| *** mlavalle has quit IRC | 00:03 | |
| *** suresh12 has quit IRC | 00:07 | |
| *** suresh12 has joined #openstack-neutron | 00:12 | |
| jog0 | markmcclain: ping | 00:17 |
|---|---|---|
| jog0 | how many lines do you want for your quantum rename | 00:17 |
| jog0 | stackalytics wise | 00:17 |
| *** amotoki has joined #openstack-neutron | 00:21 | |
| *** marun has joined #openstack-neutron | 00:31 | |
| jog0 | https://review.openstack.org/#/c/62504/ | 00:33 |
| *** yamahata has joined #openstack-neutron | 00:33 | |
| *** dims has quit IRC | 00:37 | |
| *** ashaikh has quit IRC | 00:37 | |
| *** dims has joined #openstack-neutron | 00:39 | |
| *** dims has quit IRC | 00:47 | |
| *** Mr_W has joined #openstack-neutron | 00:56 | |
| *** manjiri has joined #openstack-neutron | 00:59 | |
| *** manjiri has left #openstack-neutron | 01:00 | |
| *** dims has joined #openstack-neutron | 01:02 | |
| *** banix has joined #openstack-neutron | 01:04 | |
| *** dcahill has quit IRC | 01:10 | |
| openstackgerrit | Salvatore Orlando proposed a change to openstack/neutron: Process port_update notifications in the main agent loop https://review.openstack.org/61964 | 01:11 |
| *** suresh12 has quit IRC | 01:15 | |
| *** suresh12 has joined #openstack-neutron | 01:17 | |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: ml2: gre, vxlan type driver can leak segment_id https://review.openstack.org/61694 | 01:17 |
| *** banix has quit IRC | 01:23 | |
| *** Jianyong has joined #openstack-neutron | 01:24 | |
| *** tnadeau has quit IRC | 01:32 | |
| openstackgerrit | A change was merged to openstack/neutron: Imported Translations from Transifex https://review.openstack.org/62301 | 01:34 |
| *** djbkd has quit IRC | 01:39 | |
| *** ljjjustin has joined #openstack-neutron | 01:41 | |
| *** gdubreui has quit IRC | 01:45 | |
| *** ljjjustin has quit IRC | 01:46 | |
| openstackgerrit | Nachi Ueno proposed a change to openstack/neutron: Improve vif attributes related with firewalling https://review.openstack.org/21946 | 01:46 |
| *** ljjjustin has joined #openstack-neutron | 01:48 | |
| *** nati_ueno has quit IRC | 02:01 | |
| *** dave_tucker has quit IRC | 02:11 | |
| *** Mr_W has left #openstack-neutron | 02:12 | |
| *** banix has joined #openstack-neutron | 02:12 | |
| *** dzyu has joined #openstack-neutron | 02:17 | |
| *** networkstatic has quit IRC | 02:24 | |
| *** banix has quit IRC | 02:24 | |
| *** scotty916 has joined #openstack-neutron | 02:28 | |
| *** nati_ueno has joined #openstack-neutron | 02:30 | |
| *** scotty916 has quit IRC | 02:34 | |
| *** jp_at_hp1 has quit IRC | 02:40 | |
| openstackgerrit | berlin proposed a change to openstack/neutron: Fix VPNaaS Service driver should check whether router has gateway https://review.openstack.org/60461 | 02:46 |
| *** gdubreui has joined #openstack-neutron | 02:46 | |
| openstackgerrit | berlin proposed a change to openstack/neutron: Fix VPNaaS Service driver should check whether router has gateway https://review.openstack.org/60461 | 02:47 |
| *** neagle has joined #openstack-neutron | 02:47 | |
| *** aymenfrikha has joined #openstack-neutron | 02:48 | |
| *** vkozhukalov has joined #openstack-neutron | 02:48 | |
| *** aymenfrikha has left #openstack-neutron | 02:48 | |
| *** SushilKM has joined #openstack-neutron | 02:48 | |
| *** neagle has quit IRC | 02:50 | |
| *** banix has joined #openstack-neutron | 02:50 | |
| *** SushilKM has quit IRC | 02:52 | |
| *** dims has quit IRC | 02:53 | |
| *** dims has joined #openstack-neutron | 02:53 | |
| *** SushilKM has joined #openstack-neutron | 03:03 | |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: ml2/type_gre: Adds missing clear_db to test_type_gre.py https://review.openstack.org/62530 | 03:06 |
| *** yamahata__ has quit IRC | 03:06 | |
| *** yamahata__ has joined #openstack-neutron | 03:06 | |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: ml2/type_gre: Adds missing clear_db to test_type_gre.py https://review.openstack.org/62530 | 03:11 |
| *** ljjjustin has quit IRC | 03:21 | |
| *** clev has joined #openstack-neutron | 03:36 | |
| *** otherwiseguy has quit IRC | 03:40 | |
| *** clev has quit IRC | 03:41 | |
| openstackgerrit | Dazhao Yu proposed a change to openstack/neutron: Calculate stateless IPv6 address https://review.openstack.org/56184 | 03:47 |
| *** aveiga has quit IRC | 03:53 | |
| *** jecarey has joined #openstack-neutron | 03:54 | |
| openstackgerrit | A change was merged to openstack/neutron: NVP plugin: Avoid timeouts if creating routers in parallel https://review.openstack.org/60861 | 03:58 |
| openstackgerrit | A change was merged to openstack/neutron: Include lswitch id in NSX plugin port mappings https://review.openstack.org/54578 | 03:58 |
| openstackgerrit | A change was merged to openstack/neutron: NVP: propagate net-gw update to backend https://review.openstack.org/62322 | 03:58 |
| *** suresh12 has quit IRC | 04:08 | |
| *** chandankumar has joined #openstack-neutron | 04:16 | |
| *** banix has quit IRC | 04:25 | |
| *** iwamoto has joined #openstack-neutron | 04:27 | |
| *** banix has joined #openstack-neutron | 04:29 | |
| openstackgerrit | Micheal Thompson proposed a change to openstack/neutron: Added A10 LBaaS Driver to branch. https://review.openstack.org/62464 | 04:38 |
| *** SushilKM has quit IRC | 04:38 | |
| *** ashaikh has joined #openstack-neutron | 04:45 | |
| *** csd_ has quit IRC | 04:51 | |
| *** csd_ has joined #openstack-neutron | 04:52 | |
| *** jecarey has quit IRC | 04:59 | |
| openstackgerrit | berlin proposed a change to openstack/neutron: Support advanced NVP IPsec VPN Service https://review.openstack.org/55207 | 05:03 |
| *** qs201 has joined #openstack-neutron | 05:05 | |
| *** qs201 has quit IRC | 05:07 | |
| *** networkstatic has joined #openstack-neutron | 05:10 | |
| *** djbkd has joined #openstack-neutron | 05:12 | |
| *** yfried has joined #openstack-neutron | 05:14 | |
| openstackgerrit | berlin proposed a change to openstack/neutron: Fix VPNaaS Service driver should check whether router has gateway https://review.openstack.org/60461 | 05:15 |
| *** banix has quit IRC | 05:25 | |
| *** csd_ has quit IRC | 05:31 | |
| *** csd_ has joined #openstack-neutron | 05:32 | |
| *** gdubreui has quit IRC | 05:32 | |
| openstackgerrit | Micheal Thompson proposed a change to openstack/neutron: Wrapped text https://review.openstack.org/62544 | 05:35 |
| *** garyk has quit IRC | 05:35 | |
| *** csd_ has quit IRC | 05:40 | |
| *** csd_ has joined #openstack-neutron | 05:41 | |
| *** krast has joined #openstack-neutron | 05:47 | |
| openstackgerrit | Berezovsky Irena proposed a change to openstack/neutron: Add update from agent to plugin on device up https://review.openstack.org/53609 | 05:47 |
| *** krast has quit IRC | 05:48 | |
| *** krast has joined #openstack-neutron | 05:48 | |
| *** krast has quit IRC | 05:49 | |
| *** krast has joined #openstack-neutron | 05:49 | |
| krast | hi guys~ has some open source like vmware nsx? | 05:50 |
| *** SushilKM has joined #openstack-neutron | 06:00 | |
| *** SushilKM__ has joined #openstack-neutron | 06:04 | |
| *** dzyu has quit IRC | 06:04 | |
| *** SushilKM has quit IRC | 06:04 | |
| *** bashok has joined #openstack-neutron | 06:05 | |
| *** dzyu has joined #openstack-neutron | 06:06 | |
| openstackgerrit | Maru Newby proposed a change to openstack/neutron: Send DHCP notifications regardless of agent status https://review.openstack.org/61168 | 06:07 |
| *** SushilKM__ has quit IRC | 06:14 | |
| *** SushilKM has joined #openstack-neutron | 06:14 | |
| openstackgerrit | berlin proposed a change to openstack/neutron: Support advanced NVP IPsec VPN Service https://review.openstack.org/55207 | 06:19 |
| *** inara` has quit IRC | 06:30 | |
| *** alex_klimov has joined #openstack-neutron | 06:34 | |
| *** inara has joined #openstack-neutron | 06:37 | |
| openstackgerrit | Jenkins proposed a change to openstack/neutron: Imported Translations from Transifex https://review.openstack.org/62561 | 06:38 |
| *** alex_klimov has quit IRC | 06:39 | |
| *** SushilKM has quit IRC | 06:40 | |
| *** amritanshu_RnD has joined #openstack-neutron | 06:41 | |
| *** alex_klimov has joined #openstack-neutron | 06:41 | |
| *** harlowja is now known as harlowja_away | 06:43 | |
| *** SushilKM has joined #openstack-neutron | 06:43 | |
| *** HenryG has quit IRC | 06:43 | |
| *** marun has quit IRC | 06:49 | |
| *** vkozhukalov has quit IRC | 06:51 | |
| *** djbkd has quit IRC | 06:52 | |
| *** dzyu has quit IRC | 06:53 | |
| *** djbkd has joined #openstack-neutron | 06:54 | |
| *** djbkd has quit IRC | 06:54 | |
| *** dzyu has joined #openstack-neutron | 06:54 | |
| *** garyk has joined #openstack-neutron | 06:56 | |
| *** Abhishek_ has joined #openstack-neutron | 07:07 | |
| *** HenryG has joined #openstack-neutron | 07:10 | |
| *** jlibosva has joined #openstack-neutron | 07:28 | |
| *** Abhishek_ has quit IRC | 07:30 | |
| *** afazekas has joined #openstack-neutron | 07:37 | |
| *** zhhuabj has joined #openstack-neutron | 08:04 | |
| *** amotoki has quit IRC | 08:09 | |
| *** iwamoto has quit IRC | 08:09 | |
| openstackgerrit | Salvatore Orlando proposed a change to openstack/neutron: Process port_update notifications in the main agent loop https://review.openstack.org/61964 | 08:13 |
| *** fouxm has joined #openstack-neutron | 08:14 | |
| *** jorisroovers has joined #openstack-neutron | 08:15 | |
| *** vkozhukalov has joined #openstack-neutron | 08:26 | |
| *** dzyu_ has joined #openstack-neutron | 08:30 | |
| *** dzyu has quit IRC | 08:30 | |
| *** dzyu_ is now known as dzyu | 08:30 | |
| *** marios has quit IRC | 08:37 | |
| *** marios has joined #openstack-neutron | 08:38 | |
| *** amuller has joined #openstack-neutron | 08:43 | |
| *** HenryG has quit IRC | 08:43 | |
| *** HenryG has joined #openstack-neutron | 08:44 | |
| *** ygbo has joined #openstack-neutron | 08:47 | |
| *** fouxm has quit IRC | 08:59 | |
| *** salv-orlando has quit IRC | 09:00 | |
| *** jistr has joined #openstack-neutron | 09:00 | |
| *** jpich has joined #openstack-neutron | 09:07 | |
| *** networkstatic has quit IRC | 09:07 | |
| *** irenab_ has joined #openstack-neutron | 09:09 | |
| *** fouxm has joined #openstack-neutron | 09:17 | |
| *** yamahata has quit IRC | 09:19 | |
| *** Mierdin has quit IRC | 09:23 | |
| *** ashaikh has quit IRC | 09:23 | |
| *** safchain has joined #openstack-neutron | 09:24 | |
| irenab_ | Hi, having a get_py27 failures and need some advise. There are two tests that fails that are not related to the patch: https://review.openstack.org/#/c/53609/. Please advise | 09:27 |
| enikanorov | irenab_: looking | 09:34 |
| *** vkozhukalov has quit IRC | 09:35 | |
| irenab_ | thanks! | 09:36 |
| openstackgerrit | gongysh proposed a change to openstack/neutron: merge add-flow() calls https://review.openstack.org/61341 | 09:37 |
| *** Abhishek_ has joined #openstack-neutron | 09:39 | |
| enikanorov | irenab_: i believe your patchset has nothing to do with the test failure. please file a bug for neutron.tests.unit.openvswitch.test_ovs_tunnel.TunnelTestWithMTU.test_daemon_loop failre and do 'recheck bug xxx' where xx is the new bug id | 09:44 |
| irenab_ | enikanorov: what project should I submit the bug to? | 09:45 |
| enikanorov | neutron | 09:45 |
| irenab_ | enikanorov: the problem is when I run this test stand alone it passes OK... | 09:46 |
| enikanorov | does it fail in your environment? | 09:47 |
| enikanorov | anyway, file a bug if it is anykind of persistent or random failures | 09:48 |
| irenab_ | enikanorov: thank you, will file a bug. And yes, now it happend in my env too | 09:49 |
| *** dzyu has quit IRC | 09:50 | |
| *** vkozhukalov has joined #openstack-neutron | 09:50 | |
| *** nati_ueno has quit IRC | 09:53 | |
