Monday, 2015-11-30

« Saturday, 2015-11-28 Index Tuesday, 2015-12-01 »
*** openstack has joined #openstack-neutron15:45
*** leifmadsen_ has joined #openstack-neutron15:46
*** Sukhdev_ has joined #openstack-neutron15:46
Sukhdev_kevinbenton_ : Good Morning15:47
Sukhdev_kevinbenton_ : Will you be able to join us today at 8AM for Ironic-neutron discussion?15:48
*** lezbar has joined #openstack-neutron15:48
*** salv-orlando has joined #openstack-neutron15:49
*** jprovazn has joined #openstack-neutron15:49
*** scheuran has quit IRC15:49
*** dims has joined #openstack-neutron15:49
*** shaohe_feng has joined #openstack-neutron15:50
*** martinus__ has quit IRC15:50
*** garyk has quit IRC15:52
*** amotoki has quit IRC15:53
openstackgerritHirofumi Ichihara proposed openstack/neutron: Add functional test for availability_zone support  https://review.openstack.org/21097715:53
*** agireud has joined #openstack-neutron15:53
*** martinus__ has joined #openstack-neutron15:53
*** garyk has joined #openstack-neutron15:54
*** leifmadsen has quit IRC15:55
*** rook has quit IRC15:55
*** leifmadsen_ is now known as leifmadsen15:56
regXboiamuller: ping - am I allowed to mock a configuration parameter in a unit test or will that create bad karma?15:56
*** ajmiller has joined #openstack-neutron15:56
*** jistr has quit IRC15:57
*** krypto has quit IRC15:58
*** pbourke has quit IRC15:58
amullerregXboi: there's lots of unit tests that use set_override15:58
ihrachysregXboi: set_override?.. but note that you should not generally rely on configuration options that are not for neutron in tests, neither in code.15:58
*** pbourke has joined #openstack-neutron15:58
*** jasondotstar has joined #openstack-neutron15:58
openstackgerritShih-Hao Li proposed openstack/neutron: (WIP) Make MetadataProxyHandler configurable  https://review.openstack.org/24787415:59
regXboiamuller: thx15:59
*** krypto has joined #openstack-neutron15:59
*** edgar__ has joined #openstack-neutron15:59
regXboiihrachys: understood - this is a option that is part of the patch set15:59
*** sdague has joined #openstack-neutron15:59
*** edgar__ has quit IRC16:00
*** yfried_ has quit IRC16:01
openstackgerritvenkatamahesh proposed openstack/neutron-vpnaas: Change the order of import files to prevent pep8 errors  https://review.openstack.org/25145216:02
*** edgar__ has joined #openstack-neutron16:03
*** rook has joined #openstack-neutron16:03
*** yfried_ has joined #openstack-neutron16:03
*** haleyb has joined #openstack-neutron16:03
*** maht has joined #openstack-neutron16:04
*** banix has joined #openstack-neutron16:04
*** edgar__ has quit IRC16:04
*** edgar__ has joined #openstack-neutron16:04
*** korzen_ has quit IRC16:05
*** zhangjn has quit IRC16:05
*** korzen_ has joined #openstack-neutron16:05
*** mlavalle has joined #openstack-neutron16:05
*** brad_behle has joined #openstack-neutron16:06
*** kevinbenton_ has quit IRC16:07
*** mitchjameson has joined #openstack-neutron16:07
korzen_Hi ihrachys, was there an upgrade meeting today>16:07
*** kevinbenton has joined #openstack-neutron16:07
ihrachyskorzen_: nope. basically no one showed up, not sure whether it's turkey hangover or irc issues (rossella dropped an email that she had connectivity issues)16:08
korzen_ihrachys yes, I had the connectivity problem to join the IRC16:09
ihrachyskorzen_: good news is: you haven't missed anything16:09
korzen_ihrachys :)16:10
*** wolverineav has quit IRC16:10
*** yfried_ has quit IRC16:10
ihrachyskorzen_: as per rossella, she is working on PoC for ports, and she already has something.16:11
*** Marga_ has joined #openstack-neutron16:11
openstackgerritTang Chen proposed openstack/python-neutronclient: Trivial: Fix a typo in class ListCommand  https://review.openstack.org/25105516:12
ihrachyskorzen_: as for general OVO strategy, armax told me prev week we are good to go with one or two objects in M, no RFE needed, devref is recommended16:12
*** irenab has joined #openstack-neutron16:12
*** anshul has quit IRC16:12
*** crose has joined #openstack-neutron16:12
*** jdandrea has quit IRC16:12
*** Marga_ has quit IRC16:13
*** EinstCrazy has quit IRC16:13
*** Marga_ has joined #openstack-neutron16:14
*** rook has quit IRC16:14
korzen_ihrachys thanks for update, from my side I researched the network object but no patch yet16:14
*** rook has joined #openstack-neutron16:15
*** arif-ali_ has joined #openstack-neutron16:15
*** arif-ali has quit IRC16:15
korzen_ihrachys, are we allowed to push only one or two objects? should we choose the proper one?16:15
*** arif-ali_ is now known as arif-ali16:15
*** openstackstatus has joined #openstack-neutron16:16
*** ChanServ sets mode: +v openstackstatus16:16
ihrachyskorzen_: network and port would be those. I believe we'll have some more helper objects to serve those, but for core resources, those are two16:16
*** mohankumar has joined #openstack-neutron16:16
*** dane_leblanc has quit IRC16:18
*** jdandrea has joined #openstack-neutron16:20
*** apuimedo has quit IRC16:20
*** jlanoux has joined #openstack-neutron16:20
ihrachysmestery: how do we move forward Liberty release? should we ping someone?16:22
mesteryihrachys: I pushed the latest patch, I'll ping dhellmann and get it done.16:22
mesteryihrachys: Pinged in #openstack-release channel16:23
*** nlahouti has joined #openstack-neutron16:24
rossella_sihrachys, I was able to reconnect now, sorry for missing the meeting, I've read that nobody showed up16:25
ihrachysrossella_s: yeah. I suspect there were global issues.16:25
rossella_sihrachys, probably anyway there's nothing on freenode twitter16:27
haleybthere was a DDoS attack i believe16:27
haleybs/another16:27
mlavallerossella_s: you also had problems connecting to freenode?16:27
rossella_smlavalle, yes16:27
rossella_shaleyb, yep there was one on friday16:28
*** nlahouti has quit IRC16:28
mlavallerossella_s: thanks! I thought it was something wrong with my laptop :-)16:28
rossella_smlavalle, at first me too :P16:29
*** shaohe_feng has quit IRC16:29
*** baoli has quit IRC16:29
*** gampel has quit IRC16:31
*** baoli has joined #openstack-neutron16:31
*** neiljerram has joined #openstack-neutron16:33
*** mwagner_lap has quit IRC16:35
*** mwagner_lap has joined #openstack-neutron16:35
*** nlahouti has joined #openstack-neutron16:36
*** rook has quit IRC16:37
*** vthapar has quit IRC16:40
*** chlong has joined #openstack-neutron16:40
*** achanda_ has quit IRC16:41
regXboiamuller: I have another unit test q for you16:44
regXboiamuller: how does one go about checking if a timer thread is properly doing its thing or is that considered to "just work" ?16:44
*** mgoddard_ has joined #openstack-neutron16:44
regXboiamuller: I ask because writing a test with a time.sleep just feels wrong16:45
amullerregXboi: I need to see the patch16:46
regXboiamuller: ok, let me push it up16:46
*** LukeHinds has quit IRC16:46
*** abregman has quit IRC16:46
*** carl_baldwin_ is now known as carl_baldwin16:47
*** baoli has quit IRC16:48
*** mgoddard__ has quit IRC16:48
*** korzen_ has quit IRC16:50
*** baoli has joined #openstack-neutron16:50
*** oshvartz has quit IRC16:51
*** aranjan has joined #openstack-neutron16:53
*** armax has joined #openstack-neutron16:54
*** virtuallynathan_ has quit IRC16:55
*** edgar__ has quit IRC16:56
openstackgerritRyan Moats proposed openstack/neutron: WIP: Collector Proof of Concept DO NOT MERGE!!!  https://review.openstack.org/21347416:56
openstackgerritRyan Moats proposed openstack/neutron: Add instrumentation devref, Part I  https://review.openstack.org/20428216:56
regXboiamuller: ^^^^^^ (first patch)16:56
*** emagana has joined #openstack-neutron16:57
*** aranjan has quit IRC16:57
amullerregXboi: eek, lots of code, can I get a pointer?16:57
