| *** rpittau|afk is now known as rpittau | 07:02 | |
| opendevreview | ZhouHeng proposed openstack/neutron master: "default_availability_zones" need to be considered when validate az https://review.opendev.org/c/openstack/neutron/+/802665 | 07:36 |
|---|---|---|
| bcafarel | lajoskatona: ralonsoh: stable/ussuri started failing 100% on functional jobs (timeouts everywhere) https://bugs.launchpad.net/neutron/+bug/1938262 does this ring a bell? | 08:01 |
| bcafarel | it seems strange train is passing (aka something backported there but not in ussuri)? | 08:01 |
| ralonsoh | bcafarel, let me check | 08:01 |
| bcafarel | ralonsoh++ thanks | 08:02 |
| lajoskatona | bcafarel: thanks for headsup | 08:06 |
| opendevreview | Bodo Petermann proposed openstack/neutron-vpnaas master: VPNaaS support for OVN https://review.opendev.org/c/openstack/neutron-vpnaas/+/765353 | 09:20 |
| stephenfin | Can someone help me understand how the pecan view object...thing is built for the '/networks' API? I'm trying to figure out what APIs *do not* support pagination but getting exactly nowhere /o\ | 09:28 |
| ralonsoh | stephenfin, give me one min | 09:40 |
| ralonsoh | stephenfin, https://docs.openstack.org/api-ref/network/v2/#pagination | 09:47 |
| stephenfin | ralonsoh: Aye, I have that linked. The issue is that not all APIs support it | 09:47 |
| ralonsoh | you need the extension, that is optional (by default is enabled) | 09:48 |
| ralonsoh | yes but I doesn't matter: | 09:48 |
| ralonsoh | "If a particular plug-in does not support pagination operations, and pagination is enabled, the Networking API v2.0 will emulate the pagination behavior so that users can expect the same behavior regardless of the particular plug-in running in the background." | 09:48 |
| stephenfin | ahhhh | 09:50 |
| stephenfin | so I'll always have it | 09:50 |
| stephenfin | what about for resources that don't have an ID column? | 09:50 |
| ralonsoh | which ones? | 09:50 |
| stephenfin | sec | 09:50 |
| stephenfin | this is the change; looking to see what ones I skipped https://review.opendev.org/c/openstack/python-openstackclient/+/802482 | 09:51 |
| stephenfin | (FYI it's broken because openstacksdk doesn't currently allow you to request a specific page - it automatically fetches *all* pages) | 09:51 |
| ralonsoh | right | 09:52 |
| ralonsoh | stephenfin, but what specific resource are you talking about? | 09:53 |
| ralonsoh | without ID | 09:54 |
| stephenfin | sorry, I'm looking to find it again. I didn't right it down /o\ :) | 09:54 |
| stephenfin | *write | 09:54 |
| ralonsoh | ListIPAvailability | 09:56 |
| stephenfin | Yeah, that's one. Thanks | 09:56 |
| ralonsoh | stephenfin, and what's the problem there? this is not a resource itself but an DB view | 09:57 |
| ralonsoh | let me use the terminology | 09:57 |
| stephenfin | well all pagination in neutron is marker-based | 09:57 |
| stephenfin | so I need a marker | 09:58 |
| stephenfin | I assume that's an ID for most other resources, but if I don't have an ID what do I use | 09:58 |
| stephenfin | ? | 09:58 |
| stephenfin | Also, am I misunderstanding things or is pagination is not allowed for the '/rbac-policies' endpoint? https://github.com/openstack/neutron/blob/master/neutron/extensions/rbac.py#L129 | 09:58 |
| ralonsoh | stephenfin, about the rbac policies, it doesn't make sense. Those are DB registers | 09:59 |
| ralonsoh | we can change that | 09:59 |
| ralonsoh | about the ip availability... I don't know | 09:59 |
| ralonsoh | let me check the query done | 09:59 |
| stephenfin | the 'create_resource' call in neutron.api.v2.base also defaults to allow_pagination=False | 10:00 |
| stephenfin | so I was trying to figure out how that (or something else) was called to create e.g. the '/networks' API | 10:00 |
| stephenfin | so I could work backwards | 10:00 |
| ralonsoh | stephenfin, all resources are created using neutron.api.v2.base.create_resource | 10:02 |
| stephenfin | and the 'Controller._get_pagination_helper' method in the same module only enables the emulated pagination if 'allow_pagination' was set to True, otherwise you get the NoPaginationHelper | 10:02 |
| stephenfin | so does that mean the majority are created with no pagination support, except where they override this? | 10:03 |
| ralonsoh | stephenfin, looks like (I never saw that) | 10:03 |
| stephenfin | To further confuse things, I see 'build_resource_info' in 'neutron.api.v2.resource_helper' calls 'create_resource' with 'allow_pagination=True' but again, I'm not sure if that's used for the '/networks' API or not | 10:04 |
