Thursday, 2015-07-02

*** puranamr has quit IRC00:01
*** puranamr_ has quit IRC00:02
*** puranamr has joined #akanda00:03
openstackgerritSean Roberts proposed stackforge/akanda-rug: follow semver release versioning servers change  https://review.openstack.org/19778700:18
openstackgerritSean Roberts proposed stackforge/akanda-rug: follow semver release versioning servers change  https://review.openstack.org/19778700:27
*** puranamr has quit IRC00:28
openstackgerritSean Roberts proposed stackforge/akanda-rug: follow semver release versioning servers change  https://review.openstack.org/19778700:29
adam_gsarob, would it be more accurate to count 2015.1 as our 1.0?00:31
adam_g1.0.0, rather00:31
sarobi was thinking that at first00:31
sarobbut since akanda has been around for awhile00:32
saroband neutron is at 7.0.000:32
sarobit follows to follow neutron version for now00:32
sarobwhat do you think?00:32
adam_gon one hand it seems strange to appear like we've got 7 major releases under our belt, esp having had announcements go out celebrating our '1.0' in vancouver00:35
sarobhmm00:35
adam_gon the other, having akanda-neutron plugin versioned differently than the server its meant to pluginto could be confusing00:35
adam_gi wonder how the other neutron-* plugins are handling it00:35
sarobill peek00:36
adam_gthe neutron-* things look like they've moved to 7.0.000:37
sarobsub proj are following 700:37
adam_gthe plugins tho, are still on the old scheme00:37
adam_gstackforge/networking-*00:37
sarobi think 7 is good enough00:39
sarobthis naming numbering stuff makes my head hurt00:40
adam_gnot that it really matters for us yet, but what is the plan for upgrades with the new scheme?00:41
adam_g2015.1.0 > 7.0.000:41
sarobhow else?00:42
openstackgerritSean Roberts proposed stackforge/akanda-neutron: follow semver release versioning servers change  https://review.openstack.org/19778900:43
adam_ghow do you upgrade from 2015.1.0 to 7.0.0? thats a downgrade00:43
adam_gi assume it could just be punted to distros to use epochs in their packaging00:43
sarobi believe that is the plan00:45
sarobpunted sounds like the correct word00:45
sarobfrom my reading of the meeting logs00:45
openstackgerritSean Roberts proposed stackforge/akanda-horizon: follow semver release versioning servers change  https://review.openstack.org/19779000:47
openstackgerritSean Roberts proposed stackforge/akanda-appliance: follow semver release versioning servers change  https://review.openstack.org/19779100:49
sarobarg00:51
sarobi think i need to clear the value00:52
sarobthen change it to 700:52
sarobit doesnt like going backwards00:52
sarobhmm, neutron didnt have the same problem patching00:54
sarobadam_g: is this worth trying to figure out?00:56
adam_gsarob, if we're going to change versioning, ya00:56
sarobadam_g: well, i mean patching removing the version and then reapplying00:58
sarobadam_g: rather than why neutron was able to without pep8 failure00:58
adam_gwhy would that help?00:58
sarobadam_g: wouldnt removing the version00:59
adam_gweird, i cant reproduce that failure locally00:59
adam_gi dont know enough about pbr to know why thats borking00:59
adam_git might be because there is no 7.0.0x tag01:00
adam_gbut dont quote me on that01:00
adam_g(or go pushing tags just yet :)01:00
sarobhmm, let me chk the timing01:00
adam_goh, i got the error here now01:01
adam_gafter a pbr upgrade01:01
adam_gsarob, yeah, creating a 7.0.0b1 tag locally fixes it01:02
sarobright on01:02
adam_gwe never tagged the liberty1 milestone01:02
sarobi was getting around to it01:02
adam_gif we decide on 7.0.0, maybe tag current repos HEAD to 7.0.0b101:03
adam_gand then those versioning changes should merge01:03
adam_gassuming thats the actual issue, still kinda a guess01:03
sarobim double checking the other projects01:09
adam_gi think its just a tag issue, but ya double check before pushing anything and we may want to get consensus among the group (ie, +A on your patches) before actually pushing the tags01:11