| *** nati_ueno has joined #openstack-neutron | 09:55 | |
| *** dzyu has joined #openstack-neutron | 10:01 | |
| *** krast has quit IRC | 10:21 | |
| *** fouxm has quit IRC | 10:24 | |
| *** fouxm has joined #openstack-neutron | 10:25 | |
| *** Jianyong has quit IRC | 10:28 | |
| *** Jianyong has joined #openstack-neutron | 10:29 | |
| *** fouxm has quit IRC | 10:29 | |
| *** dzyu has quit IRC | 10:34 | |
| *** fouxm has joined #openstack-neutron | 10:37 | |
| *** vkozhukalov has quit IRC | 10:40 | |
| *** jp_at_hp has joined #openstack-neutron | 10:41 | |
| *** vkozhukalov has joined #openstack-neutron | 10:52 | |
| *** Jianyong has quit IRC | 10:52 | |
| openstackgerrit | Rajesh Mohan proposed a change to openstack/neutron: Implements: blueprint fwaas-service-insertion Implements: blueprint neutron-services-insertion-chaining-steering https://review.openstack.org/62599 | 11:04 |
| openstackgerrit | Tomoko Inoue proposed a change to openstack/neutron: Can't create a firewall for admin tenant when at least one other tenant has a firewall. https://review.openstack.org/62600 | 11:05 |
| *** amuller_ has joined #openstack-neutron | 11:10 | |
| *** nati_ueno has quit IRC | 11:10 | |
| *** nati_ueno has joined #openstack-neutron | 11:11 | |
| *** amuller has quit IRC | 11:11 | |
| *** nati_uen_ has joined #openstack-neutron | 11:12 | |
| *** pcm has joined #openstack-neutron | 11:14 | |
| *** safchain has quit IRC | 11:14 | |
| *** nati_ueno has quit IRC | 11:15 | |
| *** pcm has quit IRC | 11:16 | |
| *** pcm has joined #openstack-neutron | 11:16 | |
| *** marun has joined #openstack-neutron | 11:17 | |
| *** salv-orlando has joined #openstack-neutron | 11:28 | |
| marun | so is it just me or is devstack broken without this patch? https://review.openstack.org/#/c/61663 | 11:29 |
| marun | I know it's not acceptable in its current form, but are people just applying the patch manually or ?? | 11:30 |
| *** unicell has joined #openstack-neutron | 11:36 | |
| *** unicell has joined #openstack-neutron | 11:36 | |
| *** zoresvit has joined #openstack-neutron | 11:45 | |
| *** salv-orlando has quit IRC | 11:47 | |
| *** HenryG has quit IRC | 11:50 | |
| *** chandankumar has quit IRC | 11:53 | |
| *** chandankumar has joined #openstack-neutron | 11:53 | |
| openstackgerrit | Cyril Roelandt proposed a change to openstack/python-neutronclient: Remove a debugging print statement https://review.openstack.org/62617 | 11:58 |
| *** salv-orlando has joined #openstack-neutron | 12:10 | |
| *** salv-orlando has quit IRC | 12:10 | |
| *** Abhishek_ has quit IRC | 12:17 | |
| *** bashok has quit IRC | 12:22 | |
| *** ihrachyska has joined #openstack-neutron | 12:23 | |
| *** dave_tucker has joined #openstack-neutron | 12:24 | |
| *** ian12 has joined #openstack-neutron | 12:30 | |
| *** marios_ has joined #openstack-neutron | 12:33 | |
| *** marios_ has quit IRC | 12:33 | |
| *** jorisroovers has quit IRC | 12:40 | |
| openstackgerrit | Cyril Roelandt proposed a change to openstack/python-neutronclient: Python 3: fix a call to ugettext() https://review.openstack.org/62626 | 12:43 |
| *** afazekas has quit IRC | 13:02 | |
| *** nati_uen_ has quit IRC | 13:06 | |
| *** yamahata has joined #openstack-neutron | 13:17 | |
| *** afazekas has joined #openstack-neutron | 13:22 | |
| *** ian12 has quit IRC | 13:28 | |
| *** afazekas has quit IRC | 13:51 | |
| *** geekinutah has quit IRC | 13:55 | |
| *** geekinutah has joined #openstack-neutron | 13:56 | |
| *** HenryG has joined #openstack-neutron | 14:00 | |
| *** heyongli has joined #openstack-neutron | 14:07 | |
| openstackgerrit | Oleg Bondarev proposed a change to openstack/neutron: LBaaS: agent monitoring and instance rescheduling https://review.openstack.org/59743 | 14:19 |
| anteaya | enikanorov: nice guidance on addressing failing tests in logs, thank you | 14:21 |
| *** banix has joined #openstack-neutron | 14:22 | |
| *** afazekas has joined #openstack-neutron | 14:23 | |
| *** jorisroovers has joined #openstack-neutron | 14:23 | |
| *** yamahata__ has quit IRC | 14:27 | |
| *** yamahata__ has joined #openstack-neutron | 14:28 | |
| *** julim has joined #openstack-neutron | 14:31 | |
| *** yamahata has quit IRC | 14:32 | |
| *** irenab_ has quit IRC | 14:32 | |
| *** markmcclain has quit IRC | 14:33 | |
| *** yamahata has joined #openstack-neutron | 14:34 | |
| anteaya | this patchset is needed for a critical bug fix, it would be great to get some reviews on it please: https://review.openstack.org/#/c/21946/ | 14:39 |
| anteaya | everyone welcome to review | 14:39 |
| *** heyongli has quit IRC | 14:45 | |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron: LBaaS L7 model (WIP) https://review.openstack.org/61721 | 14:47 |
| *** mtreinish has quit IRC | 14:48 | |
| *** heyongli has joined #openstack-neutron | 14:51 | |
| *** mtreinish has joined #openstack-neutron | 14:51 | |
| *** jecarey has joined #openstack-neutron | 14:57 | |
| *** chandankumar has quit IRC | 14:59 | |
| *** markmcclain has joined #openstack-neutron | 15:01 | |
| *** SushilKM has quit IRC | 15:02 | |
| kruskakli | when patch-set is created at Gerrit, my Jenkins build is triggered. However, I'm currently running my tests against 'master' but I guess I should run on the topic-branch, or? | 15:03 |
| kruskakli | if so, can I get hold of the topic branch in quation from my Jenkins build script? | 15:04 |
| kruskakli | in question.. | 15:04 |
| anteaya | kruskakli: you should be able to | 15:05 |
| anteaya | zuul filters on the topic branch I do believe | 15:06 |
| * anteaya digs | 15:06 | |
| kruskakli | hm..ok Iäll try to dump my environment from within my build script to see if any info exist about the topic-branch | 15:06 |
| anteaya | kruskakli: this might be a place to begin: http://git.openstack.org/cgit/openstack-infra/zuul/tree/zuul/trigger/gerrit.py | 15:07 |
| anteaya | I'll keep looking | 15:07 |
| kruskakli | ok, thanx | 15:07 |
| anteaya | I'm going to go with event branch is what we are after: http://git.openstack.org/cgit/openstack-infra/zuul/tree/zuul/trigger/gerrit.py#n50 | 15:08 |
| anteaya | kruskakli: you know you can set up a zuul yourself and use it for 3rd party testing? | 15:09 |
| anteaya | http://ci.openstack.org/running-your-own.html | 15:10 |
| kruskakli | anteaya: I have a very simple start script that initiates a rather complex test case running in a vagrant instance | 15:10 |
| kruskakli | so if just can get hold of the topic-branch I would be fine | 15:10 |
| anteaya | okay | 15:11 |
| anteaya | well do you see the link I gave you? | 15:11 |
| *** edle has joined #openstack-neutron | 15:11 | |
| anteaya | that is how zuul identifies the topci branch | 15:11 |
| anteaya | does that help you? | 15:11 |
| kruskakli | I hope so, I'll see if my bash script gets that info among the environment variables | 15:12 |
| kruskakli | from Jenkins that is | 15:12 |
| edle | hello I have an issue with nova compute and api logs complaining about Neutron client authentication failed: Connection to neutron failed: Unable to find the server at has anyone ran into this issue? | 15:12 |
| *** safchain has joined #openstack-neutron | 15:13 | |
| anteaya | edle: do you have a url for a patch? | 15:13 |
| *** otherwiseguy has joined #openstack-neutron | 15:15 | |
| edle | url for a patch? im not sure what you mean url for a patch | 15:15 |
| *** amritanshu_RnD has quit IRC | 15:15 | |
| anteaya | edle: is this a development question? | 15:16 |
| *** jdev789 has joined #openstack-neutron | 15:16 | |
| edle | not a development question | 15:16 |
| anteaya | ah okay | 15:17 |
| larsks | edle: Does "keystone endpoint-get --service network" show the correct URL for your neutron server? | 15:17 |
| larsks | And can you contact that url (e.g., using curl) from your compute nodes? | 15:17 |
| *** ian12 has joined #openstack-neutron | 15:18 | |
| edle | larsks: yes shows the correct url network.publicURL | http://10.198.0.191:9696/ | 15:19 |
| larsks | How about the setting of neutron_url in /etc/nova/nova.conf? | 15:20 |
| edle | when i run curl url i get {"versions": [{"status": "CURRENT", "id": "v2.0", "links": [{"href": "http://10.198.0.191:9696/v2.0", "rel": "self"}]}]} | 15:20 |
| *** dave_tucker has quit IRC | 15:20 | |
| edle | in the nova.conf i have neutron_url=http://10.198.0.191:9696 and neutron_admin_auth_url=http://10.198.0.191:35357/v2.0 | 15:21 |
| larsks | Hmm. All that looks fine. If you run your nova services with debug logging enabled, maybe you'll see a more useful error message? | 15:23 |
| *** heyongli has quit IRC | 15:23 | |
| edle | ok ill try that | 15:23 |
| *** yamahata has quit IRC | 15:26 | |
| larsks | edle: What shows up after the "at " in your error message? Is it empty? That error actually comes from the httplib2 module, and seems to suggest a dns problem (which is odd, because you're using ip addresses). | 15:28 |
| larsks | If there's a backtrace with that error, can you post it somewhere? | 15:28 |
| *** John______ has joined #openstack-neutron | 15:29 | |
| openstackgerrit | Assaf Muller proposed a change to openstack/neutron: Add VXLAN example to ovs_neutron_plugin.ini https://review.openstack.org/62684 | 15:30 |
| John______ | Does 'security groups' rules apply for both Fixed IPs and Floating IPs or just to Floating IPs? | 15:30 |
| larsks | John______: security groups are realized as iptables rules on your compute hosts. They know nothing about floasting ips...which is to say, it doesn't matter if you use a floating ip or fixed ip to contact your host, you'll still hit the same security groups. | 15:31 |
| *** amuller_ has quit IRC | 15:33 | |
| edle | in nova api log its Caught error: Connection to neutron failed: Maximum attempts reached then File "/usr/lib/python2.7/site-packages/nova/api/openstack/__init__.py", line 119, in __call__ 2013-12-16 16:45:41.772 1432 TRACE nova.api.openstack return req.get_response(self.application) 2013-12-16 16:45:41.772 1432 TRACE nova.api.openstack File "/usr/lib/python2.7/site-packages/webob/request.py", line 1296, in send 2013-12 | 15:34 |
| kruskakli | aha, a new patch-set, I get: GERRIT_BRANCH=master in my env.vars | 15:34 |
| openstackgerrit | Oleg Bondarev proposed a change to openstack/neutron: LBaaS: agent monitoring and instance rescheduling https://review.openstack.org/59743 | 15:35 |
| edle | compute logs show Neutron client authentication failed: Connection to neutron failed: Unable to find the server at and Error: Connection to neutron failed: Maximum attempts reached File "/usr/lib/python2.7/site-packages/nova/compute/manager.py", line 1037, in _build_instance 2013-12-17 09:31:40.537 21221 TRACE nova.compute.manager [instance: e4712dae-c8c9-429e-9e9d-7fda3638b2a6] set_access_ip=set_access_ip) 2013-12-17 0 | 15:36 |
| larsks | edle: I don't see anything there that helps. Sorry! | 15:36 |
| larsks | Do you see any errors in the neutron server logs? | 15:36 |
| anteaya | kruskakli: well done | 15:37 |
| anteaya | kruskakli: anything for topic branch? | 15:37 |
| openstackgerrit | Dan Prince proposed a change to openstack/neutron: Update the descriptions for the log cfg opts https://review.openstack.org/62687 | 15:37 |
| John______ | Thanks larsks. So no iptables rules are define at a netns level, only at compute level? | 15:38 |
| edle | in neutron server logs i Mapping physical network extnet to bridge br-bond0 2013-12-17 08:36:52.359 19613 ERROR neutron.openstack.common.rpc.amqp [-] Exception during message handling, would that mess up communication with nova? | 15:38 |
| anteaya | larsks: are you active in #openstack at all? | 15:38 |
| larsks | John______: Well, there are no *security group* related iptables rules defined at a netns level. All the security group rules are associated with linux bridge devices on your compute node. | 15:39 |
