Friday, 2014-10-17

*** clu__ has joined #openstack-horizon00:00
*** ramishra has quit IRC00:02
*** clu_ has quit IRC00:03
*** clu__ is now known as clu_00:03
*** ArcTanSusan has quit IRC00:16
*** ArcTanSusan_ has joined #openstack-horizon00:16
*** ybathia has quit IRC00:36
*** ArcTanSusan_ has quit IRC00:52
*** alexpilotti has quit IRC00:53
*** exploreshaifali has quit IRC00:53
*** ramishra has joined #openstack-horizon00:58
*** hightall has joined #openstack-horizon00:59
*** cedrics has joined #openstack-horizon01:00
*** ramishra has quit IRC01:02
*** cedrics has quit IRC01:06
*** alexpilotti has joined #openstack-horizon01:10
*** alexpilotti has quit IRC01:13
*** jasondotstar has quit IRC01:17
*** nlahouti has quit IRC01:20
*** jacalcat has joined #openstack-horizon01:24
*** nlahouti has joined #openstack-horizon01:24
*** nlahouti has quit IRC01:24
*** Longgeek has joined #openstack-horizon01:37
*** alexpilotti has joined #openstack-horizon01:41
*** Longgeek has quit IRC01:42
*** ramishra has joined #openstack-horizon01:59
*** ramishra has quit IRC02:03
openstackgerritMiguel Grinberg proposed a change to openstack/horizon: Honor selected stack template and environment sources  https://review.openstack.org/12361502:07
*** rbertram has quit IRC02:16
*** clu_ has quit IRC02:25
*** jacalcat has quit IRC02:26
*** sbfox has joined #openstack-horizon02:35
*** harlowja is now known as harlowja_away02:47
*** nlahouti has joined #openstack-horizon02:48
*** nlahouti has quit IRC02:49
*** nlahouti has joined #openstack-horizon02:49
*** ramishra has joined #openstack-horizon02:52
*** ramishra_ has joined #openstack-horizon02:58
*** ramishra has quit IRC03:01
*** ramishra_ has quit IRC03:02
*** ramishra has joined #openstack-horizon03:02
*** ybathia has joined #openstack-horizon03:03
*** sbfox has quit IRC03:05
*** ramishra has quit IRC03:07
*** sbfox has joined #openstack-horizon03:15
*** Longgeek has joined #openstack-horizon03:22
*** ramishra has joined #openstack-horizon03:22
*** julim has joined #openstack-horizon03:24
*** jcoufal has joined #openstack-horizon03:27
openstackgerritA change was merged to openstack/horizon: Fixing template syntax  https://review.openstack.org/12902803:32
*** sbfox has quit IRC03:43
*** katyafervent is now known as katyafervent_awa03:52
*** tzumainn has quit IRC04:07
openstackgerritLIU Yulong proposed a change to openstack/horizon: Make no response notification(msg) level to INFO  https://review.openstack.org/11771804:07
*** hhuang has joined #openstack-horizon04:12
*** ramishra has quit IRC04:17
*** sayan|afk has quit IRC04:19
*** sbfox has joined #openstack-horizon04:19
*** ramishra has joined #openstack-horizon04:20
*** ramishra has quit IRC04:20
*** sayan|afk has joined #openstack-horizon04:31
*** ramishra has joined #openstack-horizon04:50
*** jcoufal has quit IRC04:52
*** vkmc has quit IRC04:53
*** vivekd has joined #openstack-horizon04:58
*** sseago has quit IRC05:01
*** sseago has joined #openstack-horizon05:01
*** sayan has quit IRC05:12
*** sbfox has quit IRC05:13
*** sbfox has joined #openstack-horizon05:14
*** vivekd_ has joined #openstack-horizon05:14
*** vivekd has quit IRC05:17
*** vivekd_ is now known as vivekd05:18
*** ramishra has quit IRC05:19
*** sayali has joined #openstack-horizon05:20
*** lhcheng has quit IRC05:25
*** lhcheng has joined #openstack-horizon05:25
*** lhcheng has quit IRC05:30
*** ramishra has joined #openstack-horizon05:31
*** masco has joined #openstack-horizon05:37
*** danieru has quit IRC05:46
*** danieru has joined #openstack-horizon05:46
*** lhcheng has joined #openstack-horizon05:48
*** sayan has joined #openstack-horizon05:48
*** hhuang has quit IRC05:49
*** ybathia has quit IRC05:49
*** vivekd_ has joined #openstack-horizon05:49
*** vivekd has quit IRC05:51
*** vivekd_ is now known as vivekd05:51
*** vivekd_ has joined #openstack-horizon05:53
*** vivekd has quit IRC05:56
*** vivekd__ has joined #openstack-horizon05:57
*** vivekd__ is now known as vivekd05:57
*** vivekd_ has quit IRC05:58
*** vivekd_ has joined #openstack-horizon05:59
openstackgerritLIU Yulong proposed a change to openstack/horizon: Make no response notification(msg) level to INFO  https://review.openstack.org/11771806:00
*** vivekd has quit IRC06:02
*** vivekd has joined #openstack-horizon06:04
*** vivekd__ has joined #openstack-horizon06:06
*** katyafervent_awa is now known as katyafervent06:06
*** vivekd_ has quit IRC06:06
openstackgerritOpenStack Proposal Bot proposed a change to openstack/horizon: Imported Translations from Transifex  https://review.openstack.org/12913806:06
