Thursday, 2019-03-28

*** lbragstad has joined #openstack-tc00:45
*** lbragstad has quit IRC00:57
*** jamesmcarthur has joined #openstack-tc00:58
*** jamesmcarthur has quit IRC01:07
*** ricolin has joined #openstack-tc01:07
*** sapd1 has joined #openstack-tc01:35
*** ijolliffe has quit IRC02:07
*** gouthamr has quit IRC02:10
*** gouthamr has joined #openstack-tc02:12
*** diablo_rojo has joined #openstack-tc02:24
*** diablo_rojo has quit IRC02:47
*** whoami-rajat has joined #openstack-tc02:58
*** whoami-rajat has quit IRC05:07
*** jaypipes has quit IRC05:09
*** jaypipes has joined #openstack-tc05:09
*** tdasilva has quit IRC06:15
*** whoami-rajat has joined #openstack-tc06:31
*** e0ne has joined #openstack-tc07:17
*** e0ne has quit IRC07:19
*** tosky has joined #openstack-tc08:04
*** jpich has joined #openstack-tc08:49
asettleMorning o/09:00
*** ricolin has quit IRC09:27
openstackgerritJens Harbott (frickler) proposed openstack/governance master: Rename Chef OpenStack project and update it
fricklertc-members: ^^ not sure whether this is the correct way to propose it or whether there should be some discussion first, feedback welcome09:42
*** e0ne has joined #openstack-tc09:47
evrardjpthanks frickler09:56
*** dtantsur|afk is now known as dtantsur09:59
*** melwitt has quit IRC10:26
*** e0ne has quit IRC10:31
*** melwitt has joined #openstack-tc10:32
*** e0ne has joined #openstack-tc10:36
*** e0ne has quit IRC10:58
*** e0ne has joined #openstack-tc11:05
*** weshay_pto is now known as weshay11:59
*** whoami-rajat has quit IRC12:10
*** whoami-rajat has joined #openstack-tc12:21
*** ijolliffe has joined #openstack-tc12:22
*** ijolliffe has quit IRC12:26
*** ijolliffe has joined #openstack-tc12:27
*** lbragstad has joined #openstack-tc12:30
dhellmannfrickler : thanks, that doesn't look controversial to me but a mailing list post with a bit more background would be good12:33
dhellmannmaybe it's just a matter of wanting consistency?12:33
*** e0ne has quit IRC12:41
*** e0ne has joined #openstack-tc12:43
*** jamesmcarthur has joined #openstack-tc12:45
*** lbragstad has quit IRC12:49
*** ijolliffe has quit IRC12:58
*** lbragstad has joined #openstack-tc13:02
*** Luzi has joined #openstack-tc13:03
*** ijolliffe has joined #openstack-tc13:03
*** lbragstad has quit IRC13:07
evrardjpmnaser: I think we should do the same for openstack-ansible if it's not too hard.13:16
*** lbragstad has joined #openstack-tc13:17
mnaserdhellmann: from a release mgmt perspective, is it probably advisable to land this change after we release for now?13:18
mnaserevrardjp: yeah, I think jimmy tried to push a change at some point and there was an issue around it13:18
evrardjpI seem to remember that13:19
evrardjpstill, I think it would be nice to be consistent everywhere13:19
*** e0ne has quit IRC13:20
dhellmannmnaser : that's a question for smcginnis. I think either way it's going to mean updating a bunch of files in the releases repo13:25
openstackgerritLance Bragstad proposed openstack/governance master: Elaborate on the business value of documentation owners
evrardjpif we change the releases management for OSA again mnaser that wouldn't be too bad. In fact it would help at OSA releasing :p13:34
evrardjpbut that's a longer story.13:34
smcginnismnaser: I agree, given where we are in the cycle, it would b ebest to do that change after the stein release date so we don't have a fire drill.13:35
*** ijolliffe has quit IRC13:36
mnaserevrardjp: yeah... anything to make life easy13:37
evrardjpsmcginnis: I totally agree :)13:37
mnasersmcginnis: great, mind just leaving that as a review comment when you have a few (non existent) seconds ? :P13:38
evrardjpthere is no urgency on osa side13:38
smcginnismnaser: Will do!13:38
evrardjp(I guess same applies for chef)13:38
