Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-06-21
20:12:18 dansmith and not in a container in such a way that it can't see the interfaces nova would create?
20:13:12 SpamapS yes the agent is running
20:13:22 SpamapS and they share a netns (deployed with kolla-ansible)
20:13:31 SpamapS they actually don't even have a netns
20:13:34 SpamapS they're on the host's networking
20:15:59 dansmith from looking through the agent (for the first time ever) it surely looks like it's just polling for new interfaces to pop up
20:18:49 SpamapS dansmith: I can verify that brctl show from inside the agent shows the taps
20:20:36 openstackgerrit Brianna Poulos proposed openstack/python-novaclient master: Fix trusted-image-certificate-id help text https://review.openstack.org/576555
20:23:25 dansmith yeah, so the agent runs a loop,
20:23:46 dansmith each time through it looks for new interfaces, calls process_network_devices(), which goes through each new one, ends up at plug_interface(), etc, etc
20:24:16 dansmith so I think we create a properly-named tap, it notices, does the plug into the bridge and the wiring and then emits the notification
20:24:39 dansmith which explains why I can't find any calls to neutron in the critical path while we're waiting
20:26:05 SpamapS dansmith: makes sense I 'spose
20:27:00 dansmith looks like with debug on,
20:27:12 dansmith the agent should log "Agent loop found changes!" and then the device info it found
20:27:23 dansmith do you see that happen once nova has created the taps?
20:28:24 dansmith this: https://github.com/openstack/neutron/blob/master/neutron/plugins/ml2/drivers/agent/_common_agent.py#L453-L453
20:30:38 SpamapS dansmith: right, we never get Agent loop found changes
20:30:45 SpamapS stracing now to see why
20:33:49 dansmith got confirmation from someone who would know that "plug" communication between neutron and nova just involves creating the interface
20:36:30 SpamapS k
20:36:55 SpamapS well I can see the taps from the linuxbridge container
20:37:07 SpamapS but I never see the agent actually polling brctl or ip or anything
20:37:39 dansmith well, it may use iproute directly
20:37:49 dansmith which would strace out as a socket(AF_NETLINK,...) early on
20:37:59 SpamapS "or ip" was my vague reference to iproute
20:38:07 SpamapS but .. oh that?
20:38:29 dansmith er, pyiproute I mean
20:39:01 dansmith `ip` is a CLI that speaks over a special type of socket called a "netlink" socket, which is a communication channel to the network subsystem inside the kernel
20:39:05 dansmith it's how it does all its magic
20:39:07 SpamapS yeah makes sense
20:39:11 dansmith anything else can speak that as well
20:40:28 dansmith if it did, it could also get async notification of networking events, so it would make sense
20:40:59 SpamapS no looks like it execs 'ip'
20:41:06 SpamapS or at least, sometimes it does
20:41:08 SpamapS still reading
20:41:30 SpamapS and digs around in /sys
20:42:45 dansmith https://github.com/openstack/os-vif/blob/master/os_vif/internal/command/ip/impl_pyroute2.py
20:43:12 dansmith % grep import.*pyroute2 -r neutron/ | wc -l
20:43:12 dansmith 4
20:43:21 dansmith so, neutron uses it some, in master at least
20:43:50 dansmith in privileged/agent/linux/ip_lib.py
20:44:00 dansmith anyway
20:44:11 openstackgerrit Matt Riedemann proposed openstack/nova master: Address nits from consumer generation https://review.openstack.org/577227
20:44:16 SpamapS not in pike
20:44:38 SpamapS neutron doesn't use os_vif directly even, AFAICT
20:45:00 dansmith I know it doesn't use os-vif
20:45:09 dansmith because os-vif is the integration library between nova and neutron
20:45:10 dansmith so naturally,
20:45:14 dansmith it's only used by nova
20:47:47 SpamapS in all the ways I see the agent code checking for interfaces, I can see the interface
20:47:52 SpamapS but the agent never "notices"
20:49:58 openstackgerrit Chris Dent proposed openstack/nova master: WIP: [placement] Add /reshaper handler for POST https://review.openstack.org/576927
21:03:06 SpamapS dansmith: one last question. Do you know how the neutron agent gets the tap device name it should be looking for?