*** mrunge has joined #openstack-horizon06:08
*** vivekd has quit IRC06:09
*** vivekd__ is now known as vivekd06:09
*** k4n0 has joined #openstack-horizon06:09
*** vivekd_ has joined #openstack-horizon06:20
*** exploreshaifali has joined #openstack-horizon06:21
*** r1chardj0n3s is now known as r1chardj0n3s_afk06:22
*** vivekd has quit IRC06:22
*** vivekd_ is now known as vivekd06:23
*** exploreshaifali_ has joined #openstack-horizon06:27
*** jprovazn has joined #openstack-horizon06:27
*** exploreshaifali has quit IRC06:29
*** romainh has joined #openstack-horizon06:30
*** vokhrimenko has joined #openstack-horizon06:32
*** ramishra has quit IRC06:33
*** echevemaster has quit IRC06:36
*** vivekd has quit IRC06:38
*** e0ne has joined #openstack-horizon06:38
*** amcrn has quit IRC06:39
*** ramishra has joined #openstack-horizon06:39
*** vivekd has joined #openstack-horizon06:40
*** echevemaster has joined #openstack-horizon06:41
*** echevemaster has quit IRC06:43
*** e0ne has quit IRC06:43
openstackgerritJiri Tomasek proposed a change to openstack/horizon: Restyled sidebar to resemble UX guidelines  https://review.openstack.org/12628906:44
*** nlahouti has quit IRC06:44
*** e0ne has joined #openstack-horizon06:47
*** e0ne has quit IRC06:48
*** AliceR has quit IRC06:51
*** sayan has quit IRC06:51
*** sayan has joined #openstack-horizon06:52
*** e0ne has joined #openstack-horizon06:52
*** Ala has joined #openstack-horizon06:53
*** Ala has quit IRC06:53
*** vivekd has quit IRC06:54
*** e0ne has quit IRC06:54
*** ygbo has joined #openstack-horizon06:55
*** vivekd has joined #openstack-horizon06:56
*** Ala has joined #openstack-horizon06:58
*** exploreshaifali_ has quit IRC07:00
openstackgerritAkihiro Motoki proposed a change to openstack/horizon: Do not log keystone token  https://review.openstack.org/12885907:11
*** ttrifonov is now known as zz_ttrifonov07:13
*** hightall has quit IRC07:14
*** qba73 has joined #openstack-horizon07:14
*** hightall has joined #openstack-horizon07:26
*** cedrics has joined #openstack-horizon07:28
mrungeamotoki, did you change anything for your review https://review.openstack.org/128859 ?07:31
amotoki_mrunge: I just resolved a conflict.07:31
mrungeamotoki, ah ok. thanks. viewing diffs between patch sets would be a great feature for gerrit07:32
amotoki_mrunge: for more detail, L.160-162 at https://review.openstack.org/#/c/128859/1..2/openstack_dashboard/api/keystone.py07:32
amotoki_This diff caused the conflict.07:33
mrungeamotoki, thank you07:33
openstackgerritMatthias Runge proposed a change to openstack/horizon: Refactor metering dashboard  https://review.openstack.org/10701107:34
amotoki_mrunge: when a diff between patch sets, I first reivew all files in the current set, then see diffs from a previous version and look for review mark.07:35
amotoki_mrunge: review marks help me to find related files :-)07:35
mrungeamotoki, neat. I mostly just look at latest patch set, and previous comments07:36
mrungebut it's quite easy to get lost there07:36
lhchengmrunge:  Using the "Reference Version:" in gerrit should help you see diffs between patch set no?07:39
*** ramishra has quit IRC07:41
* mrunge is just now discovering, gerrit could be configured (somehow)07:42
*** zz_ttrifonov is now known as ttrifonov07:49
amotoki_lhcheng: could you check my response to your comment in https://review.openstack.org/#/c/127642/ ? I think MultipleTimes is not needed in this case.07:50
lhchengamotoki: checking...07:51
*** e0ne has joined #openstack-horizon07:52
*** alexpilotti has quit IRC07:55
*** jpich has joined #openstack-horizon07:55
*** alexpilotti has joined #openstack-horizon07:56
*** alexpilotti has quit IRC07:56
*** alexpilotti has joined #openstack-horizon07:58
lhchengamotoki: looks good, changed my review08:00
amotoki_lhcheng: thanks!08:00
*** wuhg has joined #openstack-horizon08:04
openstackgerritA change was merged to openstack/horizon: Change django message storage backends default.  https://review.openstack.org/10179108:06
*** MaxV has joined #openstack-horizon08:12
*** akrivoka has joined #openstack-horizon08:12
openstackgerritPaul Karikh proposed a change to openstack/horizon: Remove admin role name 'admin' hardcode  https://review.openstack.org/12374108:16
openstackgerritJiri Tomasek proposed a change to openstack/horizon: Modal fixes - 2 columns, padding, header, membership  https://review.openstack.org/11351908:17
*** ramishra has joined #openstack-horizon08:18
openstackgerritPaul Karikh proposed a change to openstack/horizon: Remove admin role name 'admin' hardcode  https://review.openstack.org/12374108:18
*** aix has joined #openstack-horizon08:19
*** ramishra_ has joined #openstack-horizon08:19
*** ramishra has quit IRC08:20
*** jcoufal has joined #openstack-horizon08:24
*** hightall has quit IRC08:25
*** hightall has joined #openstack-horizon08:26
openstackgerritA change was merged to openstack/horizon: Imported Translations from Transifex  https://review.openstack.org/12913808:27
*** Longgeek has quit IRC08:40
*** Longgeek has joined #openstack-horizon08:41
*** ramishra_ has quit IRC08:42
*** ramishra_ has joined #openstack-horizon08:43
*** sayan has quit IRC08:46
*** jcoufal has quit IRC08:53
*** jcoufal has joined #openstack-horizon08:54
*** exploreshaifali has joined #openstack-horizon09:10
*** sayan|afk has quit IRC09:10
*** athomas has joined #openstack-horizon09:13
*** lhcheng has quit IRC09:14
*** exploreshaifali has quit IRC09:14
*** lhcheng has joined #openstack-horizon09:14
*** ramishra_ has quit IRC09:14
*** lhcheng has quit IRC09:18
*** ramishra_ has joined #openstack-horizon09:18
*** romainh1 has joined #openstack-horizon09:21
*** romainh has quit IRC09:21
openstackgerritA change was merged to openstack/horizon: Do not log keystone token  https://review.openstack.org/12885909:24
