| *** subscope has quit IRC | 00:27 | |
| *** jckasper has quit IRC | 00:36 | |
| *** jckasper has joined #openstack-neutron-ovn | 00:36 | |
| *** jckasper has quit IRC | 00:41 | |
| *** salv-orl_ has joined #openstack-neutron-ovn | 00:42 | |
| *** salv-orlando has quit IRC | 00:45 | |
| *** salv-orl_ has quit IRC | 01:00 | |
| *** jckasper has joined #openstack-neutron-ovn | 01:01 | |
| *** jckasper has quit IRC | 01:13 | |
| *** jckasper has joined #openstack-neutron-ovn | 01:14 | |
| *** jckasper has quit IRC | 01:18 | |
| *** jckasper has joined #openstack-neutron-ovn | 01:43 | |
| *** yamamoto has joined #openstack-neutron-ovn | 01:52 | |
| *** yamamoto has quit IRC | 02:14 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 02:38 | |
| *** salv-orlando has quit IRC | 02:41 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 02:55 | |
| *** yamamoto has joined #openstack-neutron-ovn | 02:58 | |
| *** salv-orlando has quit IRC | 03:01 | |
| *** jckasper has quit IRC | 03:26 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 03:57 | |
| *** allan_h has joined #openstack-neutron-ovn | 04:03 | |
| *** salv-orlando has quit IRC | 04:04 | |
| *** jckasper has joined #openstack-neutron-ovn | 04:15 | |
| *** brad_behle has quit IRC | 04:17 | |
| *** jckasper has quit IRC | 04:19 | |
| *** alhu_ has joined #openstack-neutron-ovn | 04:26 | |
| *** allan_h has quit IRC | 04:28 | |
| *** alhu__ has joined #openstack-neutron-ovn | 04:39 | |
| *** alhu_ has quit IRC | 04:39 | |
| *** gongysh has joined #openstack-neutron-ovn | 05:45 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 06:11 | |
| *** salv-orlando has quit IRC | 06:14 | |
| *** alhu__ has quit IRC | 06:19 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 06:54 | |
| *** salv-orlando has quit IRC | 07:02 | |
| openstackgerrit | Amitabha Biswas proposed openstack/networking-ovn: Delete ACLs of ports on provider network when port is deleted. https://review.openstack.org/282982 | 07:09 |
|---|---|---|
| openstackgerrit | Amitabha Biswas proposed openstack/networking-ovn: Reduce number of OVSDB ACL transactions https://review.openstack.org/282488 | 07:20 |
| *** salv-orlando has joined #openstack-neutron-ovn | 07:40 | |
| *** numans has joined #openstack-neutron-ovn | 07:50 | |
| *** subscope has joined #openstack-neutron-ovn | 07:51 | |
| *** subscope has quit IRC | 07:51 | |
| *** pcaruana has joined #openstack-neutron-ovn | 08:05 | |
| *** subscope has joined #openstack-neutron-ovn | 08:07 | |
| *** subscope has quit IRC | 08:17 | |
| *** fzdarsky has joined #openstack-neutron-ovn | 08:20 | |
| *** salv-orlando has quit IRC | 08:25 | |
| *** subscope has joined #openstack-neutron-ovn | 08:26 | |
| *** subscope has quit IRC | 08:40 | |
| openstackgerrit | Babu Shanmugam proposed openstack/networking-ovn: Documentation of observations on OVN integration with neutron-vpnaas https://review.openstack.org/283006 | 08:59 |
| *** subscope has joined #openstack-neutron-ovn | 09:00 | |
| *** gsagie has quit IRC | 09:22 | |
| *** gongysh has quit IRC | 09:23 | |
| *** subscope has quit IRC | 09:25 | |
| *** subscope has joined #openstack-neutron-ovn | 09:29 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 10:26 | |
| *** yamamoto has quit IRC | 10:28 | |
| *** salv-orlando has quit IRC | 10:30 | |
| *** subscope has quit IRC | 10:57 | |
| *** yamamoto has joined #openstack-neutron-ovn | 11:10 | |
| *** yamamoto has quit IRC | 11:14 | |
| *** subscope has joined #openstack-neutron-ovn | 11:30 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 11:33 | |
| *** gongysh has joined #openstack-neutron-ovn | 11:39 | |
| *** carl_baldwin has quit IRC | 11:40 | |
| *** salv-orlando has quit IRC | 11:43 | |
| *** carl_baldwin has joined #openstack-neutron-ovn | 11:44 | |
| *** subscope has quit IRC | 11:45 | |
| *** subscope has joined #openstack-neutron-ovn | 11:45 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 12:40 | |
| *** yamamoto has joined #openstack-neutron-ovn | 12:41 | |
| *** salv-orlando has quit IRC | 12:44 | |