*** ahduyck has quit IRC16:57
*** aranjan has joined #openstack-neutron16:57
openstackgerritHong Hui Xiao proposed openstack/neutron: Update related router when subnetpool change scope  https://review.openstack.org/25147016:58
regXboiamuller: I'm looking at the *only* test in tests/unit/agent/collection/test_agent.py16:58
regXboiamuller: it's trying to verify that the reporting thread works correctly, so it sets the interval to one second and then waits that interval16:58
regXboiit's really hokey16:58
*** achanda has joined #openstack-neutron16:59
regXboibut it runs with a minimum of mocks16:59
amullerregXboi: ok lemme finish a meeting I'll take a closer look16:59
regXboiamuller: take your time - it's a WiP patch with some outstanding todos on it17:00
*** aduyck has joined #openstack-neutron17:00
regXboiamuller: and I have a follow on patch to work on now17:00
*** coolsvap has quit IRC17:01
*** ildikov has quit IRC17:01
*** mgoddard_ has quit IRC17:02
*** mitchjameson has quit IRC17:02
*** ygbo has quit IRC17:02
*** mgoddard has joined #openstack-neutron17:02
*** rook has joined #openstack-neutron17:03
*** rdo_ has quit IRC17:03
*** maht has quit IRC17:05
*** coolsvap has joined #openstack-neutron17:05
*** Sukhdev_ has quit IRC17:06
*** krtaylor has joined #openstack-neutron17:06
*** maht has joined #openstack-neutron17:08
*** salv-orl_ has joined #openstack-neutron17:08
*** manjeets has joined #openstack-neutron17:09
*** minwang2 has joined #openstack-neutron17:10
*** salv-orlando has quit IRC17:11
*** Sukhdev_ has joined #openstack-neutron17:12
*** Sukhdev_ has quit IRC17:12
*** slaweq has joined #openstack-neutron17:12
*** maht has quit IRC17:12
*** apuimedo has joined #openstack-neutron17:13
*** EinstCrazy has joined #openstack-neutron17:14
openstackgerritvenkatamahesh proposed openstack/neutron-fwaas: Change the order of import files to prevent Jenkins pep8 errors  https://review.openstack.org/25142517:14
*** boris-42_ has quit IRC17:15
*** Swami has joined #openstack-neutron17:17
*** tmorin has quit IRC17:18
openstackgerritIhar Hrachyshka proposed openstack/neutron: Deprecate _ builtin translation function  https://review.openstack.org/25094617:19
*** gsagie_ has joined #openstack-neutron17:20
*** jlibosva has quit IRC17:21
*** EinstCrazy has quit IRC17:21
*** signed8bit is now known as signed8bit_ZZZzz17:21
*** jobewan has joined #openstack-neutron17:22
*** Piet has joined #openstack-neutron17:22
*** fer_ has quit IRC17:22
*** dconde has joined #openstack-neutron17:23
*** crose has quit IRC17:24
*** signed8bit_ZZZzz is now known as signed8bit17:27
*** ihrachys has quit IRC17:28
*** gsagie_ has quit IRC17:29
openstackgerritRossella Sblendido proposed openstack/neutron: Add a script to create review dashboard for a milestone  https://review.openstack.org/24864517:29
*** rdo has joined #openstack-neutron17:29
*** daneyon_ has joined #openstack-neutron17:32
*** daneyon__ has joined #openstack-neutron17:35
*** SurajD has quit IRC17:35
*** daneyon has quit IRC17:36
*** jlanoux has quit IRC17:36
*** daneyon_ has quit IRC17:38
*** rook has quit IRC17:38
*** rook has joined #openstack-neutron17:39
*** mihero has quit IRC17:41
*** krypto has quit IRC17:42
*** krypto has joined #openstack-neutron17:42
*** ildikov has joined #openstack-neutron17:42
*** paul-carlton1 has joined #openstack-neutron17:43
mordredmestery: so, I promised I'd complain at you on monday - perhaps I should just do it here17:44
*** edmondsw has joined #openstack-neutron17:44
*** Marga_ has quit IRC17:45
mordredhttps://wiki.openstack.org/wiki/Neutron/sharing-model-for-external-networks <-- we've run in to the problem in the wild that is encompassed in this wiki page17:45
mordredthe problem we have in infra, as users of multiple clouds - is being able to answer the question "does this network route packets externally"17:46
mordredadditionally we want to be able to answer the question "can I boot a vm on this network or can I only connect a router to it"17:47
mordredbecause with the answers to those two questions, it is possible to know how to interact with the networks the cloud has provided to su17:47
mordredus17:47
*** nlahouti has quit IRC17:47
*** javeriak has quit IRC17:48
openstackgerritSwaminathan Vasudevan proposed openstack/neutron: Change check_ports_exist_on_l3agent to pass the subnet_ids  https://review.openstack.org/24184317:49
amullermordred: I think that the answer to 1 is looking at the router:external flag17:49
openstackgerritSwaminathan Vasudevan proposed openstack/neutron: Check for subnets with dhcp enabled only on dvr_snat agent  https://review.openstack.org/25007517:50
mordredamuller: right. except that does not work if the cloud provider has made a per-tenant external network for me17:50
mordredamuller: because router:external implies shared17:50
Sam-I-Amrouter:external doesnt imply shared17:50
amullermordred: no it doesn't17:50
amullera network can be router:external = True, shared = False17:50
mordredit sure did make the network visible to other tenants when the updated the flag17:50
mordredat internap17:50
*** markus_z has quit IRC17:50
amullerok so17:50
Sam-I-Amit does, but only as a place to attach routers17:50
amullerif a network has router:external = True, all tenants can see it in 'neutron net-list'17:51
mordredright. but they don't want other tenants to be able to place routers in my network17:51
mordredthey make an external network per tenant at internap17:51
*** javeriak has joined #openstack-neutron17:51
amullermordred: yeah they should be able to play around with policy.json to get that17:51
amullermordred: why does it change the equation though?17:51
mordredso "if a network has router:external = True, all tenants can see it in 'neutron net-list'" - that implies shared17:51
mordredin that it's now not private to me17:52
amullermordred: the shared flag means something different17:52
openstackgerritSwaminathan Vasudevan proposed openstack/neutron: Optimize get_ports filter in check_ports_exist_on_l3_agent  https://review.openstack.org/25015017:52
mordredsorry. let me restate17:52
mordredI've been double-using a word17:52
amullerif a network has router:external = True but shared=False, that means everyone (By default) can see it, but you could not connect VMs to it17:52
mordredimplies non-private17:52
amulleronly routers17:52
mordredshared is totally a different thing17:52
amullerright17:52
mordredwhat we're looking for is a way that someone can commuicate the metadata to me, as a user, about what capabilities the network currently has17:53
mordredwhich is that it routes packets externally but is only visible to me17:53
Sam-I-Amalso if you want to connect vms directly to any public/external network, you need some additional bridge/interface mapping config17:53
mordredyes17:53
Sam-I-Amregardless of --shared17:53
mordredthey have that17:53
amullermordred: I don't think there's a way of knowing if the provider played around with policy.json to know if only you can see a network17:54
*** achanda has quit IRC17:54
mordredthat's actually not the thing I want to know - I don't care about that so much17:54
amullermordred: I'm not following then17:54
mordredwhat I want is a way for the provider to tell me that the netwrk routes packets externally17:54
mordredbut the provider in question cannot set the flag that I can see that tells me that17:55
mordredbecause doing so also makes that network visible to other tenants17:55
amullerrouter:external flag is what allows routers to be connected to a network17:55
mordredthere is no way for them to communicate to me that the network has external routing capabilities17:55
mordredok17:55
mordredso that's good to know17:55
amullerI mean the router's external leg of course17:55
mordredI think my request is for an informational metadata flag17:55