*** tosky has joined #openstack-horizon09:26
*** hightall has quit IRC09:35
*** ramishra_ has quit IRC09:39
*** cody-somerville has joined #openstack-horizon09:50
*** viklund has quit IRC09:51
*** exploreshaifali has joined #openstack-horizon09:54
*** athomas has quit IRC09:58
*** sambetts has joined #openstack-horizon09:58
openstackgerritSayali Lunkad proposed a change to openstack/horizon: Removes term inconsistency in the dashboard  https://review.openstack.org/7322210:00
*** cody-somerville has quit IRC10:02
*** vivekd has quit IRC10:05
*** sayali has quit IRC10:06
*** ZZelle has quit IRC10:07
*** ZZelle has joined #openstack-horizon10:07
*** athomas has joined #openstack-horizon10:09
*** qba73 has quit IRC10:10
zigoMaxV: Are you around?10:11
MaxVzigo: yes?10:12
*** qba73 has joined #openstack-horizon10:13
*** romainh1 has quit IRC10:14
*** romainh has joined #openstack-horizon10:14
openstackgerritSam Betts proposed a change to openstack/horizon: Pass correct project ID to get tenant_usages  https://review.openstack.org/12903710:16
*** jcoufal has quit IRC10:18
*** cedrics has quit IRC10:19
*** sayali has joined #openstack-horizon10:21
*** sambetts is now known as sambetts_10:28
*** sambetts_ is now known as sambetts10:29
*** sayali has quit IRC10:29
jpichDid flake8 always take so much CPU when running before? It's taking up 90% of my CPU lately which is pretty noticeable, slowdown wise...10:36
*** exploreshaifali has quit IRC10:38
lifelessjpich: recent upgrade to it gave it parallelisation10:41
lifelessjpich: so its now using more cores probably; always been slow :)10:41
jpichlifeless: Aah that may well be related indeed :-) Thank you!10:42
*** sayali has joined #openstack-horizon10:54
*** jcoufal has joined #openstack-horizon10:57
*** jcoufal has quit IRC10:57
*** jcoufal has joined #openstack-horizon10:57
*** romainh has left #openstack-horizon11:04
*** romainh has joined #openstack-horizon11:12
*** cedrics has joined #openstack-horizon11:20
*** exploreshaifali has joined #openstack-horizon11:21
akrivokahello Horizoners! please take some time to review this bp which should significantly improve our dev process: https://blueprints.launchpad.net/horizon/+spec/template11:22
akrivokaI also feel this deserves an email on the mailing list... just to get more opinions11:23
openstackgerritSwati Shukla proposed a change to openstack/horizon: Confirm Dialog for Delete operation  https://review.openstack.org/12887311:24
*** lhcheng has joined #openstack-horizon11:30
openstackgerritMasco Kaliyamoorthy proposed a change to openstack/horizon: cancel next previous buttons are aligned properly  https://review.openstack.org/12921211:33
*** colinmcnamara has joined #openstack-horizon11:34
*** lhcheng has quit IRC11:34
*** exploreshaifali has quit IRC11:44
*** vkmc has joined #openstack-horizon11:46
*** raildo_away is now known as raildo11:56
*** tnovacik has joined #openstack-horizon12:06
*** jacalcat has joined #openstack-horizon12:06
*** packet has joined #openstack-horizon12:10
*** mrunge has quit IRC12:11
*** packet has quit IRC12:12
*** thiagop has joined #openstack-horizon12:12
*** sayan has joined #openstack-horizon12:13
*** _crobertsrh is now known as crobertsrh12:15
*** mflobo has joined #openstack-horizon12:16
*** hightall has joined #openstack-horizon12:21
*** romainh has quit IRC12:22
*** romainh has joined #openstack-horizon12:24
*** g4rg4m3|_ has joined #openstack-horizon12:31
*** packet has joined #openstack-horizon12:38
*** jacalcat has quit IRC12:38
*** colinmcnamara has quit IRC12:40
*** sayan has quit IRC12:40
*** tnovacik has quit IRC12:41
*** masco has quit IRC12:41
*** KurtMartin has quit IRC12:44
*** KurtMartin has joined #openstack-horizon12:45
*** hightall has quit IRC12:50
openstackgerritAkihiro Motoki proposed a change to openstack/horizon: Disable router HA attribute due to Neutron bug  https://review.openstack.org/12834112:51
*** sayan has joined #openstack-horizon12:54
*** k4n0 has quit IRC12:55
*** tzumainn has joined #openstack-horizon12:55
*** ramishra has joined #openstack-horizon12:59
openstackgerritA change was merged to openstack/horizon: Restrict the length of user name input  https://review.openstack.org/12881913:00
*** johnma has joined #openstack-horizon13:00
*** cedricss has joined #openstack-horizon13:01
*** cedrics has quit IRC13:03
thiagopsambetts: Hello!13:06
sambettsthiagop: Hi!13:10
thiagopI'm taking care of some docs here to end my sprint. When you want to continue discussing that matter of yesterday, ping me please (I assume that we're going to wait for David and Eric)13:11