| *** gsagie has joined #openstack-neutron-ovn | 12:44 | |
| *** rtheis has joined #openstack-neutron-ovn | 12:44 | |
| *** yamamoto has quit IRC | 13:10 | |
| *** gongysh has quit IRC | 13:16 | |
| *** yamamoto has joined #openstack-neutron-ovn | 13:16 | |
| *** numans has quit IRC | 13:23 | |
| *** yamamoto has quit IRC | 13:33 | |
| *** jckasper has joined #openstack-neutron-ovn | 13:43 | |
| *** brad_behle has joined #openstack-neutron-ovn | 13:44 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 13:45 | |
| *** salv-orlando has quit IRC | 13:52 | |
| *** fzdarsky has quit IRC | 13:55 | |
| *** numans has joined #openstack-neutron-ovn | 13:55 | |
| *** subscope has quit IRC | 13:56 | |
| *** fzdarsky has joined #openstack-neutron-ovn | 14:01 | |
| *** subscope has joined #openstack-neutron-ovn | 14:09 | |
| *** openstackgerrit has quit IRC | 14:17 | |
| *** openstackgerrit has joined #openstack-neutron-ovn | 14:17 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 14:41 | |
| Sam-I-Am | russellb: moooooo | 14:46 |
| *** jhfeng has joined #openstack-neutron-ovn | 15:28 | |
| *** azbiswas has joined #openstack-neutron-ovn | 15:40 | |
| *** azbiswas_ has joined #openstack-neutron-ovn | 15:42 | |
| *** azbiswas has quit IRC | 15:45 | |
| *** lrichard_ has joined #openstack-neutron-ovn | 15:51 | |
| russellb | quack | 15:57 |
| *** numans has quit IRC | 15:58 | |
| *** regXboi has joined #openstack-neutron-ovn | 16:00 | |
| *** armax has joined #openstack-neutron-ovn | 16:10 | |
| *** pcaruana has quit IRC | 16:27 | |
| *** jckasper_ has joined #openstack-neutron-ovn | 16:47 | |
| *** jckasper has quit IRC | 16:48 | |
| *** azbiswas_ has quit IRC | 16:59 | |
| *** alhu__ has joined #openstack-neutron-ovn | 17:33 | |
| *** alhu__ has quit IRC | 17:33 | |
| *** arosen12 has joined #openstack-neutron-ovn | 17:40 | |
| *** allan_h has joined #openstack-neutron-ovn | 17:40 | |
| *** subscope has quit IRC | 17:47 | |
| *** arosen12 has left #openstack-neutron-ovn | 17:50 | |
| *** arosen12 has joined #openstack-neutron-ovn | 17:50 | |
| *** thumpba has joined #openstack-neutron-ovn | 17:51 | |
| *** gangil has joined #openstack-neutron-ovn | 18:08 | |
| *** gangil has joined #openstack-neutron-ovn | 18:08 | |
| *** azbiswas has joined #openstack-neutron-ovn | 18:14 | |
| *** lsp42 has quit IRC | 18:16 | |
| *** lsp42 has joined #openstack-neutron-ovn | 18:19 | |
| *** gangil has quit IRC | 18:23 | |
| *** lsp42 has quit IRC | 18:31 | |
| *** lsp42 has joined #openstack-neutron-ovn | 18:36 | |
| *** salv-orl_ has joined #openstack-neutron-ovn | 18:42 | |
| rtheis | mestery: in https://bugs.launchpad.net/networking-ovn/+bug/1545897, what is the failure continue that you noticed in your tests? | 18:43 |
| openstack | Launchpad bug 1545897 in networking-ovn "Live migration between hosts with provider networks works sporadically" [High,Confirmed] - Assigned to Richard Theis (rtheis) | 18:43 |
| *** salv-orlando has quit IRC | 18:44 | |
| regXboi | iirc, mestery is out today | 19:04 |
| *** subscope has joined #openstack-neutron-ovn | 19:04 | |
| regXboi | russellb: ping | 19:04 |
| *** lsp42 has quit IRC | 19:06 | |
| russellb | regXboi: | .·´¯`°Q(•_• ) | 19:07 |
| regXboi | russellb: the good news is that it looks like the incremental lflow patch is now working as expected under scale | 19:07 |
| russellb | fantastic :) | 19:08 |
| regXboi | russellb: the bad news is that it looks like the balloon has moved to physical_run :( | 19:08 |
| russellb | that's fine, there will always be the next bottleneck | 19:08 |
| regXboi | I'm still doing some testing - I'll update the cover letter with my results later this week | 19:08 |
| russellb | sounds good! | 19:09 |
| *** s3wong has joined #openstack-neutron-ovn | 19:10 | |
| *** lsp42 has joined #openstack-neutron-ovn | 19:10 | |
| *** lsp42 has quit IRC | 19:17 | |
| *** chandrav has joined #openstack-neutron-ovn | 19:20 | |
| *** lsp42 has joined #openstack-neutron-ovn | 19:21 | |