21:07:25 dansmith SpamapS: nova meeting right now
21:07:36 dansmith SpamapS: it's tap${uuid}
21:07:42 dansmith where uuid is the port uuid truncated
21:07:43 dansmith or something
21:07:58 dansmith you should probably talk to neutron people at this point, since it seems clearly on that side of the fence
21:13:16 dansmith if I was PTL all our meetings would be over in 12 minutes
21:13:17 dansmith just sayin'
21:13:46 tssurya dansmith: :P
21:15:37 openstackgerrit Takashi NATSUME proposed openstack/nova-specs master: Create specs directory for Stein https://review.openstack.org/573602
22:30:25 SpamapS dansmith: fyi, found the problem I think. All of the agents were stuck in the loop waiting for /run/ebtables.lock
22:30:36 SpamapS (even on restart, the lock persisted)
22:39:30 openstackgerrit Merged openstack/nova master: Add check if neutron "binding-extended" extension is available https://review.openstack.org/523548
22:44:10 openstackgerrit Zack Cornelius proposed openstack/nova master: Fix nits from change Ia7cf4414feb335b3c2e863b4c8b4ff559b275c34 https://review.openstack.org/577314
23:15:31 SpamapS now to figure out why my linuxbridge agents are using /run as their lock_path instead of /var/lib/neutron/tmp
23:58:45 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in test_compute_api.py (4) https://review.openstack.org/568462
23:59:25 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in virt/test_block_device.py https://review.openstack.org/566153
23:59:40 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in nova/tests/unit/virt/xenapi/stubs.py https://review.openstack.org/568412
#openstack-nova - 2018-06-22
01:54:14 openstackgerrit Tetsuro Nakamura proposed openstack/nova master: Add microversion for nested allocation candidate https://review.openstack.org/565487
01:54:15 openstackgerrit Tetsuro Nakamura proposed openstack/nova master: Fix nits in placement-return-all-resources series https://review.openstack.org/573604
02:02:52 openstackgerrit Merged openstack/nova master: Clarify placement DB schema migration https://review.openstack.org/576265
02:05:08 openstackgerrit Tetsuro Nakamura proposed openstack/nova master: Add microversion for nested allocation candidate https://review.openstack.org/565487
02:05:09 openstackgerrit Tetsuro Nakamura proposed openstack/nova master: Fix nits in placement-return-all-resources series https://review.openstack.org/573604
02:52:34 openstackgerrit Takashi NATSUME proposed openstack/nova master: api-ref: Fix parameters about trusted certificate IDs https://review.openstack.org/576046
02:52:52 openstackgerrit Takashi NATSUME proposed openstack/nova master: Use nova.test.TestingException https://review.openstack.org/575012
02:53:32 openstackgerrit Takashi NATSUME proposed openstack/nova master: Update admin/flavors document https://review.openstack.org/573063
02:54:03 openstackgerrit Takashi NATSUME proposed openstack/nova master: api-ref: Example verification for servers.inc https://review.openstack.org/529520
03:08:48 openstackgerrit Merged openstack/nova master: Remove network info stubbing in functional test https://review.openstack.org/570018
03:08:54 openstackgerrit Merged openstack/nova master: Cleanup nits in placement database changes https://review.openstack.org/577107
03:10:28 openstackgerrit Chen proposed openstack/python-novaclient master: Fix help text in server-group-create https://review.openstack.org/577341
03:37:34 CBR09 Hi everyone
03:37:38 CBR09 Is there anybody here
03:37:57 CBR09 I want to ask something about upgrading nova
03:39:37 openstackgerrit Ghanshyam Mann proposed openstack/nova master: Merge server create schema for BDM v2 extension https://review.openstack.org/577185
04:05:31 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in test_conductor.py (1) https://review.openstack.org/564658
04:05:48 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in test_conductor.py (2) https://review.openstack.org/564659
04:12:22 openstackgerrit jiang wei proposed openstack/nova master: Add action initiator attribute to the update payload https://review.openstack.org/536243
04:30:49 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in libvirt/test_driver.py (5) https://review.openstack.org/570842
04:31:09 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in libvirt/test_driver.py (6) https://review.openstack.org/571330
04:31:25 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in libvirt/test_driver.py (7) https://review.openstack.org/571992
04:31:43 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in libvirt/test_driver.py (8) https://review.openstack.org/571993
05:07:53 openstackgerrit Merged openstack/nova master: Fix nits from change I676291ec0faa1dea0bd5050ef8e3426d171de4c6 https://review.openstack.org/576985
05:08:00 openstackgerrit Merged openstack/nova master: Implement discard for file backed memory https://review.openstack.org/576994
06:14:30 openstackgerrit Lei Zhang proposed openstack/nova master: Add method to get cpu traits https://review.openstack.org/560317
06:23:30 openstackgerrit Dinesh Bhor proposed openstack/nova master: PCPU: Define numa dedicated CPU resource class https://review.openstack.org/561770
06:23:31 openstackgerrit Dinesh Bhor proposed openstack/nova master: NUMACell, InstanceNUMACell: Adopt 'PCPU' changes https://review.openstack.org/576021
06:23:43 openstackgerrit Merged openstack/nova master: objects: Remove NUMATopologyLimits.obj_from_db_obj https://review.openstack.org/537412

Earlier   Later