*** doug-fish has joined #openstack-horizon13:12
*** akrivoka has quit IRC13:13
jpichdoug-fish: Thanks for following up on that network topology i18n issue! I had just noticed it missed Juno. I think it'll be safe to backport too (since the strings aren't translated currently anyway...)... so my secret plan is to do just that13:17
doug-fishjpich: sneaky!13:17
doug-fishI notice that you defined the problem slightly larger than I did13:17
jpichIf we give a heads-up to the i18n team then, hopefully some will be happy to translate for the stable release :)13:17
jpichdoug-fish: That's what I do? lol13:17
jpichPartial-Fix are super welcome too13:18
jpichfeel free to unassign yourself if you don't have time to look into the rest once your own patch merges13:18
doug-fishyeah, I was just sorting out if I was going got try to fix the whole thing or just the part I initially noticed13:18
jpichThank you! Either way, same author or not, multiple patches are ok :-)13:19
*** openstackgerrit has quit IRC13:19
jpichcertainly do not let me stop you from fixing All The Things13:19
doug-fishjpich:  sure, sounds good.  Thanks for keeping us coordinated! Sometimes I wonder if people even look for an existing bug report before they open a new one.  I'm really not sure how I missed that one!13:20
*** openstackgerrit has joined #openstack-horizon13:20
jpichdoug-fish: The launchpad search is not very good unfortunately :(13:21
*** exploreshaifali has joined #openstack-horizon13:21
doug-fishyes!  let's blame the tools!13:21
doug-fishbecause there is _no way_ that it was my fault.  :-)13:21
tsufievthiagop, hi! I've studied your policy PoC. Couldn't the Keystone provide the same rules for other services as he does for himself?13:21
jpichdoug-fish: Now I'm not sure that's quite what I said ;)13:23
doug-fishhaha!13:23
* tsufiev wonders whether it was meant for all Openstack projects to migrate to Storyboard eventually13:25
jpichtsufiev: I believe that's the end goal, as we've hit some of the limits of Launchpad for our needs already13:26
tsufievjpich, let's hope the end goal arrives earlier than the Last Day :)13:28
jpichThat sounds very dramatic :-)13:28
jpichtsufiev: I hope there's a "state of storyboard" session at the Summit, I haven't looked at the schedule yet. Are you aware of how far they've come? I haven't looked at the project in a while13:28
*** peristeri has joined #openstack-horizon13:28
* tsufiev looking at it now13:29
jpichOk!13:30
*** lhcheng has joined #openstack-horizon13:32
*** ramishra has quit IRC13:32
*** sigmavirus24_awa is now known as sigmavirus2413:36
*** lhcheng has quit IRC13:36
*** hightall has joined #openstack-horizon13:38
*** masco has joined #openstack-horizon13:40
*** ramishra has joined #openstack-horizon13:42
*** cody-somerville has joined #openstack-horizon13:47
*** masco has quit IRC13:52
*** colinmcnamara has joined #openstack-horizon13:58
*** romainh has left #openstack-horizon14:00
*** ZZelle has quit IRC14:06
*** ZZelle has joined #openstack-horizon14:06
thiagoptsufiev: Good day. Sure it can, that's one of the main ideas behind the endpoint policy: you can set a policy for cinder, and the cinder itself will have to ask keystone for his policy14:07
thiagoptsufiev: If you have more than a cinder service in your stack, they can share the same policy and so on...14:07
amotoki_ FYI: devstack patch which compiles message catalogs is now approved and in the gate. https://review.openstack.org/#/c/128601/14:09
amotoki_ you can always see up-to-date translations during dev cycle.14:09
*** ramishra has quit IRC14:12
*** radez_g0n3 is now known as radez14:13
*** wuhg has quit IRC14:13
tsufievjpich, one of StoryBoard devs told me 'it will be possible to migrate to StoryBoard once it has support for milestone/branches and user groups'. Anyways, there should be discussion at Summit when it will be possible14:17
sambettsthiagop: Do you know what timezone dave and eric are in, because I won't be a run as late into my evening as I did yesterday14:18
tsufievthiagop, then it may be worth to reflect it in the PoC 'cause from what I've seen it looks like a feature supported only for Keystone service14:19
thiagopsambetts: no, sorry. :/14:19
jpichtsufiev: Sounds like they might be still far then. Thanks!14:20
thiagoptsufiev: my idea was, once the policies for each service is loaded in the endpoint, is just a matter of setting the correct URI's for each policy on POLICY_FILES14:21
*** romainh has joined #openstack-horizon14:21
thiagoptsufiev: if the service itself isn't yet ready to use the endpoint, the config can remain on the policy copy on Horizon14:22
jpichamotoki: Great, thank you!! That will avoid confusion, some people have been filing bugs about missing translations :)14:22
tsufievthiagop, ah, got it! Once, say cinder, supports it, we could use the same OS_ENPOINT_POLICY value for the 'volume' endpoint and the '%(endpoint_id)s' will be interpolated to the cinder's one and cinder will get its policies from keystone (given they're already loaded there), right?14:26
thiagoptsufiev: exactly14:27
tsufievthiagop, i like it14:27
thiagoptsufiev: :D14:27