| larsks | anteaya: Occasionally. | 15:39 |
| anteaya | larsks: can I ask you to take this conversations to that channel please? | 15:39 |
| anteaya | s/this/these | 15:39 |
| John______ | Thanks larsks and sorry anteaya | 15:39 |
| larsks | Sure. | 15:40 |
| edle | in l3agent logs i get ERROR neutron.common.legacy [-] Skipping unknown group key: firewall_driver | 15:40 |
| anteaya | larsks: thanks | 15:40 |
| anteaya | John______: I understand your position, so if we can get your questions answered in #openstack, that works best | 15:40 |
| *** carl_baldwin has joined #openstack-neutron | 15:45 | |
| *** mlavalle has joined #openstack-neutron | 15:50 | |
| *** alex_klimov has quit IRC | 15:50 | |
| *** vkozhukalov has quit IRC | 15:51 | |
| anteaya | another call for reviews on https://review.openstack.org/#/c/21946/ the latest patch passed the check tests and it fixes a critical bug | 15:51 |
| *** John______ has quit IRC | 15:52 | |
| *** salv-orlando has joined #openstack-neutron | 15:53 | |
| *** alexpilotti has joined #openstack-neutron | 15:54 | |
| openstackgerrit | Sylvain Afchain proposed a change to openstack/neutron: Re-submit "ML2 plugin should not delete ports on subnet deletion" https://review.openstack.org/62696 | 15:54 |
| *** jecarey has quit IRC | 15:56 | |
| salv-orlando | marun: it seems it is unless you enable load balancing | 15:57 |
| marun | salv-orlando: ? | 15:57 |
| salv-orlando | because lbaas migrations will add the 'agents' table which ends up being missing | 15:57 |
| marun | salv-orlando: ah, the migrations | 15:58 |
| salv-orlando | marun: why that happened its a question mark for me too | 15:58 |
| marun | salv-orlando: so it only affects people like me that don't configure devstack for lbaas | 15:58 |
| mlavalle | beagles: ping | 16:01 |
| salv-orlando | marun: likewise | 16:01 |
| beagles | mlavalle pong.. s'bout that time, yeah? | 16:02 |
| mlavalle | beagles: yeah | 16:02 |
| markmcclain | yep.. I think we're all here | 16:02 |
| *** arosen-home has joined #openstack-neutron | 16:02 | |
| beagles | okay.. sweet | 16:02 |
| arosen-home | Hi sorry i was running a few min late. | 16:03 |
| beagles | arosen-home: just in time :) | 16:03 |
| beagles | sooo... | 16:03 |
| beagles | really, a lot of the work I felt was required for nova-parity/deprecating nova-network has been happening | 16:04 |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Add support for NSX/NVP Metadata services https://review.openstack.org/49415 | 16:04 |
| beagles | mlavalle has been working on the API gaps and testing thereof | 16:04 |
| beagles | or driving if that is a better word to use | 16:04 |
| *** dhellmann has joined #openstack-neutron | 16:04 | |
| beagles | anteaya et al. have been driving getting the gate on the go | 16:04 |
| beagles | and from reading mlavalle's older work on it and anteaya's they've got a handle on it so from the parity perspective... the question would be that whether those efforts are going to hit the goals that are required by the parity effort | 16:06 |
| anteaya | etherpad links, nova-parity: https://etherpad.openstack.org/p/icehouse-neutron-nova-parity-prelim-outline | 16:07 |
| anteaya | tempest: https://etherpad.openstack.org/p/icehouse-summit-qa-neutron | 16:07 |
| anteaya | beagles: let's start with your evaluation | 16:08 |
| anteaya | do you think the current efforts will hit the nova-parity goals? | 16:08 |
| marun | salv-orlando: btw, am making good progress on dhcp agent rework. the tests are the hardest part. | 16:08 |
| salv-orlando | marun: they always are | 16:08 |
| beagles | anteaya: if you are referring to the milestones, then I think it is obvious they do not | 16:09 |
| marun | salv-orlando: shouldn't have to be :( I've started being tougher in review on testing. I think it should matter as much as the code | 16:09 |
| beagles | anteaya: at least with respect to time frames... cuz that is all blown at this point | 16:09 |
| anteaya | how can we reframe it so that we acknowledge what work has been done | 16:10 |
| beagles | if we are talking about details, I'm not sure that etherpad for the milestones is sufficient | 16:10 |
| anteaya | and is being done? | 16:10 |
| mlavalle | I think we need a sort of "checklist" or explicit "yardstick" that tells us when we have achieved parity | 16:11 |
| beagles | +1 | 16:11 |
| mlavalle | so we can all drive towards that goal | 16:11 |
| anteaya | here is my problem | 16:11 |
| beagles | way back when I started with this https://wiki.openstack.org/wiki/NovaNetNeutronParity (in a different form, mind) | 16:11 |
| anteaya | as a group neutron is great creating very detailed evaluations of status | 16:11 |
| anteaya | the issue is getting actual code to come out of this work | 16:12 |
| anteaya | I'm not in favour of another report | 16:12 |
| beagles | me neither... | 16:12 |
| anteaya | so how do we increase the volume of work done in this area? | 16:12 |
| beagles | a simple checklist of what needs to be done, who is doing it, when its done and if its verified... | 16:12 |
| anteaya | since actually producing code is the crux of the issue | 16:12 |
| beagles | I'm not sure if we need to increase it so much as put it in a directed form | 16:13 |
| anteaya | or someone active in channel recrutiing people and answering questions | 16:13 |
| anteaya | since even the best tool with result in questions | 16:13 |
| anteaya | and if folks don't have a place to ask them, they give up | 16:13 |
| beagles | let me put it a different way, with respect to parity, how do we know if we are done? | 16:14 |
| *** HenryG has quit IRC | 16:14 | |
| markmcclain | it's really 2 parts | 16:14 |
| *** jistr has quit IRC | 16:14 | |
| markmcclain | 1 we have a solution for multi-host | 16:14 |
| *** HenryG has joined #openstack-neutron | 16:14 | |
| markmcclain | 2 we have test parity with existing nova-net tests | 16:14 |
| markmcclain | especially when we're proxying calls | 16:15 |
| salv-orlando | there also some 'minor' features such as auto-assign floating ips that can be potentially classified as 'low-hanging-fruit' | 16:15 |
| salv-orlando | and where we could leverage new contributors a lot. I have already assigned the one mentioned above | 16:15 |
| salv-orlando | and following the dev | 16:15 |
| salv-orlando | but if there were a list of high-priority features needed for feature-parity it would be better | 16:16 |
| salv-orlando | this is of course as far as feature parity is concerned, which is also instrumental to testing parity | 16:16 |
| anteaya | salv-orlando: following up with the dev is great, mentoring counts for a lot | 16:17 |
| *** garyk has quit IRC | 16:17 | |
| salv-orlando | for testing parity, I think the plan would be to sort out full isolation/parallel (and we're almost there), then have a detailed report of what we need for the full tempest job to run | 16:18 |
| salv-orlando | on neutron | 16:18 |
| beagles | is "quality of default implementation" implict in the test parity high level goal? | 16:18 |
| salv-orlando | this would kind of produce also, as a side effect, a list of unsupported use cases, which will be hopefully small | 16:18 |
| mlavalle | so there it is, part of the yardstick is to get the full tempest job run, correct? | 16:18 |
| salv-orlando | Idk perhaps I'm steering the conversation in the wrong direction. | 16:19 |
| beagles | the way I see it, there are two perspectives - complementary - but not the same | 16:20 |
| beagles | simple feature parity vs. what it takes to deprecate nova-networking | 16:20 |
| markmcclain | mlavalle: yes we need to move away from the asymmetirc testing we've got now | 16:21 |
| beagles | from an engineering perspective, the former is more palatable | 16:21 |
| beagles | in the bigger picture, the latter is the more relevant | 16:21 |
| *** SushilKM has joined #openstack-neutron | 16:21 | |
| beagles | does that make sense to everybody? | 16:23 |
| salv-orlando | I think the former is just one dimension for measuring parity | 16:23 |
| salv-orlando | when you have parity along all dimensions, people would say it's ok to deprecate nova-network | 16:24 |
| *** jecarey has joined #openstack-neutron | 16:24 | |
| beagles | nice way to put it | 16:24 |
| salv-orlando | so - features, currently neutron supports more features than nova-ntetwork, but nova-net features are not a subset of neutron features | 16:24 |
| salv-orlando | second - quality. This is way harder to measure, but I think there's a general consensus that similar gate reliability would be a satisfactory measures | 16:25 |
| mlavalle | salv-orlando, exactly right… that's part of the challenge of getting the full tempest job run fixed | 16:25 |
| salv-orlando | meaning you run the same jobs, and they have a similar success rate | 16:25 |
| mlavalle | things don't translate directly | 16:25 |
| salv-orlando | I can't think of any other dimension for measuring parity now, anyone? | 16:26 |
| * markmcclain runs to another meeting… will read scollback | 16:26 | |
| *** markmcclain has quit IRC | 16:26 | |
| beagles | unless it is implied by features, use-case is a dimension | 16:27 |
| beagles | eg. how can the features be combined and deployed | 16:27 |
| beagles | the rationale behind describing how FlatManager can be emulated in neutron is a use case | 16:28 |
| beagles | I know I'm talking semantics here, but... | 16:28 |
| *** jecarey_ has joined #openstack-neutron | 16:28 | |
| beagles | feature parity alone does not imply that use-cases are met | 16:28 |
| salv-orlando | that's fine. It's a dimension, albeit not linearly independent from the others | 16:29 |
| * anteaya reminds people to click the topmost righthand button in an etherpad and add your name | 16:29 | |
| beagles | e.g we have floating IPs but the FlatDHCPManager with L3 would be considered broken from a user perspective | 16:30 |
| salv-orlando | for instance I would consider multi-host a use case rather than a feature | 16:30 |
| beagles | +2 | 16:30 |
| *** SushilKM has quit IRC | 16:30 | |
| salv-orlando | How far are we from itemising features and use cases so that gaps can be easily identified and dealt with? | 16:31 |
| *** jecarey has quit IRC | 16:31 | |
| beagles | some not far at all... | 16:32 |
| beagles | I need to grab some bandwidth to figuring out how to best describe some of them, but I think I've alluded to them in https://wiki.openstack.org/wiki/NovaNetNeutronRecipes | 16:34 |
| beagles | multi-host remains a big ? | 16:34 |
| beagles | but... | 16:35 |
| beagles | I apologize, I took this off the rails a bit... | 16:35 |
| sgran | multi-host needs a VRRP gateway, IMHO | 16:36 |
| beagles | what I was really hoping to clarify in this discussion was whether mlavalle feels that he has the API/full-tempest-gateway stuff in hand ... | 16:36 |
| *** Mierdin has joined #openstack-neutron | 16:36 | |
| beagles | and also anteaya... | 16:36 |
| sgran | DHCP agents can already be auto-scheduled to all compute hosts, for instance | 16:36 |
| beagles | at one point I fretted about having to champion that myself and I was relieved to see others taking it ;) | 16:37 |
| beagles | (with respect to parity that is) | 16:37 |
| anteaya | I have been focusing on the gate and bugs | 16:37 |
| mlavalle | beagles: I expect to have the gap analysis for the API tests to be finished this week | 16:37 |
| anteaya | mlavalle: but here is the thing | 16:37 |
| beagles | if so then I feel comfortable focusing on the use-case thing etc. | 16:37 |
| anteaya | a great gap analysis isn't getting the gap closed | 16:38 |
| *** SushilKM has joined #openstack-neutron | 16:38 | |
| anteaya | how is it going getting new tempest tests written | 16:38 |