Sam-I-Amwhat happens if you have a network with router:external owned by a non-admin tenant?17:55
mordredthat says "this network routes packets off of this cloud"17:56
*** salv-orl_ has quit IRC17:56
mesterymordred: Catching up, looks like amuller and Sam-I-Am are giving you lots of juicy info here17:56
mordredthey are - I hope I'm communicating the problem space we have generally17:57
mordredwhich is that we consume resources from clouds that have all made different deployment choices17:57
*** salv-orlando has joined #openstack-neutron17:57
amullermordred: I'm confused, internap are giving you an external network that has router:external = False?17:57
mordredyes17:57
amullerI didn't know that would work...17:57
mordredbecause it is a network that is just for me17:57
Sam-I-Ammordred: summary - if you set router:external, the network shows up for all tenants, and they can connect routers to that network even if they dont own it17:57
mesteryYikes, I wonder why they are doing that?17:57
mordredit does17:57
amullerhmm17:57
amullerhow =D17:57
Sam-I-Amand this is a Bad Thing because only one external network should appear for each tenant17:58
*** nlahouti has joined #openstack-neutron17:58
mordredya. which is why they do not set that flag17:58
Sam-I-Amwho actually owns the external networks?17:58
mordredbut then I'm left not being able to differntiate between  my public and private networks17:58
Sam-I-Ams/who/what project17:59
mordredlooking17:59
amullernow I'm curious and playing around with a devstack VM17:59
mordredhttp://paste.openstack.org/show/48040817:59
mordredthere is the public network I get from them17:59
*** madhu_ak has joined #openstack-neutron17:59
*** paul-carlton1 has quit IRC18:00
Sam-I-Amthe interesting thing is having a per-tenant external network can also be wasteful of IPs18:00
mordredthat tenant_id is my tenant_id18:00
mordredfwiw18:00
*** oshvartz has joined #openstack-neutron18:01
Sam-I-Amthe subnet on those is always public ip space?18:01
mordredyes18:01
*** salv-orlando has quit IRC18:01
Sam-I-Amare your private/project networks vxlan? always use private space?18:01
mordredSam-I-Am: this is my private network there http://paste.openstack.org/show/48040918:02
Sam-I-Amwho assigns the network names?18:02
mordredthey do18:02
Sam-I-AmLAN seems to be private, WAN is public18:02
mordredthey auto-create those two networks18:02
mordredyes. this is true18:02
*** rossella_s has quit IRC18:03
mordredhowevver, having general cloud consumption code that includes logic to look at network names to see if they contain the strings wan or lan seems like a thing that's not terribly scalable - a boolean metadata flag would be swell :)18:03
Sam-I-Amyou can attach a router that you create to the public network?18:03
mordredI don't know - I've never tried - it's not a thing I need to do18:03
Sam-I-Amquestion is, do they allow you to do that?18:04
mordred(I have no use for the private network there since the public network allows me to boot vms on it)18:04
Sam-I-Amrackspace does not18:04
Sam-I-Amhence why router:external wouldnt matter18:04
mordredtrying18:04
mordredams01:mordred@camelot:~$ neutron router-create test-router18:04
mordredQuota exceeded for resources: ['router']18:04
mordredI'm guessing "no"18:04
amullermordred: so no floating IPs either on that cloud?18:05
mordrednope18:05
Sam-I-Ammy guess is... if you want a vm on a public net, you boot it there. you can create any number of private networks, but they will never route to public network using neutron-y methods18:05
mordredyes18:05
Sam-I-Amif you want to route private-public, you create a linux box that does routing18:05
mordredthat is my understanding of this cloud18:05
amullermordred: so each VM just gets a public IP and that's it?18:05
amullerno floating IPs, no routers18:05
mordredamuller: if I boot it on the WAN network18:05
mordredwhich, btw, is AWESOME18:05
amullermordred: this is... weird18:05
mordredit's what we ALWAYS want18:05
Sam-I-Amso this makes sense that router:external is false18:05
mordredsure18:05
amullermordred: it's what you want, I wouldn't say this is a typical ask18:06
amullernot all VMs need public IPs, most don't...18:06
mordredit's more typical of openstack deployments than not, actually18:06
mordredthe majority of openstack public clouds allow direct booting of vms on public networks18:06
mordredthere are only 3 that do not18:06
mordredhowever - that's a much deeper discusssion :)18:06
amullerI thought the press declared openstack public clouds dead :)18:06
Sam-I-Amlol18:07
mordredoh them press people :)18:07
amullermordred: OK, I understand your problem, and Neutron can't give you this info presently18:07
amullermordred: it's a legit request18:07
mordredso, to sum up the current problem - because router:external is actually a flag about whether you can attach a router to a network18:07
*** aranjan has quit IRC18:07
Sam-I-Amthere are enough issues with L3 agents that its easier to separate public and private networks by disabling user-deployable routers18:07
mordredthere is _not_ a flag which communicates to users that a given network is external18:07
mordredcol18:07
mordredcool18:07
*** ankurgupta has joined #openstack-neutron18:08
amullermordred: I want you to talk to Carl Baldwin18:08
amullercarl_baldwin: ^18:08
mordredyay! I've communicated the problem sufficiently18:08
amullerhe's cooking up something called address scopes18:08
*** shashank_hegde has joined #openstack-neutron18:08
amullerwhich I think answers this ask, maybe...18:08
mordredneat18:08
amullerperhaps in a hugely overkill way to your specific ask18:08
Sam-I-Ammordred: outside of using clever net names or looking at public vs. private ip space, no.18:08
* mordred sobs18:08
amullermordred: If you could encapsulate this conversation in an email thread or bug that would be a good start18:09
mordredamuller: I can indeed!18:09
amullermordred: thank you18:09
amullermordred: yeah for now Sam-I-Am is correct I think your only way is looking at the name or the address space18:09
amullerboth approaches suck18:10
mordredamuller: thanks for talking through it with me - we're working aroud this at the moment with client-side config18:10
*** harlowja has joined #openstack-neutron18:10
Sam-I-Amin the case of rackspace, you only get 1 public network and it always has the same name18:10
*** mikeit has joined #openstack-neutron18:10
mordredso we just configure "this is the public and this is the private network" for each region of that cloud18:10
Sam-I-Amalso has a rather unique uuid18:10
mordredSam-I-Am: yah. rackspce is ALSO special cased in our code, which makes us sad18:10
*** baoli_ has joined #openstack-neutron18:10
*** chlong has quit IRC18:10
Sam-I-Amalso not 100% organic neutron either :/18:11
*** edmondsw has quit IRC18:11
*** mohankumar has quit IRC18:11
amullernot free range organic vegan neutron? :)18:11
*** sambetts is now known as sambetts-afk18:11
Sam-I-Amnope. too much GMO :)18:11
openstackgerritCedric Brandily proposed openstack/neutron: Refactor OVS-agent tunnel config validate  https://review.openstack.org/25149418:11
amullertoo many waffles I think it was in the Rackspace case18:11
Sam-I-Amgenetically modified openstack18:11
*** changbl has quit IRC18:12
amullerhttps://pypi.python.org/pypi/wafflehaus18:12
Sam-I-Amprobably worth a discussion about flags for network type18:12
amulleranyway, off to lunch, toodles18:12
Sam-I-Amenjoy18:12
*** paul-carlton1 has joined #openstack-neutron18:12
*** spandhe has joined #openstack-neutron18:12
*** amuller is now known as amuller_afk18:13
Sam-I-Ammordred: i'm kind of surprised they let you see physnet and vlan # info as a regular user18:13
*** baoli has quit IRC18:14
*** changbl has joined #openstack-neutron18:14
mordredSam-I-Am: I've never really noticed, since they are not useful information to me :)18:14