| openstackgerrit | Russell Bryant proposed openstack/networking-ovn: Document active/passive HA without shared storage. https://review.openstack.org/280907 | 19:21 |
| *** gangil has joined #openstack-neutron-ovn | 19:25 | |
| *** gangil has joined #openstack-neutron-ovn | 19:25 | |
| *** lsp42 has quit IRC | 19:26 | |
| *** lsp42 has joined #openstack-neutron-ovn | 19:29 | |
| *** lsp42 has quit IRC | 19:34 | |
| *** yamamoto has joined #openstack-neutron-ovn | 19:48 | |
| *** lsp42 has joined #openstack-neutron-ovn | 19:50 | |
| *** lsp42 has quit IRC | 19:59 | |
| openstackgerrit | Merged openstack/networking-ovn: Delete ACLs of ports on provider network when port is deleted. https://review.openstack.org/282982 | 20:01 |
| *** lsp42 has joined #openstack-neutron-ovn | 20:02 | |
| *** lsp42 has quit IRC | 20:06 | |
| *** SpamapS has quit IRC | 20:20 | |
| *** leifmadsen has quit IRC | 20:20 | |
| *** mestery has quit IRC | 20:20 | |
| *** markmcclain has quit IRC | 20:20 | |
| *** toor has quit IRC | 20:20 | |
| *** jamespd has quit IRC | 20:20 | |
| *** toor has joined #openstack-neutron-ovn | 20:20 | |
| *** mestery has joined #openstack-neutron-ovn | 20:20 | |
| *** leifmadsen has joined #openstack-neutron-ovn | 20:21 | |
| *** markmcclain has joined #openstack-neutron-ovn | 20:21 | |
| *** leifmadsen has quit IRC | 20:21 | |
| *** leifmadsen has joined #openstack-neutron-ovn | 20:21 | |
| *** jamespd has joined #openstack-neutron-ovn | 20:22 | |
| *** SpamapS has joined #openstack-neutron-ovn | 20:22 | |
| *** openstackgerrit has quit IRC | 20:47 | |
| *** openstackgerrit has joined #openstack-neutron-ovn | 20:47 | |
| *** thumpba has quit IRC | 20:55 | |
| *** thumpba has joined #openstack-neutron-ovn | 20:56 | |
| *** thumpba has quit IRC | 21:00 | |
| regXboi | russellb: ping - got a SB port binding table question for you | 21:14 |
| russellb | sure | 21:20 |
| russellb | feel free to just leave question in the channel and i'll answer when i see it | 21:20 |
| regXboi | ack... | 21:20 |
| regXboi | I'm running a scale test of 2000 copies of this openstack pattern: n1 -- r1 -- n2 | 21:20 |
| regXboi | and I see the sbrec port binding table increase in size up to 1516 rows | 21:21 |
| regXboi | and then sit there for about 45 seconds when it suddenly jumps to 1712 rows | 21:22 |
| regXboi | and then just stays there regardless of the number of copies of the openstack pattern I stamp out | 21:22 |
| russellb | I imagine those are corresponding to DHCP agent ports? | 21:22 |
| russellb | the port binding rows are created by ovn-northd though | 21:22 |
| russellb | is ovn-northd pegged at 100% CPU? | 21:23 |
| * regXboi looks | 21:23 | |
| regXboi | I'd say it's pegged between 50 and 65% | 21:23 |
| russellb | well there's several things to check here | 21:24 |
| regXboi | ah ok... hold on... it's jumped again | 21:24 |
| russellb | first, i'd let the env chill out for a bit so it reaches a steady state | 21:24 |
| regXboi | well... I'm still scaling it :) | 21:24 |
| russellb | ok | 21:24 |
| russellb | # of port binding rows should == number of neutron ports | 21:25 |
| regXboi | ah | 21:25 |
| russellb | neutron ports are likely getting auto created here for DHCP agents | 21:25 |
| russellb | you might want to create your networks with DHCP off | 21:25 |
| regXboi | and the router interfaces | 21:25 |
| russellb | yes, router interfaces if the neutron L3 agent is enabled | 21:25 |
| russellb | otherwise no | 21:25 |
| regXboi | ah | 21:25 |
| regXboi | it's not | 21:25 |
| regXboi | ok, so this looks like (1) the next long pole in the tent | 21:26 |
| regXboi | (2) something I should look at northd about | 21:26 |
| regXboi | (3) something that likely should be a candidate for incremental processing as well | 21:26 |
| regXboi | do those three statements make sense? | 21:29 |
| russellb | likely, yes | 21:29 |
| regXboi | ok... then I can let this finish and cook overnight | 21:30 |
| russellb | there was a thread about a new design for ovn-northd using "nlog" | 21:30 |