openstackgerritDoug Fish proposed a change to openstack/horizon: Make labels in Network Topology translatable  https://review.openstack.org/12927714:29
*** rwsu has joined #openstack-horizon14:31
openstackgerritSam Betts proposed a change to openstack/horizon: Pass correct project ID to get tenant_usages  https://review.openstack.org/12903714:35
*** absubram has joined #openstack-horizon14:52
*** akrivoka has joined #openstack-horizon14:52
*** Ala has quit IRC14:56
*** Ala has joined #openstack-horizon14:56
*** colinmcnamara has quit IRC15:03
*** ericpeterson has joined #openstack-horizon15:04
*** pbelanyi has quit IRC15:08
*** Ala has quit IRC15:11
*** hightall has quit IRC15:13
*** hightall has joined #openstack-horizon15:16
*** hightall has quit IRC15:21
*** Drago has joined #openstack-horizon15:21
*** Drago has quit IRC15:23
*** Drago has joined #openstack-horizon15:24
*** e0ne has quit IRC15:25
*** akrivoka has quit IRC15:26
*** sbfox has quit IRC15:27
*** hightall has joined #openstack-horizon15:33
*** lhcheng has joined #openstack-horizon15:33
*** lhcheng has quit IRC15:38
*** nlahouti has joined #openstack-horizon15:41
*** nlahouti1 has joined #openstack-horizon15:45
*** exploreshaifali has quit IRC15:45
*** nlahouti has quit IRC15:46
*** bpokorny has joined #openstack-horizon15:49
*** qba73 has quit IRC15:52
*** cedricss has quit IRC15:54
*** bpokorny has quit IRC15:55
*** cedrics has joined #openstack-horizon15:56
*** e0ne has joined #openstack-horizon15:59
*** e0ne has quit IRC15:59
*** e0ne has joined #openstack-horizon16:02
*** Dragomir has joined #openstack-horizon16:03
*** bpokorny has joined #openstack-horizon16:03
krotschecktsufiev: There may be.16:05
krotscheckAnd the state of storyboard is: “Effectively I’m the only person that’s working on it, so don’t get your hopes up.”16:05
*** e0ne has quit IRC16:07
*** Drago has quit IRC16:07
*** e0ne has joined #openstack-horizon16:13
*** e0ne has quit IRC16:14
*** lhcheng has joined #openstack-horizon16:14
*** lhcheng has quit IRC16:16
*** lhcheng has joined #openstack-horizon16:16
*** pkarikh has quit IRC16:18
*** AliceR has joined #openstack-horizon16:21
*** jpich has quit IRC16:22
*** ygbo has quit IRC16:26
*** cedrics has quit IRC16:26
lhchenghi ericpeterson16:29
ericpetersonhey lhcheng16:30
ericpetersonany updates on cookie session overflows ;) ?16:30
ericpetersonor let's just give up and go to a db backend :D16:31
*** ericpeterson has quit IRC16:32
*** ericpeterson has joined #openstack-horizon16:33
lhchengnope, I was hoping you guys already figured that out.16:34
ericpetersonkeep hoping :p16:34
lhchengmoving to file/db backend seems okay, in production deployment the operator needs to do it anyway.16:34
ericpetersonright now I have a one off patch to get my deployment going, but I'd really hope community can address this more completely16:34
lhchengone-off to use a db backend?16:35
ericpetersonone-off by not storing the allowed flags in the session16:35
*** Dragomir has quit IRC16:35
lhchenghmm thought we16:36
ericpetersonI will be moving to a db backed session soon, and at that point it becomes a non issue for me.... but it still is going to be an issue for many people getting started16:36
lhcheng*we'll revert it until a solution has been decided16:36
*** Dragomir has joined #openstack-horizon16:36
lhcheng?16:37
*** vokhrimenko has quit IRC16:37
*** EmilyW has joined #openstack-horizon16:37
ericpetersonthat would be ok with me, but I'm not the orig author.  that change had other things included also16:37
*** david-lyle has joined #openstack-horizon16:39
*** ttrifonov is now known as zz_ttrifonov16:41
*** sbfox has joined #openstack-horizon16:43
sambettsericpeterson: My vote is to change the default to a sqlite backend, that seems to be the neatest solution thats not just a bandaid for the issues16:43
*** Longgeek has quit IRC16:44
sambettsdavid-lyle, thiagop ^16:49
*** athomas has quit IRC16:51
*** exploreshaifali has joined #openstack-horizon16:51
ericpetersonyep, sambetts.  I am not strongly opposed or in favor.  I am mildly opposed to keeping that stuff on the session.   my vote is -0.12498016:52
ericpetersonI spent a lot of time on that math ;)16:52
ericpetersonI can see both sides16:52
lhchengsambetts: question on the patch, when does session['allowed']['valid_for']  get initially set?16:53
lhchenghttps://github.com/openstack/horizon/blob/master/horizon/base.py#L6316:53
*** ArcTanSusan has joined #openstack-horizon16:53
ericpetersonthe keyerror section does that lhcheng16:53
ericpetersonsession['allowed'] = {"valid_for": session.get('token')}16:53
david-lyleline 6616:53
ericpetersonit's just a bit weird to read the dict with two different methods of adding the key16:54
lhchengugh16:55
david-lylenon-obvious16:55
lhchengdavid-lyle: didn't notice that, raise an error to set a value16:55
david-lyleyeah16:55
ericpetersoncould use something like test for is in or something would be a bit clearer16:55
*** akrivoka has joined #openstack-horizon16:56
sambettsI did have a different solution in an earlier patch set than the one that merged, but one of the cores suggested that16:56
