something bad happened" . controllers are nuked, db is gone and compute nodes are rebooted | |
| 11:42:30 | sean-k-mooney | no the domains will be there | |
| 11:42:38 | sean-k-mooney | it wont be affect by the reboot | |
| 11:42:39 | admin0 | which folder | |
| 11:43:06 | sean-k-mooney | ill have to check but first you should make sure the libvirt is running | |
| 11:43:22 | sean-k-mooney | do sudo systemctl status libvirtd | |
| 11:43:47 | admin0 | its active and running | |
| 11:44:08 | sean-k-mooney | and you have no domain definitions? | |
| 11:44:24 | sean-k-mooney | did you put your instnace dir on nfs? | |
| 11:44:43 | yoctozepto | sean-k-mooney is right, I just verified instance files are really persistent | |
| 11:44:47 | yoctozepto | sorry, I got confused | |