*** ijolliffe has joined #openstack-tc13:51
*** jamesmcarthur has quit IRC13:52
openstackgerritLance Bragstad proposed openstack/governance master: Describe the business value of consistent RBAC
openstackgerritLance Bragstad proposed openstack/governance master: Elaborate on the business value of Designate
lbragstadi appreciate all the reviews jroll14:06
jrolllbragstad: no problem :)14:07
*** jamesmcarthur has joined #openstack-tc14:27
*** whoami-rajat has quit IRC14:30
*** e0ne has joined #openstack-tc14:38
*** ricolin has joined #openstack-tc14:59
fungii guess it's office hour now15:00
smcginnisMaybe a quieter week after all the highly active ones.15:01
fungithuogh i'm also in the security sig weekly meeting so may not be paying super close attention in here15:01
openstackgerritJens Harbott (frickler) proposed openstack/governance master: Rename Chef OpenStack project and update it
*** tdasilva has joined #openstack-tc15:06
mnasersomething I've shopped around is the idea of an ipv6 mgmt plane support for devstack plugins15:10
mnaserso that we can run devstack in IPv6 only mode (therefore means all services can run with IPv6)15:10
mnaserany thoughts about it?15:10
fungimaking sure our inter-service communications can work over ipv6 would be great, in my opinion15:12
jrollmy TC brain is 100% for that15:12
jrollmy ironic brain says ensuring that all the variations of pxe booting and NIC firmware we support can do v6 seems scary :)15:13
*** e0ne has quit IRC15:14
jrollI guess we don't need all of it working, just the more common ones15:14
smcginnisControl plane might be OK. But last I knew, there were still some devices that don't fully support IPv6 in all scenarios. I know we had some cinder storage vendors that did not support it for iSCSI, just management API.15:15
dtroyerI think the important bit is that it isn't our software that is the limiting factor15:15
jrollgiven that all services ostensibly support wsgi, thus support standard web servers like httpd and nginx, control plane should just be done other than CI, right?15:16
jrolldtroyer: ++15:16
jroll(big fat "should" there, as usual)15:16
dtroyerright, nice assumptions about how long a string representation of an IP address is, or that a colon in a URL is only after the scheme or before a port number, yadda yadda yadda15:17
* dtroyer has never made either of those mistakes, nope, never…15:18
jrollooo, fair.15:18
evrardjpdtroyer: :)15:20
*** e0ne has joined #openstack-tc15:20
*** Luzi has quit IRC15:25
mnaseryeah, so I think it's a fairly achievable goal15:28
mnaserand it's usually just filtered around fixing assumptions around ip scheme15:28
mnaserI agree with jroll and smcginnis that some projects will _inevitebly_ interact with ipv4 only services, but at least, openstack itself should be able to run ipv6 in terms of listeners, etc, etc15:29
fungiyeah, it would be great if our software were capable of operating without ipv4, and offered v4 support as a legacy fallback for interacting with devices/environments which lack proper v6 support15:31
gmann++, as software side if we make sure of working it with ipv6, it is much benefits for operator perspective.15:32
fungiso i think having working v6-only integration test jobs is a laudable goal to prove that's possible15:36
mnaserfungi: yeah, im thinking maybe someone can help build a job that disables ipv4 and runs IPv6 only for example15:36
fungiobviously we do still need to test that we can interoperate with v4-only suystems too15:36
mnaserwe are moving all our infra to v615:36
mnaserand we have ran into some challenges, so I thought it would make sense15:36
fungiit's an excellent suggestion indeed15:37
mnaserwould anyone like to volunteer to draft up the document and we can try and find a champion if we think it's good?15:37
gmannfungi: +1, we have support of that in devstack with base jobs ready seems (running as n-v on tempest). every project having a voting ipv6 only job15:37