ericpetersonbut it works16:56
david-lyleof course they did ;)16:56
*** ArcTanSusan_ has joined #openstack-horizon16:56
*** AliceR has quit IRC16:57
*** akrivoka has quit IRC16:57
*** akrivoka has joined #openstack-horizon16:57
*** akrivoka has quit IRC16:57
lhchengsambetts: and after session['allowed']['valid_for'] is set, where is it used?16:58
lhchengsambetts: the cache used is in session['allowed'][key]16:58
lhchengsambetts: what am I missing?16:58
*** ArcTanSusan has quit IRC16:58
*** ArcTanSusan_ is now known as ArcTanSusan16:59
*** jprovazn has quit IRC16:59
ericpetersonsession['allowed'][key] = func(self, context)16:59
ericpetersonthat is used elsewhere. and the func is called later on, cached as needed17:00
ericpetersonthe valid_for key is really just testing existence, it seems17:00
lhcheng"key" is the classname of the panel?17:00
sambettslhcheng: in the except KeyError it completely wipes the whole cache and resets it with the new allowed token17:00
*** katyafervent is now known as katyafervent_awa17:01
lhchengsambetts: ah, got it now17:01
sambettsthe valid_for is only used to decide whether or not to invalidate the cache and therefore wipe it17:01
ericpetersonI am using this patch, http://paste.openstack.org/show/121487/  which seems like something similar17:01
*** ybathia has joined #openstack-horizon17:02
ericpetersonand simpler / doesn't overflow the cookie space17:02
*** hightall has quit IRC17:02
ericpetersonbut I would not be suprised if it is slightly slower17:02
sambettssurly the problem with memoizing it is that it could lead to cross talk between sessions?17:03
ericpetersonI think it never even memoizes, b/c it uses the unique context.... which varies from request to request17:04
ericpetersonI think17:04
ericpetersontaking out the memoize might be even more clear / safer17:04
ericpetersonthe allowed checks are pretty simple / lightweight in most cases anyways17:04
lhchengyeah, the memoized doesn't really help. since it only caches per request, and per request can_access is probably only called once per panel :)17:06
sambettsthe cache was to simply reduce the amount of network spam that occured when an allowed/can_access check hits an API for verifcation17:06
lhchengyou'll probably end up just consuming extra memory17:06
sambettscan_access checks that happen locally aren't to much of an issue, but ones that perform API requests were17:07
lhchengsambetts: I understand.   I was referring to the ducttape of ericpeterson to get around the issue.17:08
*** ArcTanSusan has quit IRC17:08
ericpetersonagreed, but caching in that case also introduce new issues of being out of date.  (if we cache to prevent network calls)17:08
ericpetersonI'd rather have those allowed checks figure out if they need caching and have a common approach at a lower level maybe..... might be another approach17:09
ericpetersonlike some new util function or something, or a decorator17:09
david-lyleif we memoized anything, it should be the API call itself17:09
david-lylenot the high level wrapper17:09
ericpetersonyeah, thats even better17:09
*** e0ne has joined #openstack-horizon17:10
sambettsericpeterson: A previous patch I had, had the decorator as something that got added by whoever was writing the can_access method, but amotoki decided that added more work for those extending horizon component17:10
sambettsthats why its in the form it is now, with the allowed and the can_access function17:11
sambettsheres the review with the evolution of the patch https://review.openstack.org/#/c/120790/17:12
ericpetersonyeah, I think the difference is that amotoki was thinking that having a global cache for the allowed stuff was a good idea, and we are not all in agreement.  If we agree that it should be cached, then I'd be on the same page with you 217:13
ericpetersonbut right now we are contemplating if that is correct or not, or even the best thing to be doing17:14
ericpetersonI'm not a core also, so amotoki counts more than I do :)17:15
sambettsI think the concern is that it happens for every page load that rebuilds the nav bar, tbh if we're moving towards a more angular based approach then I would expect the number of full page reloads to greatly reduce anyway17:16
david-lylesambetts, that's certainly the hope, but we need to make it widely usable in the sorter term17:16
david-lylewhat we have now blows up in real deployments17:17
*** ArcTanSusan has joined #openstack-horizon17:17
sambettsalso there was someone who was hoping that having a cache like that would provide an easy way to mock out which panels are displayed17:17
amotoki_I have not followed the whole discussion you all are discussing here...  what I asked sambetts in the past review is to clarify what is cached and what are not, and to avoid forcing subclass to ensure the decorator.17:17
amotoki_ericpeterson: you don't need to mind discussing. I am not a superman :-)17:17
david-lyleamotoki_, you're too humble ;)17:18