| beagles | and contributing to mlavalle and anteaya's efforts, because without success in their efforts it's all for naught | 16:38 |
| anteaya | and merged? | 16:38 |
| marun | isn't gap analysis a necessary precursor to prioritizing? | 16:38 |
| salv-orlando | mlavalle: the bottom line of what anteaya suggest is to pipeline gap analysis and gap closing, I think | 16:38 |
| anteaya | yes, we have to get some tests in tempest | 16:39 |
| anteaya | so far I have seen work from enikanorov EmilienM rosella_s yfried and jlibosva | 16:39 |
| anteaya | for tempest patches | 16:39 |
| anteaya | have a I missed anyone? | 16:39 |
| mlavalle | agree, with you, The gap analysis is good so we can invite other contributos to participate…. | 16:39 |
| salv-orlando | yes, but just to make sure I am not confused. Are we talking about "API tests"? | 16:39 |
| salv-orlando | or are we including also "scenario tests" in this analysis | 16:40 |
| anteaya | I'm operating from the perspective that points on this etherpad need to be covered: https://etherpad.openstack.org/p/icehouse-summit-qa-neutron | 16:40 |
| anteaya | mlavalle: seems to be focusing on the api tests, which is great, but we need a broader focus to address all the issues | 16:41 |
| anteaya | mlavalle: do you want to stay focused on the api? | 16:41 |
| mlavalle | anteaya: not necessarilly. I want to do the best for the group. In our original conversation, we agreed I was going to do the gap analysis to drive to contributos to write api tests | 16:42 |
| * beagles backs up... | 16:43 | |
| anteaya | mlavalle: oh, I was operating under the understanding that you were going to take on that task entirely | 16:43 |
| anteaya | which includes tests being merged | 16:43 |
| beagles | what API are we talking about, neutron's API or the API that is exercised via nova? | 16:43 |
| anteaya | this might be my mis-understanding | 16:43 |
| mlavalle | anteaya: yes i can do that | 16:43 |
| beagles | we need the former but the latter is essential | 16:43 |
| anteaya | mlavalle: you can champion tests being merged into tempest? | 16:43 |
| salv-orlando | I think we are perhaps keeping derailing this conversation. Perhaps we should let beagles lead the discussion. | 16:43 |
| beagles | :) | 16:44 |
| beagles | okay... the discussion has been good and I appreciate it, but to save time I'll be direct | 16:44 |
| mlavalle | anteaya: let's continue this conversation after the meeting | 16:44 |
| beagles | we need full exercising and vetting of the interaction between nova and neutron, that means the module that is in nova/networking/neutronv2 | 16:45 |
| beagles | behavior/testing etc all has to be vetted | 16:45 |
| beagles | we need the use case thing | 16:45 |
| beagles | we need the gates | 16:45 |
| beagles | and full testing would be awesome, but outliers can be triaged | 16:46 |
| beagles | and we need to be solid and reasonably fast for normal workflows | 16:46 |
| beagles | (for default implementation) | 16:46 |
| beagles | who has what? | 16:46 |
| beagles | I have use-cases | 16:46 |
| arosen-home | I've been working on the interaction between nova <-> neutron lately. I can take lead on that. | 16:47 |
| beagles | sweet! | 16:47 |
| salv-orlando | I have speeding up the ovs agent | 16:48 |
| salv-orlando | marun has speeding and reliability for the dhcp agent | 16:48 |
| salv-orlando | this much I know. I do not see at the moment issues needing intervention on the l3 agent | 16:48 |
| salv-orlando | nati-ueno, markmcclain and in part myself did several fixes in havana (it was kind of broken before) | 16:49 |
| marun | arosen-home: i'd like to talk to you about nova/neutron integration, maybe co-work this week? | 16:49 |
| salv-orlando | I am also working on enabling parallel testing, and should deliver that by I-2 | 16:49 |
| beagles | anteaya: you've got gates covered right? | 16:50 |
| beagles | and mlavalle full tempest? | 16:50 |
| arosen-home | marun: I fly out tomorrow :/ | 16:50 |
| beagles | (damn it is a crappy time of year to get stuff done) | 16:50 |
| marun | arosen-home: ah, next year then :D | 16:50 |
| anteaya | beagles: affirmative | 16:50 |
| arosen-home | marun: sure thing. | 16:50 |
| marun | arosen-home: i'll be back for the 6th | 16:50 |
| arosen-home | sorry guys but i have to run and catch the train. I'll read back up when i reconnect . | 16:51 |
| beagles | arosen-home: cheers, and thanks | 16:51 |
| salv-orlando | adios arosen | 16:51 |
| anteaya | arosen-home: thanks for being here | 16:51 |
| *** arosen-home has quit IRC | 16:51 | |
| anteaya | beagles: if you need to see me do anything more or better regarding the gate, please tell me | 16:51 |
| beagles | anteaya: no doubt ;) | 16:52 |
| salv-orlando | I am pretty confident we can bring the various neutron agents making up the default framework to a decent reliability and scalability status | 16:52 |
| beagles | oooo | 16:53 |
| marun | nothing like a rewrite ;) | 16:53 |
| * beagles cuts and pasts | 16:53 | |
| salv-orlando | because I think those specific items are covered | 16:53 |
| beagles | pastes | 16:53 |
| beagles | mlavalle: what am I missing? | 16:53 |
| mlavalle | beagles: i can help with the full tempest job…. I thought someone else was taking tking care of that, but if that is not the case i'll glad to help | 16:54 |
| mlavalle | can I ask a stupid question? | 16:55 |
| beagles | mlavalle: I don't want to dump it on you!!! :) Was it me that you thought was doing it? ;) | 16:55 |
| beagles | fire away | 16:55 |
| mlavalle | beagles: yeah… but we can work together on it | 16:56 |
| salv-orlando | we had rossella_s working on the full tempest job. she was doing some analysis of what items needs to be fulfilled | 16:56 |
| beagles | mlavalle: okay.. cool... I had this feeling you taking it :) but I can keep it for sure. | 16:57 |
| salv-orlando | but I had no update from here in a while. Perhaps she's been syncing with you guys? | 16:57 |
| beagles | right rossella_s wasn't feeling well and couldn't make it today | 16:57 |
| beagles | let's do it this way.. I'll keep it and hookup with you and rossella_s as soon as our tz's intersect again | 16:58 |
| beagles | mlavalle: where are you TZ wise? | 16:58 |
| mlavalle | beagle rosella_s and I can synch up on that….. in any case i'll be glad to help | 16:58 |
| mlavalle | stupid question: nova deprecation is not going to happen when we say we have achieved parity…. That decision is beyond us…. Do we have a buy in with those making the decision as to what parity means? | 16:58 |
| beagles | fwiw: I'd like to see if I can scrounge some manpower for some directed efforts, so that would work out great | 16:58 |
| beagles | mlavalle: you are right on that point... | 16:59 |
| *** afazekas has quit IRC | 16:59 | |
| salv-orlando | mlavalle: not able to answer that question. You might want a nova-core for that. From my perspective it's about filling feature/use case gaps and ensuring users can trust neutron as much as nova-network | 16:59 |
| anteaya | mlavalle: including sdague and russellb in any nova-parity conversation would increase buy in on the result | 16:59 |
| beagles | mlavalle: having arosen on board doesn't hurt... I didn't have much to report at the nova meetings the past two weeks | 16:59 |
| edle | hi larsks i fixed my issue i checked everything in the docs and was exhausted of what the issues can be so i rebooted the server and that fixed it | 16:59 |
| beagles | mlavalle: but this week I have something to say | 16:59 |
| beagles | so... we can get buy-in | 17:00 |
| anteaya | edle: please chat in #openstack for support questions, thank you | 17:00 |
| beagles | fwiw: that is about the least stupid question I think I've ever heard :) | 17:01 |
| mlavalle | anteaya: shouldn't that conversation somehting we should have quickly? So we are all pointing in the right direction? | 17:01 |
| anteaya | mlavalle: sure | 17:01 |
| anteaya | I just need to know what your bottom line is | 17:02 |
| anteaya | my bottom line is tests in tempest | 17:02 |
| anteaya | you are doing good work | 17:02 |
| *** rohit404 has joined #openstack-neutron | 17:02 | |
| anteaya | but none of it counts for all of openstack unless it results in tests | 17:02 |
| anteaya | I just need to hear what you want | 17:02 |
| mlavalle | my bottom line is tests in tempest…. no oubt about it…. the gap analysis has been an exercise in trying to get more people to help, that's all | 17:03 |
| anteaya | let's evaluate it on those terms then | 17:03 |
| anteaya | how have you done in that regard? | 17:03 |
| * beagles snickers | 17:04 | |
| anteaya | not a questions to evoke snickers | 17:04 |
| beagles | anteaya: open source is a tough gig in that respect | 17:04 |
| anteaya | a reality check | 17:04 |
| anteaya | no kidding | 17:04 |
| beagles | a lot of times it is a "if you build it they will come" | 17:04 |
| anteaya | particularly in this neighbourhood | 17:04 |
| anteaya | be we have to start with being honest with ourselves and with each other | 17:05 |
| * beagles nods | 17:05 | |
| anteaya | beagles: sure, if you are ploughing up your own corn field | 17:05 |
| anteaya | but if we aren't getting the results we need when we need them | 17:05 |
| anteaya | we have to start by saying so | 17:05 |
| *** garyk has joined #openstack-neutron | 17:05 | |
| anteaya | and I can't put words in your mouth mlavalle | 17:06 |
| anteaya | so I have to hear what you have to say in this regard | 17:06 |
| anteaya | please | 17:06 |
| beagles | that's good grist actually... | 17:06 |
| mlavalle | not well, there are not enough test in tempest yet | 17:07 |
| beagles | I'll reach out to see if there are any resources that can be focused | 17:07 |
| *** Abhishek_ has joined #openstack-neutron | 17:07 | |
| anteaya | mlavalle: great, thanks for stating my own evaluation of the situation | 17:07 |
| enikanorov | mlavalle: I have two engineers in my team (in addition to myself) who can help with tempest tests | 17:07 |
| beagles | no promises, but I can at the very least complain | 17:07 |
| enikanorov | and in fact already working on some patches | 17:07 |
| anteaya | mlavalle: do you have any thoughts on how to improve the situation, or would you like some suggestions? | 17:07 |
| enikanorov | mostly on load balancer tests right now | 17:08 |
| anteaya | enikanorov: do you have urls for those patches? | 17:08 |
| enikanorov | anteaya: sure. Let me find | 17:08 |
| anteaya | that would help mlavalle track progress | 17:08 |
| anteaya | enikanorov: thanks | 17:08 |
| *** jlibosva has quit IRC | 17:08 | |
| anteaya | mlavalle: do you have any thoughts for improvement or would you like some suggestions? | 17:09 |
| mlavalle | anteaya: the gap analysis is an effort of making clear what needs to be done, so others can help. In that way, it is not just me…. There is alot that needs to be implemented in the api tsts | 17:09 |
| mlavalle | I'm also open to suggestions | 17:09 |
| anteaya | mlavalle: okay here are my suggestions | 17:10 |
| anteaya | track the names of anyone who has submitted a neutron tempest patch | 17:10 |
| anteaya | check their review page everyday | 17:10 |
| enikanorov | actually it's easier to give a link to their gerrit queue, since they both have > 2 two tempest patches on review: | 17:10 |
| enikanorov | https://review.openstack.org/#/dashboard/7249 https://review.openstack.org/#/dashboard/7293 | 17:10 |
| anteaya | ensure they have all they need in terms of reviews to get their patch merged | 17:10 |
| anteaya | if you don't see activity on a patch, ask them why, offer to help with blocks, co-ordinate with -qa to get questions answered | 17:11 |
| anteaya | also ask for reviews on any patches ready to be merged and get them merged | 17:11 |