*** sridhar_ram has joined #openstack-neutron18:14
*** adam_g has quit IRC18:14
*** adam_g has joined #openstack-neutron18:14
*** aranjan has joined #openstack-neutron18:14
*** adam_g has quit IRC18:15
*** adam_g has joined #openstack-neutron18:15
carl_baldwinamuller: mordred:  at gym now back in an hour18:15
*** shashank_hegde has quit IRC18:16
Sam-I-Ammordred: seems like a security issue18:16
Sam-I-Amregular users can learn the underlying network infra18:16
*** jamespage has quit IRC18:17
*** rm_work has quit IRC18:17
Sam-I-Amlikewise, users probably shouldnt know the network type either18:18
*** baoli_ has quit IRC18:18
Sam-I-Amthe defaults work in the way i expect18:18
*** nlahouti has quit IRC18:20
*** clev_ has quit IRC18:20
*** daneyon__ has quit IRC18:20
*** EinstCrazy has joined #openstack-neutron18:21
*** daneyon has joined #openstack-neutron18:21
*** davidlenwell has quit IRC18:21
*** davideagnello has joined #openstack-neutron18:22
*** rm_work has joined #openstack-neutron18:22
*** davidlenwell has joined #openstack-neutron18:23
*** Guest83268 has quit IRC18:23
*** rstarr has quit IRC18:23
*** maht has joined #openstack-neutron18:23
*** nlahouti has joined #openstack-neutron18:23
*** krtaylor has quit IRC18:23
*** vkmc has joined #openstack-neutron18:23
*** jamespd has quit IRC18:25
*** jamespage has joined #openstack-neutron18:25
*** jamespage has quit IRC18:25
*** jamespage has joined #openstack-neutron18:25
*** EinstCrazy has quit IRC18:25
*** clev has joined #openstack-neutron18:25
*** gberginc has joined #openstack-neutron18:25
*** chlong has joined #openstack-neutron18:26
*** BLZbubba has quit IRC18:26
*** BLZbubba has joined #openstack-neutron18:26
*** crinkle has quit IRC18:27
*** yamamoto has quit IRC18:27
*** mgagne has joined #openstack-neutron18:28
*** mgagne is now known as Guest6345318:28
*** doug-fish has quit IRC18:28
*** jpena has quit IRC18:28
PietHi folks, we wanted to let you know about a results presentation for the Nova Networks to Neutron Migration Study this Tuesday, December 1st at 10AM MST / 12PM EST / 5:00PM UTC.  A link to the virtual room for the presentation can be found at https://wiki.openstack.org/wiki/HorizonUsability_Testing#December_2015_Nova_Networks_to_Neutron_Migration_Study18:28
*** banix has quit IRC18:29
dasmPiet: ++18:29
dasmPiet: thanks18:29
*** crinkle has joined #openstack-neutron18:29
PietNo worries!  I hope you find it helpful!18:30
PietWe're also planning to record the presentation for later consumption18:30
*** jamespd has joined #openstack-neutron18:31
openstackgerritMerged openstack/python-neutronclient: Add 'baremetal' as an allowed vnic type  https://review.openstack.org/24939318:32
*** paul-carlton1 has quit IRC18:33
*** aranjan_ has joined #openstack-neutron18:34
*** claudiub has quit IRC18:34
*** fitoduarte has joined #openstack-neutron18:36
*** aranjan has quit IRC18:36
*** hogepodge has joined #openstack-neutron18:37
*** jistr has joined #openstack-neutron18:37
*** Guest63453 has quit IRC18:38
*** Guest63453 has joined #openstack-neutron18:38
*** aranjan_ has quit IRC18:38
openstackgerritRyan Moats proposed openstack/neutron: Refactor router_info.process_external  https://review.openstack.org/25150218:38
*** glavni_ninja has quit IRC18:39
*** Guest63453 is now known as mgagne18:39
*** eil397 has joined #openstack-neutron18:39
*** krtaylor has joined #openstack-neutron18:41
*** glavni_ninja has joined #openstack-neutron18:41
*** krypto has quit IRC18:41
*** banix has joined #openstack-neutron18:43
openstackgerritRyan Moats proposed openstack/neutron: WIP: Collector Proof of Concept DO NOT MERGE!!!  https://review.openstack.org/21347418:43
*** rook has quit IRC18:43
*** s3wong has joined #openstack-neutron18:44
*** shashank_hegde has joined #openstack-neutron18:44
openstackgerritRyan Moats proposed openstack/neutron: WIP: Collector Proof of Concept DO NOT MERGE!!!  https://review.openstack.org/21347418:45
openstackgerritMerged openstack/python-neutronclient: Use sphinx-build -W in [docs] target  https://review.openstack.org/25107718:45
Sam-I-AmregXboi: sounds like i have something to merge18:45
*** rstarr has joined #openstack-neutron18:45
regXboiSam-I-Am: ?!18:45
Sam-I-AmregXboi: it says do not merge :)18:46
Sam-I-Amso its just tempting18:46
regXboiSam-I-Am: you have been warned :)18:46
openstackgerritRamu Ramamurthy proposed openstack/neutron: In port_dead, handle case when port already deleted  https://review.openstack.org/24784418:46
*** tbachman has joined #openstack-neutron18:47
*** Marga_ has joined #openstack-neutron18:48
*** gampel has joined #openstack-neutron18:48
*** minwang2 has quit IRC18:49
*** abregman has joined #openstack-neutron18:49
*** wolverineav has joined #openstack-neutron18:50
*** minwang2 has joined #openstack-neutron18:50
*** mikeit has quit IRC18:51
*** sripriya has joined #openstack-neutron18:51
*** shashank_hegde has quit IRC18:51
openstackgerritShih-Hao Li proposed openstack/neutron: (WIP) Make MetadataProxyHandler configurable  https://review.openstack.org/24787418:52
*** achanda has joined #openstack-neutron18:52
openstackgerritSwaminathan Vasudevan proposed openstack/neutron: Tune _get_candidates for faster scheduling in dvr  https://review.openstack.org/24228618:52
*** rstarr has quit IRC18:55
*** tbachman has quit IRC18:56
*** tbachman has joined #openstack-neutron18:57
*** salv-orlando has joined #openstack-neutron18:57
*** maht has quit IRC18:57
*** markmcclain has joined #openstack-neutron18:57
*** sthillma has joined #openstack-neutron18:58
*** shashank_hegde has joined #openstack-neutron18:59
*** wolverineav has quit IRC18:59
*** rotbeard has quit IRC19:02
*** changbl has quit IRC19:09
*** rook has joined #openstack-neutron19:09
*** crose has joined #openstack-neutron19:10
*** rstarr has joined #openstack-neutron19:10
openstackgerritSwaminathan Vasudevan proposed openstack/neutron: Debug patch to test the gate failures on SSHTimeOut  https://review.openstack.org/24774819:11
*** Pavo has quit IRC19:13
*** manjeets has quit IRC19:15
*** Pavo has joined #openstack-neutron19:17
*** salv-orlando has quit IRC19:19
*** salv-orlando has joined #openstack-neutron19:20
*** mancdaz has quit IRC19:20
carl_baldwinmordred: amuller_afk:  I just caught up.19:20
carl_baldwinThe conversation makes sense.  router:external indeed has the semantics you claim and isn't what you're looking for.  But, you knew that already.19:21
*** xek has quit IRC19:21
Sam-I-Amcarl_baldwin: yo19:22
*** crose has quit IRC19:22
*** mancdaz has joined #openstack-neutron19:22
carl_baldwinAddress scopes may help but may not be exactly what you're looking for either.  It is basically where address overlap isn't allowed.19:23
carl_baldwinSam-I-Am: hi19:23
Sam-I-Amhowdy19:23
carl_baldwinmordred: Would a simple tag on the network suffice for this?  Those are coming in Mitaka.19:23
*** ihrachys has joined #openstack-neutron19:23
Sam-I-Amcarl_baldwin: i think this makes sense19:24
*** safchain has quit IRC19:24
*** mickeys has joined #openstack-neutron19:24
carl_baldwinSam-I-Am: Howdy back!  Could you be more clear about "this"?19:24
openstackgerritRyan Tidwell proposed openstack/neutron: [WIP] Add BGP Callback and RPC Call implementations  https://review.openstack.org/21699419:24
openstackgerritRyan Tidwell proposed openstack/neutron: BGP Dynamic Routing: introduce BgpDriver  https://review.openstack.org/20763519:24
openstackgerritRyan Tidwell proposed openstack/neutron: Add BGP Dynamic Routing DB Model and Basic CRUD  https://review.openstack.org/20162119:24
openstackgerritRyan Tidwell proposed openstack/neutron: BGP Dynamic Routing: introduce BgpDrScheduler model  https://review.openstack.org/20760719:24