ericpetersonamotoki_:  yeah, I think that point / question was good.  and you pay more attention to the changes than I do17:18
ericpetersonI just know when stuff breaks :D17:18
ericpetersonbreaking things is something I should be famous for17:19
david-lyleit's good to have a core strength17:19
ericpetersonindeed.  I just feel threatened when other people break stuff first, before I can get to it17:19
sambettshaha17:20
thiagophahah17:20
*** tqtran has joined #openstack-horizon17:20
thiagopI'm back17:20
david-lyleI think we're likely to see the session size continue to grow, so I'm in favor of a db based session store by default, but I'm still not sure caching panel access is beneficial17:22
thiagopamotoki_: if you want some context: https://bugs.launchpad.net/horizon/+bug/138207917:22
david-lylememoizing API calls yes17:22
david-lyleif this is an optimization for testing, then I think we need to focus on testing implementation rather than adding complexity to support testing17:23
sambettsSo my stand point at this stage is that I can see that a lot of improvments/optimizations need to be made the caching (if its kept in) and then the session backend needs to change to a more sutiable storage solution17:23
*** amerine has joined #openstack-horizon17:24
david-lylesambetts, I'd like to see part two for sure. I think we can provide a much better user experience if we take that route17:25
ericpetersonwhich panels make network calls for can_access stuff???17:25
david-lyleL3 networking can be disabled which hides Routers17:25
ericpetersonbecause those panels could add memoization right away, to provide another level of caching17:25
david-lylehost aggregate is an extension, that may or may not be turned on17:25
sambettsLBaas and Firewall too17:26
david-lyleFirewalls, VPNs, LBaaS are all optional17:26
*** amerine_ has quit IRC17:26
ericpetersonand once those are cached, then we could maybe pull alot of this patch and simplify it17:26
ericpetersonwould that be acceptable?  we'd have the cache, and also simplify the session stuff17:26
david-lylenetwork extension support is already memoized IIRC17:27
ericpetersonyou mean it just works™ ?17:27
david-lylenova extension support API call should be too17:27
david-lylebut not sure should == is at this point17:27
ericpetersonI am 95% nova extension supports that, I might have added it17:27
ericpetersonsambetts what do you think about ensuring we memoize the network calls, then simplify this patch???17:28
david-lylehttps://github.com/openstack/horizon/blob/master/openstack_dashboard/api/nova.py#L83217:28
david-lyleit is17:28
*** sayali has quit IRC17:29
sambettsmy concern with memomizing all those requests is having the developers remember to add them when they extend the APIs17:29
sambettsbut thats a technicallity17:30
thiagoplet me just as a question: for what I understand of memoize just returns the same result when all arguments passed are equal to the ones used in the call before. Doesn't "request" always change?17:31
ericpetersonand memoizing / caching is something we add when pain point arrise17:31
amotoki_thiagop: thanks. Does the problem occur only when we use cookie-based session?17:31
ericpetersonit does, but sometimes we call the same api 5 times in the same request17:31
ericpetersonamotoki_ yes, db backed sessions do not have this problem17:32
thiagopamotoki_: yep. It doesn't occurs with cache/file/cached_db backends17:32
sambettsamotoki_: issues occur when combining this patch, keystone v3 and cookies17:32
ericpetersonwe use v2 tokens17:32
thiagopamotoki_: since they doesn't have a fixed lenght...17:32
amotoki_hmm... if so, the problem occurs when we have more keystone endpoints.17:32
ericpetersonif you have a fairly full svc catalog, and multiple regions.... it gets big17:33
amotoki_can_access cache reveals the potential issue just earlier.17:33
thiagopmany projects...17:33
ericpetersonyep17:33
*** mikedillion has joined #openstack-horizon17:33
*** bpokorny_ has joined #openstack-horizon17:34
amotoki_I am not sure what is a good assumption or requirement.. A first big question is whether we need to support cookie-based session or not.17:35
sambettsI think SQLite is a perfectly good alternative to cookie based system, from talking yesterday we only use cookies as the default for convience in devstack17:36
*** bpokorny has quit IRC17:37
*** tqtran_ has joined #openstack-horizon17:37
*** bpokorny has joined #openstack-horizon17:39
*** e0ne has quit IRC17:39
*** tqtran has quit IRC17:39
amotoki_i see. Anyway we need to clear session cache when a token is refreshed. If it is handled properly, I think using SQlite backend is not a problem because cache is cleared when user is logged in to a re-run devstack.17:40
*** clu_ has joined #openstack-horizon17:40
*** bpokorny_ has quit IRC17:41
sambettsI must head offline now as its nearly 18:45 where I am, but summarizing my thoughts: 1) Swap to SQLite as the default session store because cookies are going to become a pain point sooner or later, 2) Maybe re-work caching to operate at a lower level to reduce the quantity of requests made to the other services whilst being more explict about what is cached and what isnt17:46