| anteaya | when current developers find that you are supportive and will drive their (tested and qualified) patches into master | 17:12 |
| anteaya | they will be motivated to submit another patch | 17:12 |
| anteaya | do this work the first thing of everyday | 17:12 |
| anteaya | enikanorov: great | 17:12 |
| anteaya | are ann and elena in channel right now? | 17:13 |
| mlavalle | anteaya: good suggestions….. how about finishing the gap analysis? | 17:14 |
| enikanorov | no, i think they working day has ended like an hour ago | 17:14 |
| anteaya | mlavalle: can you do both? | 17:14 |
| enikanorov | *their | 17:14 |
| anteaya | enikanorov: okay thanks, can you ask them to ping both myself and mlavalle in channel tomorrow | 17:14 |
| *** jorisroovers has quit IRC | 17:14 | |
| mlavalle | anteaya: sure…. I just want to make sure that part is still considred valuable…. since we are in feedback mode | 17:14 |
| anteaya | and post any blockages they are finding in their work | 17:14 |
| anteaya | mlavalle: if you can do both, my personal vote is on both, with tests in tempest getting first priority | 17:15 |
| anteaya | anyone else with an opinion? | 17:15 |
| enikanorov | ok, i'll tell them | 17:16 |
| anteaya | enikanorov: thanks | 17:17 |
| beagles | I'm vascillating on priority AND assignment.. if it were possible I'd suggest mlavalle prioritize finishing the gap and see if makes more sense to get somebody to shepherd the test creation | 17:17 |
| beagles | somebody else that is | 17:17 |
| anteaya | do we have any volunteers for shepherding test creation? | 17:17 |
| beagles | but that's just because it sucks to have a chicken and egg problem when the chicken is half created | 17:17 |
| beagles | let's look for one | 17:18 |
| anteaya | beagles: ah if someone where interested they would be following along | 17:19 |
| *** ashaikh has joined #openstack-neutron | 17:19 | |
| anteaya | were | 17:19 |
| mlavalle | beagles: what do you see me doing besides the test sheperding? | 17:19 |
| beagles | certainly out of the pool of likely candidates, they are not all listening at the moment | 17:19 |
| anteaya | how much time are we going to give to that task? | 17:19 |
| anteaya | we have about 3 weeks of actually working time remaining before i2 | 17:19 |
| anteaya | taking out holidays as actual working time | 17:20 |
| anteaya | I will be offline for a significant chunk of that | 17:20 |
| beagles | mlavalle: I don't know :) In all seriousness, if it is most appropriate that you shepherd it then by all means. | 17:21 |
| anteaya | I'm not against handing off a task to a willing volunteer, by all means, delegation and sharing is the only way to meet our goals | 17:22 |
| anteaya | but waiting for said volunteer to show up is very unproductive | 17:22 |
| beagles | no doubt | 17:23 |
| beagles | any volunteers for such an effort would have to be proactively sought out | 17:23 |
| anteaya | I'm all for that | 17:23 |
| anteaya | but in the mean time | 17:23 |
| mlavalle | yeah, that's the idea of the gap analysis…. get other people to write tests…. and to tell you the truth, I am not waiting for any volunteers…. I see my communicty job as getting neutron tests in tempest | 17:23 |
| beagles | mlavalle: cool | 17:24 |
| anteaya | are you able to sheperd patches right now? | 17:24 |
| anteaya | mlavalle: ^ | 17:24 |
| mlavalle | if you mean what you described about reviewing and guiding people, yes | 17:24 |
| anteaya | great | 17:25 |
| anteaya | thanks | 17:25 |
| anteaya | let me know how I can support you | 17:25 |
| mlavalle | and also, once the I finsh the gap analysis, writing api tests myself | 17:25 |
| anteaya | great | 17:26 |
| mlavalle | my focus lately on the gap analysis has been an attemot to scale the effort up | 17:26 |
| anteaya | did we need to discuss anything else on this topic? | 17:26 |
| anteaya | I understand | 17:26 |
| beagles | the API coverage or parity? | 17:27 |
| anteaya | we seem to have dicussed both | 17:27 |
| beagles | :) | 17:27 |
| mlavalle | before you and Beagles go, I want to know who is working on the ful tempest gate job. I think that is importnat and thought that was being covered | 17:27 |
| anteaya | I think salv-orlando felt rosella_s was but hasn't heard from her and she is sick today | 17:28 |
| anteaya | mlavalle do you and beagles want to talk to her about that and find out if she is indeed working on it, and if yes what her status is? | 17:29 |
| mlavalle | A couple of weeks ago, you asked me to talk to her and I did | 17:29 |
| anteaya | did you follow up with her since that time? | 17:29 |
| mlavalle | I was hoping today's meeting was the ocassion…. but since she is sick, i'll follow up with her | 17:29 |
| anteaya | mlavalle: thanks | 17:30 |
| mlavalle | specifically, I will probe whether she thinks she is the point person for that…. If not, that's ok, but we need to know | 17:30 |
| anteaya | if you can do so in channel, so much the better then the rest of us can read the logs/scrollback and follow along | 17:31 |
| anteaya | yes | 17:31 |
| anteaya | I totally agree | 17:31 |
| anteaya | we need to know what her understanding of her commitment is | 17:31 |
| anteaya | her bottom line in this regard | 17:31 |
| anteaya | not to change that but to address gaps if need be | 17:31 |
| mlavalle | correct | 17:32 |
| anteaya | great | 17:32 |
| beagles | at the moment the full tempest run chokes on similar things to the smoke testing | 17:32 |
| beagles | the multiple network thing...etc | 17:32 |
| anteaya | beagles: do you have any urls for the chokes? | 17:32 |
| beagles | to properly assess each test at the moment would be to run each test in isolation | 17:33 |
| beagles | what? | 17:33 |
| beagles | did I just write? | 17:33 |
| beagles | gah | 17:33 |
| * anteaya stands by | 17:33 | |
| beagles | basically you would have to run each test independently | 17:33 |
| beagles | and no | 17:33 |
| beagles | but you know the deal | 17:33 |
| anteaya | okay | 17:33 |
| anteaya | well actually I don't | 17:33 |
| beagles | if a previous test screws up the cleanup, the subsequent tests do not run | 17:34 |
| beagles | because there are multiple networks for that tenant | 17:34 |
| anteaya | ah | 17:34 |
| beagles | and because neutron doesn't order networks, then it cannot create by default order and requires that the client specify one | 17:34 |
| anteaya | is rossella_s aware of this situation? | 17:36 |
| anteaya | or do we know? | 17:36 |
| anteaya | mlavalle beagles are the two of you available for a daily status meeting? | 17:36 |
| anteaya | at least for the rest of this week? | 17:36 |
| anteaya | 2200 utc would work fine for me | 17:36 |
| mlavalle | beagles: to talk about the full tempest job? | 17:37 |
| mlavalle | anteaya^^^ | 17:37 |
| anteaya | to talk about progress | 17:37 |
| anteaya | on any issue dealing with nova-parity and tempest | 17:38 |
| anteaya | dkehn: where do you clone from for your devstack? | 17:39 |
| anteaya | git.o.o or github? | 17:39 |
| mlavalle | Thursday I won't be available…. tomorrow and friday I'm fine | 17:39 |
| dkehn | anteaya: trunk | 17:39 |
| beagles | I think same | 17:39 |
| beagles | also not avail today | 17:39 |
| anteaya | beagles: we did today already, let's talk tomorrow and friday | 17:39 |
| anteaya | and thanks | 17:39 |
| dkehn | anteaya: the only thing that is manipulated there is the lcoalrc | 17:39 |
| dkehn | s/lcoalrc/localrc | 17:40 |
| beagles | what.. you aren't going to get loads of stuff done this afternoon! | 17:40 |
| anteaya | dkehn: I'm reading your comment on https://review.openstack.org/#/c/21946/ | 17:40 |
| anteaya | and I had a conversation yesterday, with mattymo about him not seeing all the recent commits in his git log when he cloned from github | 17:41 |
| anteaya | dkehn: do you see https://review.openstack.org/#/c/57600/ in your devstack git log? | 17:41 |
| dkehn | anteaya: basically if you don't mock the udevadm it no work locally | 17:41 |
| dkehn | haven't checked. | 17:41 |
| anteaya | k | 17:41 |
| *** zoresvit has quit IRC | 17:41 | |
| dkehn | anteaya: I was assuming it was merge, says it was merged, until I tested it here and then checked the code | 17:42 |
| anteaya | let's look at git logs | 17:42 |
| anteaya | I'll check git.o.o | 17:42 |
| *** zoresvit has joined #openstack-neutron | 17:43 | |
| anteaya | dkehn: merged: http://git.openstack.org/cgit/openstack/neutron/log/?qt=grep&q=I0e16295be66513c2257831bf62fd6a0c19f3b8a4 | 17:43 |
| *** vkozhukalov has joined #openstack-neutron | 17:44 | |
| dkehn | anteaya: ok, let me do a freesh clone and see if its there then I'll do a fetch and see if its there again, | 17:45 |
| anteaya | dkehn: sorry here is the merge: http://git.openstack.org/cgit/openstack/neutron/commit/?id=512b4e71fb4bc864e240fe34286ce67c7332da4c | 17:45 |
| anteaya | dkehn: k | 17:45 |
| dkehn | anteaya: this is what I did before | 17:45 |
| yfried | marun: I was told that a +1 from a neutron dev will get my tempest patches higher priority for review. can you please take a look? I tagged you on all of them. | 17:46 |
| dkehn | anteaya: without the the checking part | 17:46 |
| *** ygbo has quit IRC | 17:46 | |
| *** alagalah has quit IRC | 17:46 | |
| anteaya | yfried: link to urls? mlavalle and I would like to help you get your tempest patches merged | 17:47 |
| anteaya | actually linking mlavalle to all your patches as well would be good, he is supporting neutron tempest patches | 17:48 |
| yfried | https://review.openstack.org/#/c/55101/ | 17:48 |
| yfried | https://review.openstack.org/#/c/55146/ | 17:48 |
| yfried | https://review.openstack.org/#/c/60757/ | 17:48 |
| yfried | https://review.openstack.org/#/c/56889/ | 17:48 |
| yfried | https://review.openstack.org/#/c/62702/ | 17:48 |
| yfried | https://review.openstack.org/#/c/62697/ | 17:48 |
| *** SumitNaiksatam has quit IRC | 17:48 | |
| yfried | anteaya: mlavalle tnx | 17:48 |
| yfried | mlavalle: what's you're name on gerrit? can't find it to tag you | 17:49 |
| anteaya | miguel | 17:49 |
| mlavalle | yfried: minsel | 17:49 |
| anteaya | ah sorry | 17:49 |
| *** zoresvit has quit IRC | 17:49 | |
| anteaya | did I spell your first name correctly at least? | 17:49 |
| anteaya | I am a horrible speller | 17:49 |
| mlavalle | anteaya; mi name is Miguel….. minsel was assigned to me by Launchpad when I originally signed up…. I don't know why | 17:50 |
| yfried | mlavalle: tnx | 17:51 |
| yfried | mlavalle: you can expect me to spam you with tempest patches from now on.. | 17:51 |
| yfried | :) | 17:51 |
| anteaya | yfried: go go go | 17:51 |
| mlavalle | yfried: that's coll… that's what we want :-) | 17:52 |
| mlavalle | cool^^^ | 17:52 |
| *** jlibosva has joined #openstack-neutron | 17:52 | |
| anteaya | jlibosva: hello, how goes the grenade patch? | 17:53 |
| anteaya | did you get my message I left in the comments | 17:53 |
| yfried | anteaya: mlavalle: btw - since tempest-network is my main focus currently, I would love pointers, suggestions, and ideas for tests | 17:54 |
| anteaya | yfried: awesome | 17:55 |
| anteaya | what are you using for guidance so far? | 17:55 |
| jlibosva | anteaya: I found problem probably in ovs | 17:55 |
| anteaya | jlibosva: okay, what do you want to do about that? | 17:55 |
| *** markmcclain has joined #openstack-neutron | 17:55 | |
| jlibosva | anteaya: I'm trying to figure out why floating_ip test is failing | 17:55 |
| anteaya | do you need to file a bug? or talk to someone about it? | 17:56 |
| anteaya | jlibosva: okay let's start by linking to the failing test so others can follow along | 17:56 |
| jlibosva | anteaya: I want to talk to someone, I have a lot of failed to get flags for network device qr-7f907021-82: No such device in ovs log | 17:56 |
