Earlier  
Posted Nick Remark
#openstack-nova - 2021-02-19
14:34:55 sean-k-mooney bauzas: the name is not used in code
14:35:06 bauzas okay, so 'segment' then ?
14:35:10 sean-k-mooney the alisa is whats used internally
14:35:21 bauzas why is this so complicated to know which extension name to use ?
14:35:38 sean-k-mooney the name is the huma readable name more or less you would use the alis in the config files for example
14:35:51 sean-k-mooney where are you using it
14:35:56 bauzas gibi__: tbh, I copied the semantics from multi_provider_network extension
14:36:11 sean-k-mooney there are two filed you can check for either
14:36:15 bauzas this was a single name and with all non-capitals
14:36:36 bauzas sean-k-mooney: context https://review.opendev.org/c/openstack/nova/+/773976/9/nova/network/constants.py#32
14:36:45 sean-k-mooney thanks jsut went lookign for that
14:37:13 sean-k-mooney bauzas: stephenfin has a patch to normallyise you use of names to the alias
14:37:57 gibi__ I've tried, in devstack. I enabled the segment service_plugin in neutron and the pre-filter did not triggered as it looked for the name 'segments' but the extensions dict in nova contains 'Segments'
14:38:04 gibi__ bah
14:38:07 gibi__ segment and Segment
14:38:08 gibi__ singular
14:38:19 bauzas weirdo
14:38:25 gibi__ I change the constant to Segment locally and it now triggers the prefilter
14:38:28 bauzas I'll need to respin the patch either way
14:38:54 gibi__ but now I'm stuck troubleshooting neutron segment plugin as it does not create the placement RP inventory and aggregate
14:39:09 sean-k-mooney bauzas: https://review.opendev.org/c/openstack/nova/+/705792
14:39:20 gibi__ I will let you know where am I when I leave for the weekend
14:40:06 sean-k-mooney bauzas: currently we use the name so you would need Segment for your patch
14:40:35 bauzas ok, let me respin then
14:40:39 bauzas gibi__: thanks for checking
14:40:54 gibi__ no problemo
14:41:06 bauzas gibi__: https://docs.openstack.org/neutron/latest/admin/config-routed-networks.html
14:46:35 gibi__ bauzas: ack
14:51:58 openstackgerrit Sylvain Bauza proposed openstack/nova master: Add net & utils methods for routed nets & segments https://review.opendev.org/c/openstack/nova/+/773976
14:52:05 openstackgerrit Sylvain Bauza proposed openstack/nova master: Add a routed networks scheduler pre-filter https://review.opendev.org/c/openstack/nova/+/749068
14:56:19 gibi bauzas, sean-k-mooney: another finding http://paste.openstack.org/show/802828/ in this simple net config the segments plugin in neutron is not triggered as the subnet does not show any segment_id, but in the other hand the network has a segment, and the segment list shows the segment id
14:56:24 gibi this will be a problem for us
14:56:53 gibi as in this case nova things there is a segment (based on segments list) but the segmetns plugin in neturon did not created any aggregates as the subnet does not show a segment_id
14:57:35 bauzas wait
14:58:51 sean-k-mooney gibi: i think you need to be an admin to see it
14:59:03 bauzas gibi: I'll enter some bug triage meeting in a second (and I have to host it), could we postpone our discussion by one hour ?
14:59:13 gibi bauzas: sure, I will be her
14:59:17 gibi sean-k-mooney: I'm admin
14:59:38 sean-k-mooney gibi: can you try listing the filed explictly?
14:59:42 gibi sean-k-mooney: also verifyed that in this case the subnet has no segment_id from the neutron code perspective
14:59:48 gibi sean-k-mooney: sure
14:59:50 bauzas gibi: my wife will look at me as we have curfew at 6pm, so I won't be able to stay late as we need to visit supermarkets but we can continue for 10 mins
14:59:51 sean-k-mooney or doign it with debug
14:59:56 sean-k-mooney osc may not render it
15:00:11 sean-k-mooney by default that is
15:00:54 sean-k-mooney just reading your pastbin but need to jump to same meeting as bauzas brb
15:01:10 gibi sean-k-mooney: ack, have a nice meeting :)
15:01:11 gibi http://paste.openstack.org/show/802829/
15:01:26 gibi so the REST API returns segment_id null
15:01:59 stephenfin sean-k-mooney: I've about three patches that bauzas would benefit from, IMO, but I obviously don't want to make a review chain even longer /o\
15:02:09 stephenfin (the secure boot one is 12 patches long already)
15:03:38 sean-k-mooney stephenfin: ya i know just point it out to show we currently use the Name field
15:03:47 sean-k-mooney so Segment
15:04:06 sean-k-mooney not the alias field
15:04:06 stephenfin ah, yes, good point
15:04:17 stephenfin who ever thought that was a good idea...
