Earlier  
Posted Nick Remark
#openstack-nova - 2020-06-04
12:35:26 gmann it should be ok to remove as we already doing warning for 4 years but good to check in PTG if anyone still using that way.
12:36:11 sean-k-mooney gibi: i bought a https://www.amazon.co.uk/Samson-SAQ2U-Microphone-Recording-Pack/dp/B001R747SG dnymic mic and stand so i can turn on my ceiling fan without out it being picked up without having to use a headset
12:37:17 gibi gmann: Is it OK to for you if I move that topic up to the first slot today where we anyhow talk about removals and deprecation?
12:37:43 gibi sean-k-mooney: that mic look serious :)
12:37:55 gmann gibi: yeah, that is totally fine.
12:38:00 gibi gmann: cool
12:38:04 gmann thanks
12:40:56 sean-k-mooney gibi: well i work form home permently and have calls regualarly but ya its a live stage mic that alot of twitch/game stream also use. for like 60 euro its good value too
12:42:14 openstackgerrit Merged openstack/nova-specs master: Make Feature Liaison optional https://review.opendev.org/730638
12:46:41 gibi sean-k-mooney: yeah, sounds like a good idea for a home office. I only do full home office due to COVID
12:48:48 gmann johnthetubaguy: you want to do quick fix on release name in https://review.opendev.org/#/c/732173/1
12:50:23 johnthetubaguy gmann: oops, looking
12:51:03 openstackgerrit John Garbutt proposed openstack/nova-specs master: Re-propose Unified Limits in Nova https://review.opendev.org/732173
12:53:06 alex_xu sean-k-mooney: gibi so for neutron sriov agent, I guess it has no way to discover the VF? since it won't read from libvirt
12:59:35 sean-k-mooney alex_xu: it does but not in a useful way
12:59:51 sean-k-mooney you have to list the netdev names of the PF in its config
13:00:05 sean-k-mooney and it uses iptool i think
13:00:17 gibi alex_xu: nova sends pci_slot information to neutron in the binding profile but yeah as sean-k-mooney said the pci whitelist is on nova side today to specify what VF and PF resource is usable
13:00:41 sean-k-mooney gibi: we do also send the pci addres in the pci_slot key
13:00:43 gibi so nova has the VF resource view
13:01:06 sean-k-mooney but that is used only when we are plugging or unplugging the port
13:07:03 artom gdi, y'all are talking and I can't hear you, can I?
13:11:16 alex_xu_ gibi: sean-k-mooney I'm just thinking of that the neutron sriov agent won't be able to manage vf even in the future. (not today, today nova manage it)
13:12:03 openstackgerrit Merged openstack/nova-specs master: Re-propose Unified Limits in Nova https://review.opendev.org/732173
13:16:23 sean-k-mooney alex_xu_: right i dont think it will or should manage the vfs
13:25:23 artom Dammit, you need a Zoom account to change your audio output settings >_<
13:29:07 sean-k-mooney artom: you should not
13:29:15 sean-k-mooney i can change mine and i dont have one
13:29:22 artom sean-k-mooney, in the client?
13:29:33 sean-k-mooney in both the client and the browser
13:29:40 artom sean-k-mooney, 'cuz I didn't find any way to do it. Even their official doc states you have to be logged in
13:29:46 dansmith advanced settings need an app I think
13:29:47 artom sean-k-mooney, https://support.zoom.us/hc/en-us/articles/201362283-Testing-computer-or-device-audio
13:29:50 dansmith including advanced network
13:30:05 sean-k-mooney artom: what are you trying to change
13:30:15 artom sean-k-mooney, audio output to go to my BT headphones
13:30:26 sean-k-mooney if you click the arror on mic
13:30:31 artom It was outputting to stereo out, where nothing is currently plugged in
13:30:34 sean-k-mooney then you can change the output
13:30:49 artom sean-k-mooney, oh christ, yes, it's there, you're right
13:38:17 stephenfin artom: your gain is waaaay too high
13:39:20 sean-k-mooney artom: the zoom app auto adjust your volume
13:39:43 sean-k-mooney or can so maybe look at that it in the extra options via the same arror menue
13:41:20 artom stephenfin, but my gainz are way too low :( Stupid gyms closing
13:41:35 stephenfin artom: you know where the door is
13:50:50 sean-k-mooney artom: you are now a bit quite but that is an imporvement
13:54:06 gibi gmann: it is policy stuff in the nova room
13:54:11 bauzas gibi: patches from dansmith about the compute 5.0 API bump https://review.opendev.org/#/q/topic:queens-rpc-bump+(status:open+OR+status:merged)
13:55:00 dansmith bauzas: yeah
13:55:06 gmann gibi: 1 min joinin
13:55:08 bauzas gibi: the proxy dansmith told https://review.opendev.org/#/c/541005/6/nova/compute/manager.py@7750
13:55:11 gibi gmann: we will discuss the policy stuff after the break at 5 minute past the hour
13:55:27 gibi gmann: so no rush
13:55:32 bauzas anyway, less a deal to write the 6.0 proxy than the 5.0
13:55:35 gibi bauzas: thanks!