| jlibosva | the ports are in ovs-vsctl though | 17:56 |
| anteaya | jlibosva: great | 17:56 |
| *** nati_ueno has joined #openstack-neutron | 17:56 | |
| anteaya | jlibosva: let's have a link to something | 17:57 |
| anteaya | so the conversation can focus on the stacktrace | 17:57 |
| yfried | anteaya: I was using our (RedHat) test plan but I think I got the basic stuff done. | 17:57 |
| anteaya | yfried: okay, mlavalle is going to look at your patches and then he can better asses | 17:57 |
| anteaya | yfried: have you seen this etherpad: https://etherpad.openstack.org/p/icehouse-summit-qa-neutron | 17:57 |
| jlibosva | anteaya: you want me to put the logs to pastebin or somewhere? | 17:57 |
| anteaya | jlibosva: yes please | 17:58 |
| yfried | anteaya: mlavalle: I'm also using this to coordinate but so far I'm the only one - https://blueprints.launchpad.net/tempest/+spec/neutron-advanced-scenarios | 17:58 |
| marun | yfried: ok | 17:58 |
| *** ihrachyshka has joined #openstack-neutron | 17:58 | |
| yfried | so me, myself, and I are working in perfect coordination | 17:58 |
| *** safchain has quit IRC | 17:58 | |
| sdague | so danpb brought up an interesting point in nova, nova doesn't specify a timeout for ovs commands at all (it waits forever), which is a different approach than on neutron - https://review.openstack.org/#/c/61105/4 | 17:59 |
| *** ihrachyska has quit IRC | 17:59 | |
| sdague | has anyone thought about making that timeout super huge, like 300s, so only gets hit in the worse possible cases | 17:59 |
| anteaya | yfried: well perfect coordination is nice, can you add your name to the scenario tests on the etherpad please, as well as a link to the blueprint and your patches | 18:00 |
| dkehn | anteaya: figured it out, din't do a rebase | 18:00 |
| dkehn | anteaya: will update the review | 18:00 |
| *** zz_ajo is now known as ajo | 18:00 | |
| anteaya | dkehn: so the issue was local to you? | 18:00 |
| anteaya | dkehn: thanks | 18:00 |
| yfried | anteaya: I see someone already wrote my name | 18:01 |
| anteaya | markmcclain: sdague poses a good question if you are around | 18:01 |
| anteaya | salv-orlando: ^ | 18:01 |
| anteaya | yfried: we don't let good workers get passed us | 18:01 |
| anteaya | :D | 18:01 |
| *** jecarey_ has quit IRC | 18:01 | |
| markmcclain | sdague: we could make the timeout larger | 18:02 |
| anteaya | yfried: I don't see your name under scenario tests, just at the bottom of the etherpad in the list | 18:02 |
| salv-orlando | in a meeting will get back at you later | 18:02 |
| *** nati_uen_ has joined #openstack-neutron | 18:02 | |
| dkehn | markmcclain: any luck with the devstack, from yesterdays migration issues? | 18:02 |
| mlavalle | marun: I am going to review yfried patches now, but a core Neutron reviewer like you will have more weight in getting his patched merged. Are you going to take a look on them? | 18:02 |
| marun | mlavalle: i will | 18:02 |
| mlavalle | marun: thanks | 18:02 |
| yfried | anteaya: tnx for the compliment, but it seems to me that in this area you don't have much choice :) | 18:03 |
| anteaya | yfried: buddy you are singing my song | 18:03 |
| yfried | anteaya: I'm right there under "Scenario Tests" | 18:03 |
| sdague | markmcclain: the nova objections were basically that small timeouts would actually cause races, and if ovs was deadlocking for some reason, that's a different bug | 18:04 |
| *** nati_uen_ has quit IRC | 18:04 | |
| *** ian13 has joined #openstack-neutron | 18:04 | |
| anteaya | yfried: I'm blind, yes you were there all along | 18:04 |
| *** nati_uen_ has joined #openstack-neutron | 18:04 | |
| *** nati_uen_ has quit IRC | 18:05 | |
| yfried | anteaya: now I just need to see what is this doc and what am I listed for. hope no one has dibs on my kidney without asking me | 18:05 |
| anteaya | nati_ueno: nice work on your patch | 18:05 |
| *** nati_ueno has quit IRC | 18:05 | |
| *** nati_uen_ has joined #openstack-neutron | 18:05 | |
| anteaya | yfried: don't answer any craigslist ads | 18:05 |
| openstackgerrit | rcurran proposed a change to openstack/neutron: *** PROPOSAL ONLY. *** https://review.openstack.org/61583 | 18:06 |
| *** ian12 has quit IRC | 18:07 | |
| anteaya | nati_uen_: yfried would like https://review.openstack.org/#/c/21946/ tested against https://review.openstack.org/#/c/62702/ | 18:08 |
| anteaya | nati_uen_: are you able to coordinate with him to make this happen? | 18:08 |
| *** jdev789 has quit IRC | 18:08 | |
| markmcclain | sdague: the other downside with large timeouts is that they hide commands which aren't executing fast enough | 18:12 |
| *** clev has joined #openstack-neutron | 18:12 | |
| markmcclain | sometimes those indicate other problems | 18:12 |
| sdague | markmcclain: that's true, but "fast enough" is easy to prematurely shoot something in the head as well | 18:13 |
| sdague | anyway, I wanted to bring up the thought, because not having an ovs timeout on nova hasn't seemed to cause a problem | 18:13 |
| *** ian13 has quit IRC | 18:14 | |
| *** ashaikh has quit IRC | 18:15 | |
| *** dhellmann is now known as dhellmann_ | 18:17 | |
| *** suresh12 has joined #openstack-neutron | 18:20 | |
| *** SumitNaiksatam has joined #openstack-neutron | 18:21 | |
| *** harlowja_away is now known as harlowja | 18:22 | |
| anteaya | jlibosva: waiting to hear from you when you have something | 18:23 |
| anteaya | do you have to run the tests again to get the stacktraces? | 18:24 |
| *** terry_howe has joined #openstack-neutron | 18:24 | |
| jlibosva | anteaya: I;m on meeting | 18:24 |
| jlibosva | anteaya: no, it happens after stack.sh | 18:24 |
| jlibosva | anteaya: http://paste.openstack.org/show/55182/ | 18:24 |
| anteaya | great, I'll look at the paste, thanks | 18:25 |
| openstackgerrit | rcurran proposed a change to openstack/neutron: Calls in delete methods should be in reverse order as create https://review.openstack.org/61583 | 18:26 |
| terry_howe | I'm working on the unified CLI and it seems to me the neutron CLI command dhcp-agent-network-add would be better called net-dhcp-add | 18:28 |
| terry_howe | doesn't it make more sense to add DHCP agents to networks than networks to DHCP agents? | 18:29 |
| anteaya | carl_baldwin: are you available for some questions? | 18:29 |
| carl_baldwin | yes | 18:29 |
| anteaya | carl_baldwin: jlibosva is working on this patch: https://review.openstack.org/#/c/58695/ | 18:29 |
| anteaya | awesome | 18:29 |
| anteaya | and he is hitting some walls | 18:29 |
| anteaya | carl_baldwin: here is a paste from running stack.sh: http://paste.openstack.org/show/55182/ | 18:30 |
| anteaya | he needs some help addressing a floating ip issue | 18:30 |
| anteaya | he is in a meeting atm but the grenade patch is very important | 18:30 |
| anteaya | if you could help him with it we would both be greatful | 18:31 |
| anteaya | hi terry_howe do you have a link to a patch or blueprint? | 18:32 |
| anteaya | I just give better responses if I can see the same thing you are working from | 18:32 |
| carl_baldwin | anteaya: I'll take a look. | 18:33 |
| anteaya | carl_baldwin: thanks | 18:33 |
| *** vkozhukalov has quit IRC | 18:36 | |
| *** fouxm has quit IRC | 18:37 | |
| *** vkozhukalov has joined #openstack-neutron | 18:37 | |
| *** fouxm has joined #openstack-neutron | 18:38 | |
| *** fouxm has quit IRC | 18:38 | |
| *** Abhishek_ has quit IRC | 18:38 | |
| *** networkstatic has joined #openstack-neutron | 18:38 | |
| mlavalle | yfried: are you aware of isolated credentials? | 18:41 |
| *** alexpilotti has quit IRC | 18:42 | |
| *** jistr has joined #openstack-neutron | 18:44 | |
| *** ijw has joined #openstack-neutron | 18:44 | |
| *** garyk has quit IRC | 18:49 | |
| *** garyk has joined #openstack-neutron | 18:49 | |
| *** edle has quit IRC | 18:53 | |
| *** nati_ueno has joined #openstack-neutron | 18:53 | |
| *** networkstatic has quit IRC | 18:56 | |
| *** nati_uen_ has quit IRC | 18:57 | |
| *** ashaikh has joined #openstack-neutron | 18:57 | |
| *** dhellmann_ has left #openstack-neutron | 18:58 | |
| *** SushilKM__ has joined #openstack-neutron | 18:58 | |
| *** SushilKM has quit IRC | 18:58 | |
| *** networkstatic has joined #openstack-neutron | 18:58 | |
| terry_howe | anteaya: I'm thinking of mapping the neutron client commands like this: | 19:01 |
| terry_howe | https://wiki.openstack.org/wiki/OpenStackClient/Commands#DHCP_Agent | 19:01 |
| terry_howe | just not sure that entirely makes sense to implement it that way | 19:01 |
| anteaya | terry_howe: thanks just heading into a meeting, will look and respond | 19:01 |
| anteaya | thanks for the link | 19:01 |
| jlibosva | carl_baldwin: hi | 19:04 |
| jlibosva | carl_baldwin: have you seen the errors from http://paste.openstack.org/show/55182/ before? | 19:05 |
| *** salv-orlando has quit IRC | 19:05 | |
| enikanorov | terry_howe: agents host networks, hence cli command name | 19:05 |
| carl_baldwin | jlibosva: Can you give me a specific line number to look at. There is a lot in that paste. | 19:06 |
| enikanorov | terry_howe: while openstack client is not widely used it is particular project clients which driver usage patterns and names. It doesn't make sense to change cli command name in neutron client | 19:07 |
| *** ihrachyshka has quit IRC | 19:07 | |
| enikanorov | *which drive | 19:07 |
| jlibosva | carl_baldwin: oh, sure. I think it tries to create a port in ovs (line 165) but then on line 176 you can see that some ioctl calling failed | 19:08 |
| *** clev has quit IRC | 19:09 | |
| jlibosva | carl_baldwin: I have problems that dhcpdicovery from instances don't get to router nor dhcp namespace | 19:10 |
| *** nati_ueno has quit IRC | 19:10 | |
| *** nati_ueno has joined #openstack-neutron | 19:13 | |
| carl_baldwin | jlibosva: I see those errors in the log. This isn't something that I have seen before. Can you confirm that the devices do exist? Looks like this error is occurring with a number of tap devices. I also see the same error with some qr-* and qg-* devices. | 19:15 |
| jlibosva | carl_baldwin: I can see them in ovs-vsctl show | 19:17 |
| carl_baldwin | jlibosva: How about "ip addr show" ? | 19:18 |
| jlibosva | carl_baldwin: nope | 19:18 |
| jlibosva | carl_baldwin: just ovs bridges | 19:19 |
| *** ihrachyshka has joined #openstack-neutron | 19:20 | |
| carl_baldwin | I would except to see them in "ip addr show". I'll admit, though, that my knowledge of the ovs plugin is limited. I've only recently started to use it in devstack. | 19:21 |
| *** ashaikh has quit IRC | 19:23 | |
| carl_baldwin | jlibosva: Is it possible that we somehow got in to a network namespace somewhere along the way where we shouldn't be? | 19:23 |
| carl_baldwin | jlibosva: Forgive me if I missed some context along the way that would answer my questions. | 19:23 |
| jlibosva | carl_baldwin: you are right. I can see the one of tap devices in dhcp namespace. But I think that's correct. | 19:25 |
| *** rpodolyaka1 has joined #openstack-neutron | 19:25 | |
| jlibosva | the one, that reported No device found | 19:25 |
| *** jdev789 has joined #openstack-neutron | 19:26 | |
| yfried | mlavalle: I know it's a field in the conf that's supposed to allow parallel testing. and that all I know | 19:27 |
| terry_howe | enikanorov I'm not asking to change the neutron cli command name, I'm asking if what I proposed for OSC makes sense | 19:29 |
| carl_baldwin | jlibosva: Right, I think that it is correct that you can see that in the dhcp namespace. Also, my expectation that you can see it in the default namespace may not be correct with the ovs plugin or driver. Again, my experience with ovs is limited. | 19:30 |
| *** harlowja is now known as harlowja_away | 19:30 | |