15:04:21 stephenfin *whoever
15:04:39 sean-k-mooney whomever?
15:05:00 sean-k-mooney clearly the most important question of the day
15:06:28 stephenfin whoever
15:06:42 stephenfin I will give it to whomever wants it
15:07:04 stephenfin you can sub whomever with him, her, us etc.
15:07:11 sean-k-mooney imperitive vs questioning ?
15:07:15 stephenfin Whoever wants it can have it
15:07:33 stephenfin you can sub whoever for he, she, we, etc.
15:08:08 stephenfin so if e.g. "him" would make sense, choose whomever, else choose whoever
15:08:16 stephenfin at least I think that's how it works :)
15:08:59 stephenfin Yup, Google agrees with me https://www.grammarly.com/blog/2-effective-ways-to-deal-with-whoeverwhomever-confusion/
15:10:15 openstackgerrit Artom Lifshitz proposed openstack/nova master: pci: implement the `socket` NUMA affinity policy https://review.opendev.org/c/openstack/nova/+/772779
15:17:50 gibi sean-k-mooney, bauzas: I think I figured out another interesting thing. A subnet can be created without providing a segment id but also with --network-segment the user can provide a segment_id for a subnet. In the former case the segments plugin ignore the subnet and no placement and nova aggregates are created. But in the latter case the segment plugin acts on the subnet. However in both cases nova
15:17:56 gibi considers the network having a a segment as the segments list command returns the segment regardless of how the subnet was created.
15:29:34 openstackgerrit Artom Lifshitz proposed openstack/nova master: pci: implement the `socket` NUMA affinity policy https://review.opendev.org/c/openstack/nova/+/772779
15:58:45 bauzas gibi: I'm back
15:59:14 bauzas wrapping up on your comments
16:07:31 bauzas gibi: what you said last doesn't look an issue to me, right?
16:07:48 bauzas it's more the operator who messed up
16:09:40 gibi bauzas: no, both config is valid and both network is a valid network
16:10:06 gibi I think neutron considers net3 as a non routed network while net4 as a routed network
16:10:16 gibi but now consider both as routed and try to find the aggergate for net3
16:10:18 gibi and fails
16:10:52 gibi s/but now/but nova/
16:11:39 bauzas gibi: you refer to http://paste.openstack.org/show/802828/ ?
16:12:10 gibi sorry I refer to https://review.opendev.org/c/openstack/nova/+/773976/10/nova/scheduler/utils.py#1370
16:12:56 sean-k-mooney gibi: yes subnets dont need to have segment ids
16:13:11 sean-k-mooney gibi: but a network requires that all subnets either have them or none of them do
16:13:54 sean-k-mooney gibi: we shoudl have support for both routed and non routed netowrks in teh same cloud
16:14:16 gibi sean-k-mooney: ^^ yeah this is what my example in the comment tries out
16:14:56 sean-k-mooney no what i mean is i pointed this out already and bauzas added a func test to make sure this works if i am not rembereing
16:15:03 gibi so we can have one network with a segment and with subnet without that segment, and also another network with a segmetn and a subnet that has that segment
16:15:16 sean-k-mooney gibi: we will check every network but if a netwrok does not have segment it wont break anything
16:15:36 sean-k-mooney we are ment to catch InvalidRoutedNetworkConfiguration in the prefilter
16:15:37 gibi sean-k-mooney: yes but if the network has a segment but the subnet does not
16:15:58 bauzas gibi: and you're passing a port ?
16:16:00 sean-k-mooney segments are mapped to subnets
16:16:04 sean-k-mooney not to networks
16:16:14 gibi sean-k-mooney: but nova queries segment list
16:16:14 bauzas that ^
16:16:18 gibi for networks
16:16:34 bauzas gibi: for net3, it won't return segments, right?
16:16:43 sean-k-mooney no we get the segment on a network and look those up

Earlier   Later