openstackgerritRyan Tidwell proposed openstack/neutron: Add Queries For BGP Route Lookups  https://review.openstack.org/24219219:24
openstackgerritRyan Tidwell proposed openstack/neutron: BGP Dynamic Routing: introduce BgpDrAgent  https://review.openstack.org/20762519:24
openstackgerritRyan Tidwell proposed openstack/neutron: Add L3 Notifications To Enable BGP Dynamic Routing  https://review.openstack.org/24106219:24
*** javeriak has quit IRC19:25
*** gampel has quit IRC19:26
*** mkoderer has quit IRC19:26
*** yamamoto has joined #openstack-neutron19:28
*** mkoderer has joined #openstack-neutron19:28
*** yamamoto has quit IRC19:33
*** aranjan has joined #openstack-neutron19:35
*** shivrao has joined #openstack-neutron19:35
*** amuller_afk is now known as amuller19:38
*** markmcclain has quit IRC19:41
*** markmcclain has joined #openstack-neutron19:41
*** abregman is now known as abregman|ZzZzZz19:42
*** abregman|ZzZzZz is now known as abregman19:43
*** javeriak has joined #openstack-neutron19:43
*** lindycoder has quit IRC19:46
*** minwang2 has quit IRC19:47
*** lindycoder has joined #openstack-neutron19:47
*** manjeets has joined #openstack-neutron19:48
*** aranjan has quit IRC19:57
*** virtuallynathan_ has joined #openstack-neutron19:57
*** Pavo has quit IRC19:58
*** mickeys has quit IRC20:00
sc68calSam-I-Am: you're a busy bee today :)20:00
Sam-I-Amsc68cal: nahhhh20:02
*** busterswt has joined #openstack-neutron20:03
*** azbiswas has joined #openstack-neutron20:06
*** vishy has joined #openstack-neutron20:06
*** nyechiel_ has quit IRC20:08
openstackgerritPaul Michali proposed openstack/neutron-fwaas: Remove dependency on neutron for topics  https://review.openstack.org/25152120:09
*** abregman is now known as abregman|afk20:11
*** doug-fish has joined #openstack-neutron20:14
*** minwang2 has joined #openstack-neutron20:14
*** jwy has joined #openstack-neutron20:19
*** Marga_ has quit IRC20:19
*** martinus__ has quit IRC20:21
*** tongli has joined #openstack-neutron20:21
*** martinus__ has joined #openstack-neutron20:22
*** ihrachys has quit IRC20:23
*** jwy has quit IRC20:25
*** jistr has quit IRC20:27
*** nlahouti has quit IRC20:29
*** jwy has joined #openstack-neutron20:29
*** achanda has quit IRC20:31
*** adelia has joined #openstack-neutron20:31
*** RichardRaseley has joined #openstack-neutron20:34
*** salv-orlando has quit IRC20:35
*** salv-orlando has joined #openstack-neutron20:35
*** neiljerram has quit IRC20:38
*** gampel has joined #openstack-neutron20:39
*** oshvartz has quit IRC20:41
openstackgerritAaron Rosen proposed openstack/neutron: Fix bug updating port_bindings when port has been deleted  https://review.openstack.org/24489620:41
*** EinstCrazy has joined #openstack-neutron20:42
*** maht has joined #openstack-neutron20:43
*** matrohon has joined #openstack-neutron20:44
*** EinstCrazy has quit IRC20:47
*** rook has quit IRC20:48
*** Marga_ has joined #openstack-neutron20:50
*** abregman|afk has quit IRC20:50
*** doug-fish has quit IRC20:51
*** gampel has quit IRC20:51
*** yfried_ has joined #openstack-neutron20:51
*** dims_ has joined #openstack-neutron20:53
*** doug-fish has joined #openstack-neutron20:54
*** aranjan has joined #openstack-neutron20:54
*** doug-fis_ has joined #openstack-neutron20:55
*** dims has quit IRC20:56
*** Pavo has joined #openstack-neutron20:57
*** busterswt has quit IRC20:57
*** doug-fi__ has joined #openstack-neutron20:57
*** gvrangan has joined #openstack-neutron20:57
gvranganHi!! , We are currently working with Opendaylight Project looking for some help about L2 Population20:58
*** doug-fish has quit IRC20:58
*** aranjan has quit IRC20:58
gvranganIs there any use cases page docuemntd fpr that? or some tests?20:58
*** doug-fis_ has quit IRC21:00
*** tongli has quit IRC21:00
*** javeriak has quit IRC21:01
*** doug-fi__ has quit IRC21:02
*** aranjan has joined #openstack-neutron21:02
*** dims_ has quit IRC21:03
Sam-I-Amgvrangan: in what sense?21:03
*** Piet has quit IRC21:04
gvranganwe  wanted to know all the use cases for the L2 Population implemetation21:04
*** changbl has joined #openstack-neutron21:04
*** doug-fish has joined #openstack-neutron21:04
*** Marga_ has quit IRC21:05
*** Marga_ has joined #openstack-neutron21:06
Sam-I-Amthere's really only one. when using tunnel-type networks, you dont want a bunch of broadcast floating around everywhere. so... with l2pop, neutron handles populating the fdb and blocks conventional arp.21:06
mordredcarl_baldwin: heya!21:09
*** doug-fish has quit IRC21:09
mordredcarl_baldwin: yes. simple tag would be awesome21:09
mordredcarl_baldwin: well, let me be clear though ...21:09
carl_baldwinmordred: hi21:09
mordredcarl_baldwin: it would want to be a simple tag on the network that was NOT different per provider21:09
mordredcarl_baldwin: free-form tags here would be worse than useless21:09
*** Marga_ has quit IRC21:10
mordred_but_ - if we have general tagging per network and can piggyback that in some known and published way - that woudl be killer awesome21:10
mordredcarl_baldwin: in fact, there are like 4 pieces of information I need to know about networking in any cloud, if I could just define a set of tags that deployers should set, then we could totally defcore that up and whatnot21:12
carl_baldwinmordred: Then there may a possibility here.  The general tag mechanism will be a thing in Mitaka as far as I know.  I haven't heard anything about any kind of standardization on tag usage.21:12
*** saksham has joined #openstack-neutron21:12
Sam-I-Ammordred: so something like external:true ?21:12
Sam-I-Amor public21:12
Sam-I-Amwords can be funny21:12
mordredyah. external:true would be fine - northbound would be fine - "gets-off-cloud" would be fine21:12
Sam-I-Ami like gets-off-cloud :)21:12
mordredhonestly, "bunny-rabbit" would be fine if I knew it was always there :)21:12
mordredcarl_baldwin: so, I mention standards bcause glance has image tagging - and it's completely useless21:13
mordredas a consumer of more than one cloud21:13
*** rook has joined #openstack-neutron21:13
mordredhttp://paste.openstack.org/show/479871/21:14
*** dims has joined #openstack-neutron21:14
mordredthat's a print out of the image dict for each cloud for the tihng they have that's closest as my script could find to "an ubuntu image"21:14
mordredwhich you'd think would be a normal thing a user would want to find21:14
Sam-I-Aminteresting, there's a simple defined difference between external and internal networks21:15
Sam-I-Ampublic:true, external:true, private:false, provider:true ... lots of ways to skin the cat21:16
mordredcitycloud has "tags: os:ubuntu" internap has "properties: distro: ubuntu" rax has "org.openstack__1__os_distro: com.ubuntu" ultimum has "os_distro: ubuntu" and most people have nothing21:17
mordredSam-I-Am: yup21:17
*** rdo has quit IRC21:18
Sam-I-Ameither way it should only be a boolean21:18
mordredSam-I-Am: I'm mostly concerned, as a user of a given cloud with the decision tree for "I want to route traffic off the cloud": "do I need floating IPs?" yes: "which network should I attach a router to?" no: "which network should I boot VMs on"21:18
*** mickeys has joined #openstack-neutron21:18
mordredif you add to that "can I made a private network?" - it's covers all of the clouds I have found so far21:19
mordredand is the set of questions I need to know before I can use the cloud21:19
Sam-I-Ammakes sense21:19
*** tbachman has quit IRC21:19
*** Marga_ has joined #openstack-neutron21:19
Sam-I-Amsort of need a neutron capabilities-list21:19
mordredyeah - kinda21:20
*** tbachman has joined #openstack-neutron21:20
mordredI mean, I have a bunch of code that gets this mostly right currently21:20
mordredthe internap/rackspace deploy is the one I can't sort out21:20
Sam-I-Amwell, rackspace isnt much more than the neutron api21:21