| enikanorov | it may make sense, but it vice versa of what neutron cli has so it will create confusion for users i believe | 19:30 |
| jlibosva | carl_baldwin: also the logs might not be cause of the issue I run into...I think. It might be security groups as well, right? | 19:32 |
| jlibosva | carl_baldwin: packets can be thrown away before they enter ovs | 19:32 |
| carl_baldwin | jlibosva: You could be right. | 19:33 |
| *** dave_tucker_ has joined #openstack-neutron | 19:38 | |
| *** jdev789_1 has joined #openstack-neutron | 19:39 | |
| anteaya | mestery: any comments for the open discussion in -meeting regarding multi-node testing? | 19:41 |
| mestery | anteaya: Nothing this work, no. | 19:42 |
| anteaya | k | 19:43 |
| *** jlibosva has quit IRC | 19:45 | |
| *** jlibosva has joined #openstack-neutron | 19:46 | |
| roaet | what was the magic to fix the lbaas devstack failure thing? add lbass to the localrc where? | 19:47 |
| *** dave_tucker_ has quit IRC | 19:48 | |
| *** fouxm has joined #openstack-neutron | 19:49 | |
| rpodolyaka1 | anteaya: hey! did you want to discuss something? | 19:52 |
| *** fouxm has quit IRC | 19:53 | |
| anteaya | rpodolyaka1: you had a patch you needed some attention on | 19:53 |
| anteaya | can you link to it please? | 19:53 |
| *** fouxm has joined #openstack-neutron | 19:53 | |
| rpodolyaka1 | https://review.openstack.org/#/c/61677/1 | 19:53 |
| rpodolyaka1 | Yong's patch | 19:53 |
| anteaya | 7 minutes left in -infra meeting | 19:53 |
| anteaya | you are welcome to outline how we can help you | 19:54 |
| rpodolyaka1 | just review it :) | 19:54 |
| enikanorov | it's has -1 from Mark | 19:54 |
| rpodolyaka1 | I left some clarifying comments there | 19:55 |
| rpodolyaka1 | yep | 19:55 |
| rpodolyaka1 | so I might be wrong | 19:55 |
| enikanorov | i guess the bottom line is whether to allow to modify existing migration | 19:55 |
| rpodolyaka1 | that's I want more eyes to review it | 19:55 |
| rpodolyaka1 | *why | 19:55 |
| *** gdubreui has joined #openstack-neutron | 19:56 | |
| *** fouxm has quit IRC | 19:56 | |
| enikanorov | markmcclain: can we discuss https://review.openstack.org/#/c/61677 ? I think Roman has presented good analysis of different cases, and it justifies the approach taken in this patch | 19:56 |
| *** gdubreui has quit IRC | 19:56 | |
| *** gdubreui has joined #openstack-neutron | 19:56 | |
| *** networkstatic has quit IRC | 19:56 | |
| *** gdubreui has quit IRC | 19:56 | |
| *** SushilKM__ has quit IRC | 19:59 | |
| markmcclain | enikanorov: the fact still remains that we cannot revise released migrations | 19:59 |
| *** networkstatic has joined #openstack-neutron | 19:59 | |
| *** SushilKM__ has joined #openstack-neutron | 19:59 | |
| dkehn | //review.openstack.org/#/c/57600/) as one would expect, but after doing a fetch, with not merge issues its gone. I'm sure it | 19:59 |
| enikanorov | markmcclain: why? | 19:59 |
| markmcclain | it breaks continuity for those following trunk or stable/havana | 19:59 |
| enikanorov | right, but stable/havana doesn't work for ml2 | 20:00 |
| enikanorov | because the migration is broken | 20:00 |
| rpodolyaka1 | enikanorov: in fact, it does... | 20:00 |
| markmcclain | it does for those running create_all() | 20:00 |
| enikanorov | doesn't it make sense to fix it in stable too? | 20:00 |
| rpodolyaka1 | ML2 works only due to create_all() | 20:00 |
| enikanorov | ah, ok, I see | 20:01 |
| rpodolyaka1 | so why can't we fix this migration in-place? (it's a bad habit, of course, but can't this be an exception?) | 20:01 |
| rpodolyaka1 | ML2 would be broken, if create_all() failed | 20:01 |
| rpodolyaka1 | so for existing installations the fix will be no-op | 20:02 |
| rpodolyaka1 | because that's what we need | 20:02 |
| rpodolyaka1 | and we could fix new installations properly | 20:02 |
| rpodolyaka1 | even by a dirty hack like patching existing migrations | 20:02 |
| rpodolyaka1 | so my point is - users would have noticed their ML2 not working long time ago | 20:04 |
| rpodolyaka1 | both trunk chasers and stable/havana ones | 20:04 |
| rpodolyaka1 | neutron just woudn't be usable | 20:04 |
| rpodolyaka1 | as this migration is almost one year old, we have ML2 merged for a few months now | 20:04 |
| *** gdubreui has joined #openstack-neutron | 20:08 | |
| anteaya | dkehn: is you above comment a ghost from the past? | 20:09 |
| *** ashaikh has joined #openstack-neutron | 20:09 | |
| dkehn | anteaya: hmm from the past | 20:10 |
| anteaya | rpodolyaka1: the problem for advocating for the breaking of some rules to shoehorn a patch in, is that this approach has kind of been the status quo | 20:10 |
| dkehn | anteaya: past meaning an hour or so ago? | 20:10 |
| anteaya | a situation which I am trying hard to change and fortunately am seeing good results | 20:10 |
| anteaya | dkehn: yeah I think that is the time frame when that comment was applicaple | 20:11 |
| dkehn | yepper | 20:11 |
| *** vkozhukalov has quit IRC | 20:11 | |
| anteaya | rpodolyaka1: so personally I can't advocate for a relaxing of the rules for one exception | 20:11 |
| anteaya | important as that exception might be | 20:11 |
| anteaya | rpodolyaka1: is there any other option on the table that will help you achieve your goals? | 20:12 |
| rpodolyaka1 | anteaya: no, thank you! | 20:12 |
| anteaya | okay thanks, if you want to talk again please do ping | 20:12 |
| rpodolyaka1 | cool! | 20:13 |
| anteaya | :D | 20:13 |
| *** harlowja_away is now known as harlowja | 20:14 | |
| *** clev has joined #openstack-neutron | 20:15 | |
| enikanorov | the exception could be preferable because the patch does the job in a better way then the fix which floows the rules, that's it | 20:15 |
| enikanorov | *follows | 20:15 |
| *** salv-orlando has joined #openstack-neutron | 20:16 | |
| *** alexpilotti has joined #openstack-neutron | 20:17 | |
| *** mtaylor is now known as mordred | 20:20 | |
| *** suresh12 has quit IRC | 20:22 | |
| *** jdev789 has quit IRC | 20:23 | |
| *** jdev789_1 has quit IRC | 20:23 | |
| yfried | mlavalle: can you please show me how to use tenant isolation instead of main/alt tenants? | 20:25 |
| mlavalle | yfried: sure, hang on a minute | 20:26 |
| anteaya | enikanorov: if you can make a convincing argument, that perhaps includes addressing a need to change the rules then let me know how I can help you set up that discussion | 20:27 |
| sc68cal | markmcclain: Is there any current use for the external property of a network, for v6? | 20:27 |
| markmcclain | sc68cal: possibly yes | 20:28 |
| sc68cal | wondering if a combo of enable_dhcp=false on a v6 subnet, and external=true on the network could be used to mean that a provider router is announcing RA's | 20:28 |
| markmcclain | I'd rather have an explicit setting on the subnet | 20:28 |
| markmcclain | otherwise the behavior seems a bit too subtle | 20:28 |
| sc68cal | ok - perhaps subnet_modes = ['provider-slaac'] ? | 20:29 |
| ijw | Not external, but provider:… presumablyt | 20:29 |
| ijw | external is slightly different | 20:29 |
| enikanorov | anteaya: the arguments were presented in the review. rpodolyaka1 did the analysis of different cases (including 'trunk chasers'). The alternative solution is possible, but it will require migration rearrangement as far as i understand. which is also not desirable | 20:29 |
| sc68cal | ijw: Yeah the only real use I know for external is floating ips | 20:30 |
| ijw | I think for v4 this works by disabling DHCP and setting the DHCP server address, and I was trying to come up with something that worked the same way | 20:30 |
| *** jlibosva has quit IRC | 20:30 | |
| mlavalle | yfried: can you open a copuple of modules? let's look at the code. tempest.scenario.manager and tempest.common.isolated_creds | 20:30 |
| ijw | sc68cal: Yeah, that's basically it, I think, and it's not really v6 relevant or appropriate because for v6 routing you need a nexthop, not an address | 20:31 |
| ijw | (when you don't have NAT) | 20:31 |
| ijw | On the IPv6 front, people, look at https://docs.google.com/document/d/1rOBOOu_OwixMStm6XJOb5PKkJA6eFbL_XCE7wlTfaPY/ - it's intended as where we could take it, not a 'we must do all of this by Icehouse', and in particular I think we'd skip the internal DHCP server stuff | 20:32 |
| sc68cal | Just soo close I can taste it - instances get v6 addresses and work correctly, but openstack is clueless about the prefix being advertised northbound | 20:33 |
| *** jecarey has joined #openstack-neutron | 20:33 | |
| sc68cal | because the moment you add a v6 subnet in openstack the whole thing comes crashing down :) | 20:33 |
| anteaya | enikanorov: okay, thanks for the summary | 20:33 |
| *** SumitNaiksatam has quit IRC | 20:34 | |
| sc68cal | ijw: do you want to share that google doc on the ML? | 20:35 |
| sc68cal | we'll def talk about it in the subteam meeting | 20:35 |
| *** suresh12 has joined #openstack-neutron | 20:35 | |
| geekinutah | carl_baldwin: you around? | 20:36 |
| carl_baldwin | geekinutah: yes | 20:36 |
| ijw | Yeah, can do. I was just warming up with it, really | 20:36 |
| geekinutah | I was just looking at https://review.openstack.org/#/c/58017/ | 20:36 |
| yfried | mlavalle: can we talk in 15min? | 20:37 |
| geekinutah | I'm wondering, the whole ip recycling behavior doesn't seem appropriate for an API | 20:37 |
| mlavalle | yfried: yeah, ping me | 20:37 |
| geekinutah | have you ever considered moving it out and having it be triggered on a more periodic basis? | 20:37 |
| yfried | mlavalle: are you on US time? | 20:37 |
| mlavalle | yfried: I am in us central time. how about you? | 20:38 |
| yfried | Israel. | 20:38 |
| geekinutah | so the API could send a notification or something to another service or thread rather than trying to deal with it in it's runtime | 20:39 |
| mlavalle | yfried: ok, i'll be here for a good while | 20:39 |
| markmcclain | geekinutah: IP reclamation is not easy | 20:39 |
| markmcclain | with large ranges we can delay | 20:40 |
| markmcclain | with smaller ranges collection would need to occur more synchronously | 20:40 |
| geekinutah | yeah, I'm just trying to think around how to do it without serializing other API stuff | 20:40 |
| carl_baldwin | My thinking was that if an IP has just been released to a pool then one should be immediately available. | 20:42 |
| yfried | mlavalle: im back | 20:43 |
| geekinutah | oh wait, your patch will only serialize on subnets that need rebuilding | 20:43 |
| geekinutah | I misread it first time around | 20:43 |
| yfried | mlavalle: both modules are open | 20:44 |
| carl_baldwin | geekinutah: Right, and rebuilding will not come around very often unless the subnet is nearly full. | 20:44 |
| mlavalle | yfried: the big idea is to provide a set of user credentials for each test. That way, each test will run in it's own tenant and won't interfere with others. That is important if we want to run tests in parallel in the gate | 20:44 |
| * anteaya is afk for a bit | 20:45 | |
| carl_baldwin | ... or more precisely "unless the pool is nearly exhausted" | 20:45 |
| mlavalle | yfried: right now, tenant isolation is not enabled in the gate for neutron. But we hope it will be soon, to enable parellel testing | 20:45 |
| * carl_baldwin is running out for a bike ride. Beautiful weather here in Colorado. | 20:46 | |
| mlavalle | the isolated_creds module has methods to generate credentials for a primary user, an admin user and a tenant user | 20:46 |
| *** rpodolyaka1 has quit IRC | 20:46 | |
| mlavalle | I menat alt user (last one) | 20:46 |
| yfried | mlavalle: ok | 20:47 |
| mlavalle | yfried: if you look at class OfficialClientTest in manager.py, there is a method credentials. It generate credentials for the primary user, i.e. demo | 20:48 |