mnaseryeah, if we aim for a voting ipv6 only job as a way to know if its complete, that seems pretty good way to not regress15:38
mnaseranyways, by doing this, we'll probably make our code more robust too because i.e. if someone was parsing code in a questionable way that made assumption on an ip, now they can use $some_python_lib to do that (hopefully we don't have people implementing ipv6 parsing libs :P)15:39
*** e0ne has quit IRC15:40
jrollmnaser: too late :P
jrollat least it's in oslo and not a bunch of projects, though :)15:40
mnaserjroll: that's a good thing to actually list in that document15:41
mnaseryeah, Oslo.utils is pretty accessible15:41
jrollmnaser: yeah, the whole file has some helpful stuff for this goal15:41
mnaseryeah ipv6 platform detection and all15:42
* mnaser looks at history of that file and suggests them all as potential goal champions15:42
fungiworks for me!15:49
mnaserI guess we don't have anyone volunteer to push this up as a potential goal15:51
mnaserI have a lot going on right now (cloud upgrades, openstack-ansible release) so I couldn't do it right _now_ but I could do it after release, but that might be a bit late for PTLs15:52
mnaserI think it would be neat if someone could help out at least draft that stuff out15:52
jrollI would if I had time, but I don't, sorry15:52
jrollmnaser: by "too late for PTLs", do you mean you wanted to have something before the PTG to talk about for train? or U?15:56
*** diablo_rojo has joined #openstack-tc16:02
mnaserjroll: train, correct, similar to the concerns lbragstad had at the time16:16
fungizigo also mentioned a potentially interesting goal on openstack-discuss just a few minutes ago16:21
TheJuliamy tc brain says that if we want v6 to be a priority for teams, we're going to need to partner with the deployment tools to force it forward16:27
*** e0ne has joined #openstack-tc16:27
TheJuliaAnd more so from a standpoint of not doing or avoiding doing things that are v6 only and driving things in projects that are v6 only16:27
mnaserTheJulia: yeah, this is one of the rare goals that deployment tools can actually participate in too!16:29
TheJuliaMy ironic brain, like jroll's cries as well, considering we'll likely ahve to hold on to some custom roms to do bare metal16:29
mnaserwe usually were left out of the fun =P16:29
fungii suspect pushing it on deployment tools before we even test that it works under devstack may be putting the cart before the horse16:30
evrardjpfungi: that's indeed an interesting goal, it simplifies the user interface16:31
TheJuliathe cloud vendors are another issue16:32
*** whoami-rajat has joined #openstack-tc16:32
*** smcginnis is now known as smcginnis_pto16:48
*** e0ne has quit IRC17:07
*** e0ne has joined #openstack-tc17:08
*** e0ne has quit IRC17:17
*** jamesmcarthur_ has joined #openstack-tc17:17
*** e0ne has joined #openstack-tc17:21
*** jamesmcarthur has quit IRC17:21
*** ricolin has quit IRC17:26
*** jpich has quit IRC17:30
*** ricolin has joined #openstack-tc17:38
*** e0ne has quit IRC17:45
*** ricolin has quit IRC17:46
*** Sundar has joined #openstack-tc17:58
*** dtantsur is now known as dtantsur|afk18:08
*** jamesmcarthur_ has quit IRC18:47
*** e0ne has joined #openstack-tc19:33
*** jamesmcarthur has joined #openstack-tc19:51
*** jaypipes_ has joined #openstack-tc19:54
*** jaypipes has quit IRC19:54
*** e0ne has quit IRC20:00
*** openstackgerrit has quit IRC21:07
*** ijolliffe has quit IRC21:37
*** whoami-rajat has quit IRC22:52
*** jamesmcarthur has quit IRC22:57
*** jamesmcarthur has joined #openstack-tc23:14
*** Sundar has quit IRC23:15
*** adriant has quit IRC23:34
*** jamesmcarthur has quit IRC23:42
*** jamesmcarthur has joined #openstack-tc23:43
*** jamesmcarthur has quit IRC23:48
*** adriant has joined #openstack-tc23:49

Generated by 2.15.3 by Marius Gedminas - find it at!