| stephenfin | ralonsoh: You can probably see why I'm so confused :-D | 10:04 |
| ralonsoh | I know | 10:04 |
| ralonsoh | stephenfin, for ip availability I think we need (1) change the controller and allow pagination and (2) change the primary key, if possible | 10:22 |
| ralonsoh | let me check that | 10:22 |
| ralonsoh | I'm playing with this right now | 10:22 |
| ralonsoh | stephenfin, ok, I know how to make it "paginable": change the API in neutron-lib, adding an API extension to make the network_id as primary_key | 10:32 |
| ralonsoh | and then modify the create_resource of Network_ip_availability, adding allow_pagination=True | 10:32 |
| ralonsoh | can you give me a bug ID? to document that | 10:33 |
| ralonsoh | and push the patches | 10:33 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN][Placement] Drive binding by placement allocation https://review.opendev.org/c/openstack/neutron/+/786478 | 11:05 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN][Placement] Add a SB Chassis event to track changes in BW config https://review.opendev.org/c/openstack/neutron/+/801712 | 11:25 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Remove "_get_network_lock_id" compatibility method https://review.opendev.org/c/openstack/neutron/+/800967 | 11:29 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Add "network_id" to "_after_router_interface_deleted" payload https://review.opendev.org/c/openstack/neutron/+/802687 | 11:29 |
| stephenfin | ralonsoh: will do | 11:51 |
| *** whoami-rajat is now known as Guest2569 | 12:41 | |
| *** whoami-rajat__ is now known as whoami-rajat | 12:41 | |
| lajoskatona | ralonsoh, obondarev: Hi, could You please check again this bgpvpn patch: https://review.opendev.org/c/openstack/networking-bgpvpn/+/799346 , not sure if we need a new ps | 12:57 |
| ralonsoh | sure | 12:57 |
| ralonsoh | ah this one | 12:57 |
| ralonsoh | lajoskatona, I'm really not sure about https://review.opendev.org/c/openstack/networking-bgpvpn/+/799346/3/networking_bgpvpn/neutron/db/bgpvpn_db.py#461 | 12:59 |
| ralonsoh | in any case, that won't affect the result | 12:59 |
| ralonsoh | lajoskatona, done | 13:00 |
| lajoskatona | ralonsoh: thanks | 13:07 |
| obondarev | lajoskatona: approved | 13:10 |
| opendevreview | Maxim Korezkij proposed openstack/neutron-vpnaas master: Added strongswan supported diffie-helllman-groups https://review.opendev.org/c/openstack/neutron-vpnaas/+/802714 | 13:27 |
| *** tbachman is now known as Guest2572 | 13:28 | |
| mjwales | Hi, we have a Victoria deployment where VMs are getting an I/O timeout when contacting the metadata server. The network the VMs are deployed to is a HA network but the metadata proxy is only running on the primary node. Is this correct or should the metadata proxy be running across all the HA nodes? | 13:31 |
| fungi | i know slaweq is on vacation, anybody happen to know if mlavalle is around this week? | 14:15 |
| ralonsoh | yes he is around, he'll chair the drivers meeting | 14:15 |
| fungi | thanks, hadn't seen him in irc since the weekend so thought maybe he was otherwise occupied | 14:16 |
| fungi | will try reaching out by e-mail instead | 14:16 |
| fungi | haleyb: amotoki: are you still active in https://launchpad.net/~neutron-coresec/+members ? just trying to figure out who to engage on some private security bugs | 14:18 |
| *** n98205 is now known as haleyb_ | 14:23 | |
| *** haleyb_ is now known as haleyb__ | 14:24 | |
| *** haleyb__ is now known as haleyb_ | 14:24 | |
| opendevreview | Merged openstack/neutron master: Sanitize MAC addresses https://review.opendev.org/c/openstack/neutron/+/789831 | 14:27 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN][Placement] Add support for minimum bandwidth QoS rules https://review.opendev.org/c/openstack/neutron/+/786441 | 14:54 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN][Placement] Drive binding by placement allocation https://review.opendev.org/c/openstack/neutron/+/786478 | 14:54 |
| *** rpittau is now known as rpittau|afk | 16:21 | |
| opendevreview | sean mooney proposed openstack/os-vif stable/wallaby: add configurable per port bridges https://review.opendev.org/c/openstack/os-vif/+/802475 | 16:48 |
| opendevreview | Terry Wilson proposed openstack/networking-ovn stable/train: Create neutron_pg_drop Port Group on init https://review.opendev.org/c/openstack/networking-ovn/+/802530 | 16:58 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!