adam_gi gotta run now01:11
adam_gbaby's official past-due so maybe ill talk to you tomorrow. :)01:12
adam_gmaybe not01:12
sarobroger that good buddy01:12
sarobim taking tomorrow off01:12
sarobill be in on friday01:12
*** puranamr has joined #akanda01:38
*** puranamr has quit IRC01:42
*** JustinRK has quit IRC08:14
*** JustinRK has joined #akanda08:15
*** openstackgerrit has quit IRC09:19
*** openstackgerrit has joined #akanda09:20
*** puranamr has joined #akanda13:58
*** puranamr has quit IRC14:14
*** puranamr has joined #akanda15:17
*** puranamr has quit IRC15:26
*** puranamr has joined #akanda17:10
*** rods_ has joined #akanda17:13
*** rods_ has quit IRC17:31
*** puranamr has quit IRC18:12
*** puranamr has joined #akanda18:13
*** puranamr_ has joined #akanda18:33
*** puranamr has quit IRC18:37
adam_gryanpetrello, would you guys be okay with the file-based debug modes going away in favor of rug-ctl debug modes being persistent?18:52
ryanpetrellothat's fine with me18:56
ryanpetrelloas I recall, the file-based thing was just a cheap way to get persistence18:56
*** fzylogic has joined #akanda18:56
drwahlya, i don't think file-based debug modes going away would be a big deal18:57
* fzylogic agrees18:58
fzylogicthe only problem I see is that some sort of persistence would be nice18:58
adam_gcool19:01
adam_gwhat if that meant akanda requiring a database? :)19:01
fzylogicshouldn’t be a problem with something like berkeleydb or sqlite19:02
fzylogicno reason to use anything more powerful than that19:03
adam_gfzylogic, well this in the context of scaling out to multiple rug processes across nodes19:03
fzylogicright19:03
adam_gso, mysql/postgres/etc19:04
fzylogicbut a router wouldn’t move around between rug processes very much, right?19:04
fzylogicideally, never (except in the case of a node being removed from the pool?)19:04
adam_gya19:04
adam_ger, right19:04
adam_grugs maintaining their own sqlite dbs for this is not much different then them using the filesystem19:05
fzylogicno, it’s really just the user interface that changes19:06
fzylogicif you use a generic db interface, the actual database used could just be an implementation detail too19:06
adam_gright,  it would be (oslo.db)19:07
fzylogicdefault to sqlite and let the user change that to something shared if they need it19:07
fzylogicor heck, default to a noop19:07
fzylogicand just don’t persist unless the user asks for it explicitly19:07
ryanpetrellowe still on for our meeting today adam_g markmcclain ?19:15
adam_gim down to hang19:15
adam_gsarobs out19:15
*** puranamr_ has quit IRC19:25
*** puranamr has joined #akanda19:32
*** rods has quit IRC20:19
*** puranamr has quit IRC20:31
*** puranamr has joined #akanda20:38
*** puranamr has quit IRC21:47
*** puranamr has joined #akanda21:50
*** puranamr_ has joined #akanda22:16
*** puranamr has quit IRC22:19
*** puranamr has joined #akanda22:21
*** puranamr_ has quit IRC22:24
*** puranamr has quit IRC22:28
*** puranamr has joined #akanda22:28
*** puranamr_ has joined #akanda22:30
*** puranamr has quit IRC22:33
*** puranamr_ has quit IRC22:35
*** puranamr has joined #akanda22:35
*** puranamr_ has joined #akanda22:38
*** puranamr has quit IRC22:39
*** puranamr has joined #akanda22:40
*** puranamr_ has quit IRC22:42
*** puranamr has quit IRC22:48
*** puranamr has joined #akanda22:48
*** puranamr_ has joined #akanda22:59
*** puranamr has quit IRC23:03
*** puranam__ has joined #akanda23:04
*** puranamr_ has quit IRC23:07
*** puranam__ has quit IRC23:09
*** puranamr has joined #akanda23:10
*** puranamr has quit IRC23:11
*** puranamr has joined #akanda23:14
*** puranamr has quit IRC23:14
*** puranamr has joined #akanda23:15
*** puranamr_ has joined #akanda23:16
*** puranam__ has joined #akanda23:18
*** puranamr_ has quit IRC23:18
*** puranamr has quit IRC23:19
*** rods_ has joined #akanda23:40
*** rods_ has quit IRC23:42

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