| russellb | but maybe there is some lower hanging fruit in ovn-northd we could get in the meantime | 21:30 |
| regXboi | yeah, I think I remember reading it - need to go back and do a refresher on it | 21:30 |
| regXboi | well - travel safe to MN | 21:31 |
| russellb | thanks! i'm headed there tomorrow | 21:31 |
| russellb | you going? | 21:31 |
| regXboi | I'll finish this experiment, and let it sit overnight to see if it stabilizes out at 8000 ports | 21:31 |
| russellb | cool | 21:31 |
| regXboi | unfortunately no - we are dealing with some lingering health issues with my daughter and so I want to be around here in case things get interesting | 21:32 |
| russellb | would be interesting to know how long it's taking ovn-northd to recalculate as you reach higher #s of resources | 21:32 |
| regXboi | I can infer that from the logs of this run | 21:32 |
| russellb | i'm sorry to hear that, i hope things settle down there | 21:32 |
| regXboi | me to | 21:32 |
| russellb | family >>>>>>> neutron | 21:32 |
| regXboi | but my plan was to instrument ovn-northd tomorrow and repeat the experiment with some eyes on where it is spending its time to look for low hanging fruit | 21:33 |
| regXboi | yes family >>>>>>>>>>> neutron :) | 21:33 |
| russellb | sounds good | 21:33 |
| regXboi | I can say though that the logical flow patch is working as advertised - I'm running into the physical and ovn-northd points now | 21:33 |
| regXboi | which is goodness | 21:33 |
| azbiswas | regXboi: Just to confirm - these new ports are on different logical switches right? | 21:35 |
| regXboi | all I'm doing is stamping out copies of n1---r1---n2, so I believe the answer is yes | 21:36 |
| azbiswas | I want to remove ACL (my new fav topic) from the equation of processing :) | 21:36 |
| azbiswas | so that's good | 21:36 |
| russellb | yeah, should be 1 port per network in this env | 21:36 |
| russellb | as i understand it | 21:36 |
| regXboi | we'll know for sure when I'm done | 21:37 |
| russellb | regXboi: note that you're spawning thousands of dnsmasq processes here too heh | 21:37 |
| regXboi | on purpose :) | 21:38 |
| russellb | if on a small # of nodes, it could have a big impact, hurting how much you can stress the OVN bits | 21:39 |
| russellb | as long as you're aware! | 21:40 |
| regXboi | oh yes, I'm very aware of what I'm doing | 21:40 |
| openstackgerrit | Merged openstack/networking-ovn: Document active/passive HA without shared storage. https://review.openstack.org/280907 | 21:50 |
| *** thumpba has joined #openstack-neutron-ovn | 22:03 | |
| *** armax has quit IRC | 22:06 | |
| *** rtheis has quit IRC | 22:07 | |
| *** thumpba has quit IRC | 22:13 | |
| *** thumpba has joined #openstack-neutron-ovn | 22:13 | |
| *** regXboi has quit IRC | 22:19 | |
| *** thumpba has quit IRC | 22:23 | |
| *** thumpba has joined #openstack-neutron-ovn | 22:28 | |
| *** gangil1 has joined #openstack-neutron-ovn | 22:40 | |
| *** gangil has quit IRC | 22:41 | |
| *** switchcade has joined #openstack-neutron-ovn | 22:42 | |
| *** azbiswas has quit IRC | 22:49 | |
| *** azbiswas has joined #openstack-neutron-ovn | 22:49 | |
| *** allan_h has quit IRC | 22:50 | |
| *** allan_h has joined #openstack-neutron-ovn | 22:52 | |
| *** azbiswas has quit IRC | 22:54 | |
| *** jckasper_ has quit IRC | 23:03 | |
| *** jckasper has joined #openstack-neutron-ovn | 23:04 | |
| *** salv-orl_ has quit IRC | 23:08 | |
| *** jckasper has quit IRC | 23:08 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 23:11 | |
| *** allan_h has quit IRC | 23:14 | |
| *** allan_h has joined #openstack-neutron-ovn | 23:16 | |
| *** allan_h has quit IRC | 23:17 | |
| *** jckasper has joined #openstack-neutron-ovn | 23:26 | |
| *** jckasper has quit IRC | 23:31 | |
| *** armax has joined #openstack-neutron-ovn | 23:34 | |
| *** allan_h has joined #openstack-neutron-ovn | 23:40 | |
| *** subscope has quit IRC | 23:50 | |
| *** azbiswas has joined #openstack-neutron-ovn | 23:50 | |
| *** azbiswas has quit IRC | 23:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!