13:55:38 gmann gibi: ok
13:56:20 bauzas err, the 5.x proxy I mean
14:27:33 openstackgerrit Hervé Beraud proposed openstack/nova master: Always reset CONF when starting the wsgi app https://review.opendev.org/733627
14:27:33 openstackgerrit Hervé Beraud proposed openstack/nova master: Always reset CONF when starting the wsgi app https://review.opendev.org/733627
14:47:11 stephenfin kashyap: you're up next
14:49:17 kashyap stephenfin: Just a moment, wrapping up another thing, sorry
14:51:14 kashyap stephenfin: What's the topic? /me goes to check the Etherpad...
14:51:24 stephenfin q35
14:52:49 kashyap Line-435? We already talked a bit; see line-445 on the AGREED bit
14:52:59 kashyap stephenfin: ^
14:53:25 kashyap But sure, if there's anything else to be touched on that, can certainly speak up...
14:56:05 kashyap gibi: stephenfin: Has the topic flew by? Which line this whole ARM topic on?
14:56:33 stephenfin the previous one (line 427)
14:56:35 kashyap Ah, we're on line-427 currently
14:56:44 gibi kashyap: 427 yes
14:56:49 kashyap stephenfin: Yep, just see it, "virtualized architectures"
14:58:01 dansmith sean-k-mooney: so you could just float one DNM patch to some thing to add that queue to get one run of nova on arm yeah?
14:58:06 dansmith I mean, I assume we've never seen that work?
15:00:11 sean-k-mooney yep
15:00:30 dansmith yep which? can float a job or yep, we've seen it work?
15:01:13 sean-k-mooney you can just do a DNM patch
15:01:30 sean-k-mooney and yes i was looking at patches that do that previously
15:01:44 sean-k-mooney i think there are still a few issue they are working through
15:01:50 dansmith I didn't know we could do that so easily.. now I'm curious if it works as is
15:02:41 sean-k-mooney you just need to refecne the correct pipeline ill see if i can find the patch i was looking at before
15:14:32 sean-k-mooney its 8 on 3.x kernels
15:14:37 johnthetubaguy sean-k-mooney: ah...
15:14:41 sean-k-mooney becuase that si the most that vhost kernel module supports
15:14:50 johnthetubaguy sean-k-mooney: guest limit, or host limit, or both?
15:14:54 johnthetubaguy ah
15:14:56 sean-k-mooney in 4.X its 256
15:15:08 johnthetubaguy ah, interesting, thanks
15:15:22 sean-k-mooney we now have a config option so you can tell nova what it is if its different
15:15:32 kashyap gibi: I need to drop to run a quick errand, afraid
15:15:46 gibi kashyap: do you have other topics on the etherpad?
15:16:02 gibi I don't see any
15:16:02 kashyap gibi: Not that I know of
15:16:08 gibi OK. thanks for joining
15:16:53 kashyap I'm curious to hear stephenfin's topic of "documentation is awful" (line-474) - please take some notes on it on the Etherpad :)
15:17:39 gibi kashyap: ack
15:30:23 sean-k-mooney https://docs.openstack.org/nova/latest/configuration/config.html#filter_scheduler.max_io_ops_per_host
15:56:20 melwitt sean-k-mooney: just curious, how can you know there's not enough swap configured before any instances are booted so you know the flavor?
15:56:29 CeeMac hi sean-k-mooney and melwitt ; regarding or chat yesterday around shelving, we've done some testing today and it would appear that even in shelved_offloaded state an instance is accumulating usage and the resources still appear allocated in the project quota
15:57:09 CeeMac im back to being perplexed
15:57:24 melwitt CeeMac: oh... this is sounding familiar :(

Earlier   Later