| mlavalle | yfried: this happens if tenant isolation is enabled. Otherwise, it just goes to the config file for the demo credentials. | 20:49 |
| mlavalle | take a look | 20:49 |
| beagles | mlavalle: quick question.. how easily is that stuff used for a test that itself uses multiple tenants? is it pretty much obvious once you take a peek? | 20:50 |
| mlavalle | beagles: it's pretty easy. you just need to get used that your demo user, alt user and admin user are coming from a class, instead form the config file | 20:51 |
| beagles | mlavalle: cool | 20:51 |
| mlavalle | yfried: are with me? | 20:52 |
| yfried | yeah | 20:52 |
| yfried | looking at "get_primary_creds" | 20:52 |
| yfried | mlavalle: ok | 20:52 |
| mlavalle | yfried: now look at NetworkScenarioTest(OfficialClientTest), which inherits from OfficialClientTest | 20:53 |
| openstackgerrit | Sean M. Collins proposed a change to openstack/neutron: Create a new attribute for subnets, to store v6 dhcp options https://review.openstack.org/52983 | 20:53 |
| yfried | mlavalle: ? | 20:54 |
| mlavalle | yfried: you will see in the setUpClass method that the tenant id is initialized either from isolated credentials or from the config file | 20:54 |
| mlavalle | yried: for the demo user | 20:54 |
| mlavalle | yfried^^^ | 20:54 |
| yfried | ok | 20:55 |
| mlavalle | yfried: in isolated)ceds thera are parallel methods for the alt user: generate credentials, get the tenant id, etc | 20:55 |
| mlavalle | yfried: look at get-alt_creds for example | 20:56 |
| mlavalle | get_alt_creds^^^^ | 20:56 |
| *** fouxm has joined #openstack-neutron | 20:57 | |
| *** SushilKM__ has quit IRC | 20:58 | |
| mlavalle | yfried: so what you need to do in your test or somewhere in the classes in manager.py is to check if tenant isolation is enabled and if it is, use isolated_creds to generate the alt_user. Otherwise, you go to the config file | 20:59 |
| yfried | mlavalle: why aren't there wrapper methods that return creds based on "isolation" flag - ie - you either get your isolated creds from this method or you get them form file | 20:59 |
| mlavalle | yfried: well, i didn't write the credentials isolation piece. But I suspect that we wanted to maintain backwards compatibility | 21:00 |
| ijw | sc68cal: mail sent | 21:00 |
| *** fouxm has quit IRC | 21:01 | |
| sc68cal | ijw: thanks | 21:01 |
| *** fouxm has joined #openstack-neutron | 21:01 | |
| ijw | If you want to use your patch to propose the attrs and values that's fine, I can just update the doc to match | 21:01 |
| yfried | mlavalle: I'm asking if there might be a reason for me not to create a standalone patch with this method | 21:02 |
| yfried | mlavalle: it seems pretty simple and I'm wondering why it wasn't done already | 21:02 |
| mlavalle | yfried: there is no reason. If you have a better way of doing this, by all means, let's do it | 21:02 |
| *** jistr has quit IRC | 21:02 | |
| *** geekinutah has left #openstack-neutron | 21:02 | |
| mtreinish | yfried: it's actually an oversight I think, we probably want to store the creds in a object level variable for easy reference. In the api tests this is stored in the clients manager object, which it looks like I didn't do when I rewrote the scenario tests to use isolated creds | 21:03 |
| yfried | mlavalle: all I'm saying is that what you are suggesting should be done regardless of my patch | 21:03 |
| mtreinish | yfried: see http://git.openstack.org/cgit/openstack/tempest/tree/tempest/clients.py#n180 for how it is stored in the api tests | 21:04 |
| mtreinish | yfried: but helper methods would defintely be a welcome addition for the scenario tests | 21:04 |
| yfried | I'll do it and then rebase the patch on it, but since it's not a problem in the patch, can you please recall your -1 so it is merged? | 21:05 |
| *** fouxm has quit IRC | 21:06 | |
| yfried | mlavalle: ^ | 21:06 |
| yfried | mlavalle: I don't want to block this patch because of a feature that's not used in other network tests. | 21:07 |
| *** dims has quit IRC | 21:08 | |
| mlavalle | yfried: fair enough. But keep in mind that all network tests should take this into consideration going forward | 21:08 |
| yfried | mlavalle: np | 21:08 |
| *** dims has joined #openstack-neutron | 21:09 | |
| mlavalle | yfried: in fact, th other scenario test you are working on, basic net ops, was modified by me about a couple of months ago to work with the isolation code | 21:09 |
| *** yamahata has joined #openstack-neutron | 21:10 | |
| yfried | mlavalle: which part | 21:11 |
| mlavalle | yfried: here, i removed some lines to let the parent class take care of managing credentials and tenant: https://review.openstack.org/#/c/45578/12/tempest/scenario/test_network_basic_ops.py | 21:14 |
| openstackgerrit | A change was merged to openstack/neutron: Imported Translations from Transifex https://review.openstack.org/62561 | 21:17 |
| *** nati_ueno has quit IRC | 21:20 | |
| *** nati_ueno has joined #openstack-neutron | 21:28 | |
| *** 17SADQLUP has joined #openstack-neutron | 21:31 | |
| yfried | nati_ueno: do you think you could test your patch against mine? | 21:32 |
| nati_ueno | yfried: hi | 21:32 |
| nati_ueno | yfried: which patch? | 21:32 |
| yfried | nati_ueno: https://review.openstack.org/#/c/21946/ tested against https://review.openstack.org/#/c/62702/ | 21:33 |
| nati_ueno | yfried: Sure the patch need also https://review.openstack.org/#/c/44596/ | 21:34 |
| nati_ueno | I can definitely do that, however we have server migration in office in this week, so may be I can't do it in today | 21:35 |
| openstackgerrit | Sylvain Afchain proposed a change to openstack/neutron: Add param and iptables rules to protect dnsmasq ports https://review.openstack.org/61994 | 21:36 |
| *** 17SADQLUP has quit IRC | 21:38 | |
| yfried | nati_ueno: np. so let me know when you can and if you need help. You'll probably have some rebase issues by next week | 21:39 |
| nati_ueno | yfried: sure Thank you for your sharing test patch | 21:39 |
| *** briancline has quit IRC | 21:39 | |
| yfried | nati_ueno: thank you for fixing it so my patch will finally be merged | 21:40 |
| nati_ueno | yfried: This tempest patch is important, otherwise we could break sec-group function again | 21:40 |
| *** skromnica has joined #openstack-neutron | 21:41 | |
| yfried | nati_ueno: yeah. tell me about it. I'm just amazed no one saw this before me... | 21:41 |
| nati_ueno | yfried: Thank you for your contribution on this :). | 21:42 |
| yfried | nati_ueno: anyway - tnx for the work. I'm guessing it wasn't a 1line fix | 21:42 |
| nati_ueno | your welcome | 21:42 |
| yfried | asking again here (redirected from infra): | 21:43 |
| yfried | Hi guys, I'm looking to get the external connectivity (ie - ping 8.8.8.8) on the devstack-neutron gate so this patch can work: https://review.openstack.org/#/c/55146/ | 21:43 |
| yfried | can any one help me? | 21:43 |
| *** SushilKM has joined #openstack-neutron | 21:43 | |
| yfried | or to put it another way - are we sure the gate doesn't have external connection at the moment? | 21:44 |
| nati_ueno | yfried: I guess we should setup NAT rule in the VM if we wan't to do this | 21:45 |
| nati_ueno | or correct GW settings in the VM | 21:45 |
| yfried | nati_ueno: and who can do that? | 21:46 |
| *** skromnica has quit IRC | 21:46 | |
| nati_ueno | I can do that, if it is nexe week | 21:46 |
| nati_ueno | s/nexe/next/ | 21:46 |
| nati_ueno | We should update devstack | 21:46 |
| yfried | nati_ueno: can you point me to anyone else? I would really like it to happen earlier. this patch have been stuck in review limbo for a while now. | 21:47 |
| yfried | nati_ueno: but tnx for the offer. I will nag you next week if I can't find other volunteers | 21:48 |
| nati_ueno | yfried: how about asking it in the openstack-dev ? | 21:48 |
| yfried | nati_ueno: will do | 21:48 |
| nati_ueno | The change should be easy, just setup ip masquerade between floating ip range and external network connection in the VM | 21:50 |
| nati_ueno | it should be one or two lines in devstack | 21:50 |
| *** briancline has joined #openstack-neutron | 21:52 | |
| *** SushilKM has quit IRC | 21:56 | |
| *** clev has quit IRC | 21:58 | |
| *** nati_uen_ has joined #openstack-neutron | 22:01 | |
| *** nati_uen_ has quit IRC | 22:03 | |
| *** SumitNaiksatam has joined #openstack-neutron | 22:03 | |
| *** nati_ueno has quit IRC | 22:03 | |
| *** nati_uen_ has joined #openstack-neutron | 22:03 | |
| *** jroovers has joined #openstack-neutron | 22:06 | |
| *** ian12 has joined #openstack-neutron | 22:08 | |
| *** markmcclain has quit IRC | 22:10 | |
| *** klindgren_ has joined #openstack-neutron | 22:13 | |
| klindgren_ | Is it expected that when using the ml2 plugin neutron agent-list results in no results being returned, but under the neutron_ml2 database the agents table is populated with agents that are checking in? This is under devstack on Ubuntu | 22:14 |
| *** pcm__ has joined #openstack-neutron | 22:17 | |
| *** ian12 has quit IRC | 22:18 | |
| *** dave_tucker has joined #openstack-neutron | 22:19 | |
| *** pcm has quit IRC | 22:20 | |
| *** pcm__ has quit IRC | 22:26 | |
| anteaya | terry_howe: sorry for my delayed response | 22:43 |
| anteaya | I think the openstackclient is a great project | 22:43 |
| anteaya | it really makes things easier for people to get a handle on openstack as a whole, rather than just one project | 22:44 |
| anteaya | given the command format for the other projects you have on the wiki, I don't see how else you could format the commands for DHCP agent | 22:44 |
| anteaya | if there are other ways that fit the format used by openstackclient currrently, I am interested to hear them | 22:45 |
| openstackgerrit | Sukhdev Kapur proposed a change to openstack/neutron: This is just a test for Gerrit for third party testing- plz ignore https://review.openstack.org/62771 | 22:45 |
| anteaya | now how the relationship is conveyed by the command, between network and agent, we would need more opinions on that | 22:46 |
| openstackgerrit | Sukhdev Kapur proposed a change to openstack/neutron: This is just a test for Gerrit for third party testing- plz ignore https://review.openstack.org/62771 | 22:49 |
| *** dave_tucker has quit IRC | 22:51 | |
| *** dave_tucker has joined #openstack-neutron | 22:52 | |
| openstackgerrit | mouad benchchaoui proposed a change to openstack/neutron: Make the metadata namespace proxy transparent https://review.openstack.org/28137 | 22:56 |
| *** yfried has quit IRC | 22:56 | |
| openstackgerrit | Sylvain Afchain proposed a change to openstack/neutron: L3 Agent can handle many external networks. https://review.openstack.org/59359 | 22:57 |
| *** dave_tucker has quit IRC | 22:59 | |
| *** dave_tucker has joined #openstack-neutron | 22:59 | |
| *** dims has quit IRC | 23:01 | |
| ijw | argh | 23:01 |
| *** jpich has quit IRC | 23:01 | |
| *** harlowja is now known as harlowja_away | 23:02 | |
| *** harlowja_away is now known as harlowja | 23:05 | |
| anteaya | arosen this bug just jumped: https://bugs.launchpad.net/neutron/+bug/1250168 | 23:05 |
| anteaya | enikanorov: this bug has jumped in frequency as well https://bugs.launchpad.net/nova/+bug/1210483 | 23:06 |
| anteaya | and markmcclain this bug continues to cause problems: https://bugs.launchpad.net/tempest/+bug/1253896 | 23:09 |
| *** aymenfrikha has joined #openstack-neutron | 23:15 | |
| *** dims has joined #openstack-neutron | 23:16 | |
| *** yamahata has quit IRC | 23:20 | |
| anteaya | currently we have a 57% pass chance in the gate across all of openstack | 23:21 |
| *** djbkd has joined #openstack-neutron | 23:30 | |
| *** jroovers has quit IRC | 23:36 | |
| *** SumitNaiksatam has quit IRC | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!