Sam-I-Ami dont know what internap is doing21:21
Sam-I-Amdocumentation might be one way to solve this, but its not something you can get back by querying an api21:21
Sam-I-Amwhether or not you can boot a vm on a public/provider network is another story21:21
mordredSam-I-Am: well, there are two things:21:22
Sam-I-Amyou'd have to know whether that network was mapped to the particular compute node21:22
mordredSam-I-Am: in rackspace, the neutorn API is not exposed21:22
mordredso it's not possible for a user to query for neutron network flags21:22
mordredhowever, since there is only one network in rackspace21:23
mordredthe question becomes moot - "which network should I boot on" doesn't have to be answered21:23
mordredfor internap, there is a per-tenant public and a per-tenant private21:24
Sam-I-Ammordred: well, neutron net-list works with rax21:24
mordredSam-I-Am: oh? that's new21:24
mordrednope21:24
mordred404 Not Found21:24
mordredThe resource could not be found.21:24
mordredthat's what I get21:24
mordredyou must be special21:24
*** amuller has quit IRC21:24
Sam-I-Amit only shows private networks21:24
*** rdo has joined #openstack-neutron21:26
*** yfried_ has quit IRC21:27
Sam-I-Amit doesnt show the public or service-net networks21:27
Sam-I-Amif you only have those, not sure what it does21:27
Sam-I-Am(i cant delete my private networks to try it)21:27
bloganSam-I-Am: it should if you do --shared in net-list21:27
*** mlavalle has quit IRC21:27
Sam-I-Amblogan: ding21:28
Sam-I-Ammordred: if you do that, you'll also see the unique uuids for the special networks21:28
mordredI still get 40421:28
Sam-I-Amwhats in your rc file?21:29
mordredhttp://paste.openstack.org/show/48043221:29
mordredmy catalog has https://dfw.networks.api.rackspacecloud.com/v2.0 for the publicURL of DFW for network21:30
roaetHenryG: hey there! Any new work for Online Migrations?21:30
*** jprovazn has quit IRC21:30
pc_mFYI, looks like PEP8 is broken on all repos. I'm looking at neutron-vpnaas, rm_work is looking at neutron-lbaas (that's where I first saw it). blogan says neutron fails too.21:31
*** mlavalle has joined #openstack-neutron21:32
carl_baldwinmordred: I just added a link to this conversation on this spec review:  https://review.openstack.org/#/c/216021/21:32
carl_baldwinmordred: I thought the spec had been approved but I guess it is still under discussion.21:32
Sam-I-Ammordred: thats what i see too21:32
*** doug-fish has joined #openstack-neutron21:32
mordredcarl_baldwin: sweet!21:33
Sam-I-Ammordred: this is my rather organic file - http://paste.openstack.org/show/95Rt35Xp3KZHiGtZjlP9/21:34
Swamipc_m: I have seen "test_restart_l3_agent_on_sighup" fail in neutron dsvm_functional21:34
*** doug-fis_ has joined #openstack-neutron21:35
*** doug-fis_ has quit IRC21:35
Swamipc_m: but I am not sure if it is related to pep8 that you mentioned above.21:35
*** doug-fis_ has joined #openstack-neutron21:35
*** Pavo has quit IRC21:35
mordredSam-I-Am: yah - none of that should matter21:35
pc_mSwami: not sure, but most of these seem to be noise - import ordering, for example.21:35
mordredSam-I-Am: I think your tenant has access to things mine does not21:36
Swamipc_m:ok21:36
Sam-I-Ammordred: which is odd?21:36
HenryGroaet: Hi! Yes, work needs to be done to reboot https://review.openstack.org/12494621:36
mordredSam-I-Am: nah - there are features that you have to ask to have enabled21:36
mordredI'm guessing that as docs folks they enable more of them for you so you can document everything21:36
*** doug-fish has quit IRC21:37
Sam-I-Ammordred: except this account is just for the upstream docs team21:37
roaetHenryG: ah. reboot as in get the patch mergable?21:37
openstackgerrityalei wang proposed openstack/neutron: Add network_update RPC into sriov agent  https://review.openstack.org/23349921:37
Sam-I-Ammordred: this particular account is one of those 'open source contributor' accounts... basically gets a credit every month21:38
HenryGroaet: I am going to be mostly offline Tue and Wed. Can we sync up on Thur/Fri?21:38
Sam-I-Ambut i did not ask for any specific access21:38
roaetof course, what time would be best for you?21:38
mordredSam-I-Am: weird21:38
Sam-I-Ammordred: might open a ticket21:38
HenryGroaet: It's more than that. It needs to be refactored quite a bit, into two parts.21:38
roaetHenryG: ok we can talk more about it on Thur/Fri?21:39
blogananyone else run tox pep8 lately? just want to make sure i'm not the crazy one and its failing just for me21:39
HenryGroaet: sounds good21:39
bloganotherwise working on a fix21:39
roaetHenryG: did you want to schedule something?21:39
HenryGroaet: we can do that. What TZ you in?21:39
roaetCST21:40
*** Piet has joined #openstack-neutron21:41
HenryGroaet: 10am CST Friday OK for you?21:42
roaetexcellent. Ill meet you in here?21:42
HenryGroaet: yep21:42
*** dims_ has joined #openstack-neutron21:42
*** baoli has joined #openstack-neutron21:43
*** baoli has quit IRC21:43
*** dims has quit IRC21:43
roaetHenryG: Sorry I just realized I am interviewing someone at 10am21:44
roaetCould we something around that?21:44
roaet+set21:44
roaetit is 1 hour long21:44
HenryGroaet: 11am CST works for me21:44
roaetHenryG: thanks! I may be 5 min late or so to walk from the interview area to my desk21:45
HenryGroaet: no problem21:45
Sam-I-Amcarl_baldwin: i also commented on that spec21:45
carl_baldwinSam-I-Am: Good stuff21:47
*** rook has quit IRC21:51
*** psanchez has quit IRC21:51
*** rook has joined #openstack-neutron21:52
*** tflynn has joined #openstack-neutron21:52
*** baoli has joined #openstack-neutron21:52
*** psanchez has joined #openstack-neutron21:53
*** fitoduarte has quit IRC21:54
*** matrohon has quit IRC21:55
*** dedery has joined #openstack-neutron21:56
*** mwagner_lap has quit IRC21:57
*** rook has quit IRC21:57
*** tflynn_ has joined #openstack-neutron21:57
Sam-I-Amever have one of those days where you expected to do X and ended up doing Y?21:59
Sam-I-Amyet it some way, Y is related to X21:59
anteayaarmax: who might be working with bgpvpn to bring them back into sync with the branching model for neutron?22:00
*** nlahouti has joined #openstack-neutron22:00
*** tflynn has quit IRC22:01
anteayaarmax: I'm not sure if anyone on that project saw sdague's comment on this patch which he left after this patch merged: https://review.openstack.org/#/c/230358/9/zuul/layout.yaml22:01
openstackgerritPaul Michali proposed openstack/neutron-vpnaas: Fixing pep8 failures  https://review.openstack.org/25155622:01
anteayaarmax: also do we have any confidence other drivers are not going down that same road?22:01
*** Guest87643 has joined #openstack-neutron22:03
*** Marga__ has joined #openstack-neutron22:05
*** thorst_ has quit IRC22:05
*** Marga_ has quit IRC22:06
*** thorst_ has joined #openstack-neutron22:06
*** markvoelker has joined #openstack-neutron22:07
*** Pavo has joined #openstack-neutron22:07
openstackgerritSlawek Kaplonski proposed openstack/neutron: Add extension_manager and support for extensions in linuxbridge agent  https://review.openstack.org/25054222:07
*** Prithiv has joined #openstack-neutron22:08
*** prithivm has joined #openstack-neutron22:08
*** Marga__ has quit IRC22:09
*** thorst_ has quit IRC22:10
armaxanteaya: tmorin is the guy’s nick22:12
openstackgerrityalei wang proposed openstack/neutron: Add network_update RPC into sriov agent  https://review.openstack.org/23349922:12
armaxanteaya: iirc, but what road are you talking about?22:13
anteayaarmax: is anyone working with him to come back into alignment? or is he doint this alone22:13
anteayawell somehow he is in a situation of feeling he needs to test his plugin not with neutron master22:13
anteayawhich from an infra perspective feels a bit odd22:13
anteayaas he now has to set up tests to test against liberty22:14
pc_majmiller: vpn commit is up.22:14
anteayaso if this is a design model neutron is allowing it would be great if you could chat with infra about it22:14
*** virtuallynathan_ has quit IRC22:15
ajmillerpc_m: OK22:15
anteayaas we are treating this as an outlier that we hope will soon be fixed22:15
pc_majmiller: thanks! I'll be going offline in about 10 mins.22:15
anteayaso we are hoping someone in neutron leadership is working with him to bring this back into an expected workflow22:15
openstackgerritSwaminathan Vasudevan proposed openstack/neutron: Static routes not added to qrouter namespace for DVR  https://review.openstack.org/22802622:15
armaxanteaya: to be honest I personally don’t like this either…22:15
anteayaarmax: okay well this isn't working for infra22:15
anteayaso if it also isn't working for neutron22:16
*** julim has quit IRC22:16
anteayalet's address the situation together and then ensure noone else feels this is a good idea22:16
anteayaas this is creating extra burden for both of us22:16
anteayaarmax: current patch in review: https://review.openstack.org/#/c/243725/222:17
armaxanteaya: actually I was about to send an email to the ML to clarify my position about this22:17
anteayaarmax: don't let me keep you22:18
armaxbecause I can’t deal with the explosion of patches that has come as a result of the arrangement we’re in22:18
*** nlahouti has quit IRC22:18
anteayaarmax: if you can manage a review on that patch, I would be grateful, but post to the mailing list is higher priority22:18
anteayaarmax: oh thank you22:18
anteayaarmax: you have no idea how happy I am to hear you say that22:18
armaxI think people/project should be empowered22:18
armaxbut not at the expenses of a few22:18
anteayaarmax: I support you 100&22:18
anteayaagreed22:19
*** emagana has quit IRC22:19
anteayaarmax: please by all means, compose your post22:19
armaxanteaya: ok, hopefully it’l match your expectations22:19
armax:)22:19
ajmillerpc_m It looks like it's check queue  pep8 has failed22:20
pc_majmiller: :(22:20
ajmillerhttps://jenkins02.openstack.org/job/gate-neutron-vpnaas-pep8/179/console22:20
anteayaarmax: you already exceed my expectations, good sir22:21
* armax blushes22:21
anteaya:)22:21
pc_majmiller: Seems to be reporting more files than local22:22
*** tiswanso has quit IRC22:22
pc_majmiller: I'll delete all .pyc files and retry in my local area.22:23
ajmillerpc_m OK, sounds good22:23
* pc_m that's what I did in neutron and all hell broke loose.22:23
*** ZZelle has joined #openstack-neutron22:23
pc_marmax: are you aware of the pep8 carnage?22:23
*** daneyon has quit IRC22:23
armaxpc_m: carnage?22:24
pc_mpep8 failing in all repos22:24
armaxall branches too?22:24
pc_mI have a patch for VPN (but still has issues), there is one for LB, neutron fails, and haven't checked FW.22:24
pc_marmax: Seen on master22:25
armaxpc_m: link?22:25
openstackgerritHirofumi Ichihara proposed openstack/neutron: Add availability_zone support for router  https://review.openstack.org/22441822:25
ajmillerarmax: also looks like stable/liberty (cf https://review.openstack.org/#/c/249638/)22:25
pc_marmax: Well I just saw it in my local LB repo. Confirmed by others. Patches up are 251546 and 25155622:25
openstackgerritHirofumi Ichihara proposed openstack/neutron: Add functional test for availability_zone support  https://review.openstack.org/21097722:25
pc_marmax: I just cloned neutron and ran pep8 and it has tons of failures.22:25
openstackgerritBrandon Logan proposed openstack/neutron: Fix some pylint errors that came with new pylint  https://review.openstack.org/25156022:26
pc_marmax: I think jobs are failing pep8 also.22:26
armaxit must be pylint22:26
bloganpc_m, armax: ^22:26
pc_marmax: yes22:26
*** mancdaz has quit IRC22:26
*** manjeets has left #openstack-neutron22:26
armaxhttps://pypi.python.org/pypi/pylint22:26
armaxit released yesterday22:26
pc_myeah, what blogan said :)22:26
armaxand we’re back to the usual problem22:27
bloganpushed up a patch that fixes it for m locally, but it is adding a global rule disable22:27
armaxwhich we never fixed here:22:27
armax https://review.openstack.org/#/c/163506/22:27
armaxhere: https://review.openstack.org/#/c/177193/2/global-requirements.txt22:27
armaxwe seem we pin it22:27
pc_majmiller: I removed .pyc files and .tox dir in vpn area, and it is passing pep8. Now sure why we see more failures in the job running.22:28
armaxso I am a bit at loss to what’s going on22:28
*** mancdaz has joined #openstack-neutron22:28
ZZellearmax, pc_m, hi22:28
pc_mZZelle: yo22:29
bloganhmm22:29
armaxhere: http://logs.openstack.org/38/249638/1/gate/gate-neutron-lbaas-pep8/ba595bb/console.html#_2015-11-30_16_27_40_81022:29
ZZellearmax, pc_m, the last trouble we add with pylint was not pylint itself BUT astroid22:29
armaxit looks like we’re pulling pylint 1.5.022:29
armaxso something is not working with the global pin22:29
blogani'm getting 1.4.422:29
bloganbut yeah it snot being pinned22:30
armaxperhaps sdague might be able to help22:30
pc_mI have pep8 1.5.7 pylint 1.4.422:30
ZZellearmax, the behavior of pylint depends on the "raw" information provided by astroid ... so a new release of astroid could change pylint behavior22:31
armaxajmiller: the global requirement pin happened on master22:31
armaxon Apr 2822:31
armaxso that should apply to liberty too22:32
*** regXboi has quit IRC22:32
armaxZZelle: but shouldn’t still pull the version we’re pinned to?22:32
ajmillerarmax: It looks like it is explictly pulling 1.5.022:32
ajmillerhttp://logs.openstack.org/38/249638/1/gate/gate-neutron-lbaas-pep8/ba595bb/console.html22:33
armaxajmiller: right, it is22:33
pc_mMy astroid is at 1.3.822:33
armaxajmiller: does this mean that astroid needs pinning too?22:33
ZZellearmax, iirc, there is no constraint on astroid version22:33
armaxno there’s isn’t22:33
ajmillerWish I knew.  stable/liberty is pulling astroid==1.4.122:33
ZZellearmax, i think we have a winner: https://pypi.python.org/pypi/astroid/1.4.122:34
ZZellereleased yesterday22:34
armaxZZelle: both did22:34
armaxZZelle: pylint released too22:35
armaxyesterday22:35
pc_mZZelle: yeah, seems so. I have 1.38 locally and it is working.22:35
pc_m1.3.822:35
pc_mLocally, i updated astroid (to 1.4.1) and ran pep8 again. Still passed.22:36
ZZellearmax, pylint is pinned not astroid22:36
*** adelia_ has joined #openstack-neutron22:37
*** manjeets_ has quit IRC22:38
armaxZZelle: neutron changes on the constrainted pep8 job are running fine22:38
armaxso the constraints are actually doing their job22:38
armaxso there are multiple ways to handle this breakage22:39
*** mihero has joined #openstack-neutron22:39
pc_mI just updated pylint from 1.4.1 to 1.5.0 locally, and it too passed. Removed *.pyc, .tox dir and re-ran pep8, still passed - locally22:40
armaxwith switching to constrained jobs for the affected projects being the most effective way to handle this once and for alll22:40
*** adelia has quit IRC22:41
bloganarmax: would that change the tox -e pep8 to do what the contrainted jobs do?22:41
bloganfor running tests lcoally?22:41
ZZelleblogan, i doubt it22:41
ZZellethe requirements.txt enforces minimal versions only22:42
*** adelia_ has quit IRC22:42
armaxblogan: you have a a pep8-constraints job22:42
armaxhttps://github.com/openstack/neutron/blob/master/tox.ini#L11622:42
*** rtheis has quit IRC22:42
bloganarmax: so would you want to just make pep8 do what pep8-contraints does?22:43
*** kawa2014 has quit IRC22:44
armaxblogan: no22:44
*** doug-fis_ has quit IRC22:44
armaxblogan: what would be the point?22:44
bloganwell wouldn't tox -e pep8 fail all the time then?22:44
*** doug-fish has joined #openstack-neutron22:44
armaxnot all the time22:45
ZZellearmax, on my computer, tox -epep8 is working with astroid 1.3.8 but not astroid 1.4.0 and 1.4.1 (both released yesterday)22:46
ZZellearmax, i would be in favor of pinning astroid version as the aim of the change you pointed was to pin pylint behavior22:47
armaxZZelle: it might take a while to prove that it works, but that’s one interim solution22:47
bloganarmax: what i mean is right now if i run tox pep8 it fails, if i run pep8-constraints it passes.  so if we just started doing pep8-constraints then pep8 won't be fixed and thats what most people use before they push up patches22:47
bloganperhaps i'm missing something22:48
pc_mZZelle: I'm confused. I had pylint 1.4.1, pep8 1.5.7, astroid 1.3.8 (worked). Updated pylint to 1.5.0 and astroid to 1.4.1 and it still passed.22:48
bloganpc_m: i thought you were getting a ton of errors too22:48
armaxblogan: you want to detect the breakage in a controlled fashion, that’s all22:48
*** nlahouti has joined #openstack-neutron22:48
*** doug-fish has quit IRC22:49
pc_mZZelle: job in check queue has pylint 1.5.0, pep8 1.5.7, and astroid 1.4.1 and it is failing tests.22:49
ZZellepc_m, perhaps the trouble is astroid>=1.4 with pylint < 1.522:49
*** doug-fish has joined #openstack-neutron22:49
ZZellepc_m, sure? according to test-requirements.txt, it should be pylint 1.4.422:49
pc_mblogan: I'm trying vpn right now, which passed locally, but is failing in check queue.22:49
pc_mZZelle: in my local repo?22:50
*** doug-fis_ has joined #openstack-neutron22:50
pc_mZZelle: right. 1.4.4 pylint in my area.22:51
*** oldcrowew has joined #openstack-neutron22:52
oldcrowewis anyone having issues updating kilo to liberty using ovs?22:52
pc_mSo I had 1.4.4(pylint) and 1.38 (astroid), for vpn repo and tests passed. Failed in check queue with 1.5.0 and 1.4.1. I updated to the same version and it still works in my local area.22:53
*** emagana_ has joined #openstack-neutron22:53
*** gissi has joined #openstack-neutron22:53
*** doug-fish has quit IRC22:54
*** doug-fis_ has quit IRC22:55
*** Marga_ has joined #openstack-neutron22:56
ZZellepc_m, vpnaas is not a right example22:56
openstackgerritMerged openstack/neutron-lib: Remove pypy as tox env  https://review.openstack.org/24942322:56
pc_mI went to my neutron repo (with pylint at 1.5.0 and astroid at 1.4.1), cleared .pyc files and .tox dir and ran pep8. Still get failures.22:56
ZZellepc_m, it uses an unpinned version of pylint22:56
*** Marga_ has quit IRC22:57
*** Marga_ has joined #openstack-neutron22:57
pc_mZZelle: Guess Im wondering why neutron-vpnaas passes locally, with same versions as are used in the check queue job.22:57
*** jobewan has quit IRC22:57
*** mwagner_lap has joined #openstack-neutron22:57
*** yamahata has joined #openstack-neutron22:58
oldcrowewi see this is the ovs-agent log: ERROR neutron.agent.linux.ovsdb_monitor  Interface monitor is not active22:58
oldcrowewany ideas?22:58
*** gberginc has quit IRC22:58
oldcrowewhttps://bugs.launchpad.net/neutron/+bug/149542322:59
openstackLaunchpad bug 1495423 in neutron "Cannot get dom0's ovsdb monitor result" [Undecided,New]22:59
*** banix has quit IRC22:59
oldcrowewis close but its xen not kvm22:59
oldcrowewsame setup though22:59
ZZellearmax, pc_m, neutron-vpnaas is not a really good example as pylint dependency is added in tox.ini file instead of test-requirements.txt which means pylint is unpinned23:01
*** Marga_ has quit IRC23:01
*** dims has joined #openstack-neutron23:02
*** doug-fish has joined #openstack-neutron23:03
*** dims_ has quit IRC23:03
*** anshul has joined #openstack-neutron23:03
ZZellearmax, pc_, imo, my proposal would be to pin astroid==1.3.8 (to get the stable fashion way) in neutron and update *-aas to move pylint dependency in test-requirements and pin astroid23:03
openstackgerritMichael Krotscheck proposed openstack/neutron: Added CORS support to Neutron  https://review.openstack.org/23698823:04
*** changbl has quit IRC23:04
*** Marga_ has joined #openstack-neutron23:04
*** mickeys has quit IRC23:04
*** mickeys has joined #openstack-neutron23:05
openstackgerritMichael Krotscheck proposed openstack/neutron: Added CORS support to Neutron  https://review.openstack.org/23698823:05
*** changbl has joined #openstack-neutron23:05
*** kbringard has quit IRC23:05
*** doug-fish has quit IRC23:07
pc_mZZelle: OK. So in neutron. I had pylint 1.5.0, asteroid 1.4.1. When I run, test-requirements sets pylint to 1.4.4 and the test fails locally. If I change tox-requirements to 1.5.0 for pylint then the test has other errors. So it looks like 1.4.4 and 1.38 work, and 1.5.0 and 1.4.1 give new failure output for import ordering, and other such errors, but no longer have undefined variable.23:08
pc_mZZelle: Your plan sounds good.23:08
*** minwang2 has quit IRC23:09
ZZellepc_m, i would prefer we stay on pylint==1.4.4 for the moment (it worked so it should continue to work) and moves to pylint==1.5.0 latter (typically after mitaka1)23:10
pc_mZZelle: Agree, as going to 1.5.0 with 1.4.1 has other failures.23:11
ZZellepc_m, more precisely evaluate pylint 1.5.0 after mitaka123:11
ZZelleand moves or not to 1.5.023:11
pc_mZZelle: roger.23:13
*** breitz has quit IRC23:15
*** oldcrowew has left #openstack-neutron23:15
*** breitz has joined #openstack-neutron23:15
ZZellepc_m, the question is: could we add a constraint on astroid in neutron* test-requirements.txt and after update global requirements to unblock neutron* pep823:16
pc_mZZelle: Is astroid set in global requirements?23:18
ZZellepc_m, nope23:18
pc_mZZelle: not sure I follow the concern... can it just be added to test-requirements?23:20
ZZellepc_m, normally requirements.txt test-requirements.txt are subset of https://github.com/openstack/requirements requirements files23:21
pc_mZZelle: ah. right.23:22
pc_mso need to add to global and then pick a pinned version for neutron?23:22
ZZellepc_m, i think we should ask the best ever PTL!23:25
ZZellearmax, ^^23:25
armaxZZelle: I am hardly the best PTL, ever, but thanks for the compliment23:26
armaxZZelle: as for the requirements machinery, I think it’s best to relay the question in openstack-infra23:26
*** sridhar_ram has quit IRC23:26
armaxpc_m, ZZelle: even though pinning astroid should do the trick I believe23:27
*** hichihara has joined #openstack-neutron23:30
*** kawa2014 has joined #openstack-neutron23:30
*** lori is now known as lori|away23:32
ZZellepc_m, could you handle it? It's more than time for me to leave23:32
openstackgerritJohn Davidge proposed openstack/python-neutronclient: Add support for default subnetpools API  https://review.openstack.org/24254723:32
pc_mit's late here too. I can try to initiate.23:33
ZZelleI will continue if needed tomorrow23:33
ZZellebye23:33
*** ZZelle has quit IRC23:33
pc_mZZelle: ok23:33
pc_mtrying now...23:33
*** sridhar_ram has joined #openstack-neutron23:35
*** djan has joined #openstack-neutron23:38
*** barra204 has quit IRC23:38
*** shakamunyi has joined #openstack-neutron23:38
*** jerrygb has quit IRC23:39
*** jerrygb has joined #openstack-neutron23:39
openstackgerritMerged openstack/neutron: Support Unicode request_id on Python 3  https://review.openstack.org/25085323:41
*** lindycoder has quit IRC23:43
*** nlahouti has quit IRC23:46
*** kawa2014 has quit IRC23:47
*** signed8bit is now known as signed8bit_ZZZzz23:48
*** kawa2014 has joined #openstack-neutron23:48
*** prithivm has quit IRC23:51
*** Prithiv has quit IRC23:51
*** nlahouti has joined #openstack-neutron23:53
*** RichardRaseley has quit IRC23:54
*** RichardRaseley has joined #openstack-neutron23:56
*** manjeets has joined #openstack-neutron23:58
*** miyagishi_t has joined #openstack-neutron23:58
*** tbachman has quit IRC23:59
*** tbachman has joined #openstack-neutron23:59
« Saturday, 2015-11-28 Index Tuesday, 2015-12-01 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!