*** ArcTanSusan has quit IRC17:48
thiagopanybody does have a point in favor of continuing using cookie based sessions? :P17:48
thiagopdoes anybody*17:48
sambettsthe only one that came up yesterday was for HA but we decided that at that point you'll be running a real DB anyway17:49
*** ericpeterson has quit IRC17:49
*** harlowja_away is now known as harlowja17:49
thiagopsambetts: +117:50
david-lyleswitching to db backed session store by default should go to the ML17:51
sambettsML?17:51
david-lyledev mailing list17:52
sambettsAh I see, yes :-)17:52
david-lyleimpacts more than just horizon17:52
david-lyledevstack, QA17:52
david-lyleoperators that may be watching17:52
david-lylewould be nice to get feedback on any concerns17:53
sambetts+117:55
amotoki_+117:56
*** ArcTanSusan has joined #openstack-horizon17:56
*** sayali has joined #openstack-horizon17:59
sambettsGood Night Horizon!17:59
*** sambetts has quit IRC17:59
*** bpokorny_ has joined #openstack-horizon18:00
*** bpokorny has quit IRC18:03
thiagopwho sends the e-mail to ML?18:03
david-lyleI suppose I can18:04
*** radez is now known as radez_g0n318:04
*** ericpeterson has joined #openstack-horizon18:06
*** radez_g0n3 is now known as radez18:07
*** bpokorny has joined #openstack-horizon18:07
*** bpokorny_ has quit IRC18:10
*** nlahouti1 has quit IRC18:11
thiagopdavid-lyle: I think people will give you more attention :)18:12
*** TravT has joined #openstack-horizon18:15
*** arborism has joined #openstack-horizon18:16
openstackgerritDavid Lyle proposed a change to openstack/horizon: Juno Release Notes  https://review.openstack.org/12874418:17
*** arborism is now known as amcrn18:17
*** ArcTanSusan has quit IRC18:20
*** sbfox1 has joined #openstack-horizon18:26
*** packet has quit IRC18:26
*** MaxV has quit IRC18:27
*** MaxV has joined #openstack-horizon18:27
openstackgerritA change was merged to openstack/horizon: Disable router HA attribute due to Neutron bug  https://review.openstack.org/12834118:28
*** sbfox has quit IRC18:28
*** Dragomir has quit IRC18:31
*** MaxV has quit IRC18:31
ericpetersonthiagop I'm good with switching session backends, there will just need to be some communication on how this impacts people18:36
*** cody-somerville_ has joined #openstack-horizon18:40
*** cody-somerville has quit IRC18:41
*** amerine has quit IRC18:47
*** amerine has joined #openstack-horizon18:47
*** ArcTanSusan has joined #openstack-horizon19:06
*** ArcTanSusan has quit IRC19:11
openstackgerritDoug Fish proposed a change to openstack/horizon: Make labels in Network Topology translatable  https://review.openstack.org/12927719:12
*** bpokorny has quit IRC19:15
openstackgerritA change was merged to openstack/horizon: Pass correct project ID to get tenant_usages  https://review.openstack.org/12903719:16
*** ArcTanSusan has joined #openstack-horizon19:16
*** radez is now known as radez_g0n319:17
*** tosky has quit IRC19:17
*** radez_g0n3 is now known as radez19:18
*** jpomero_ has joined #openstack-horizon19:18
*** jpomero has quit IRC19:21
*** MaxV has joined #openstack-horizon19:27
*** jcoufal has quit IRC19:29
*** cody-somerville_ has quit IRC19:31
*** MaxV has quit IRC19:32
openstackgerritA change was merged to openstack/horizon: Improve sub-navigation styling  https://review.openstack.org/12824719:32
openstackgerritDoug Fish proposed a change to openstack/horizon: Fix bad performance on project update panel  https://review.openstack.org/11948719:37
*** colinmcnamara has joined #openstack-horizon19:38
*** e0ne has joined #openstack-horizon19:44
*** TravT has quit IRC19:58
openstackgerritA change was merged to openstack/horizon: Remove selenium dependency when not using selenium tests  https://review.openstack.org/12677720:02
*** jacalcat has joined #openstack-horizon20:11
openstackgerritDavid Lyle proposed a change to openstack/horizon: Replaces always disabled context selection link  https://review.openstack.org/12937920:20
*** MaxV has joined #openstack-horizon20:20
*** sayali has quit IRC20:23
*** david-lyle has quit IRC20:24
*** MaxV has quit IRC20:25
*** crobertsrh is now known as _crobertsrh20:28
*** ybathia has quit IRC20:34
*** bpokorny has joined #openstack-horizon20:34
*** e0ne has quit IRC20:50
*** doug-fish has left #openstack-horizon20:55
*** rwsu has quit IRC20:56
*** bpokorny has quit IRC20:57
*** nlahouti has joined #openstack-horizon20:59
*** sbfox has joined #openstack-horizon20:59
*** colinmcnamara has quit IRC20:59
*** lhcheng has quit IRC20:59
*** lhcheng has joined #openstack-horizon21:00
*** sbfox1 has quit IRC21:00
*** radez is now known as radez_g0n321:01
*** romainh has left #openstack-horizon21:04
*** ayoung has joined #openstack-horizon21:24
*** bpokorny has joined #openstack-horizon21:26
*** TravT has joined #openstack-horizon21:29
*** colinmcnamara has joined #openstack-horizon21:32
*** lhcheng has quit IRC21:43
*** lhcheng has joined #openstack-horizon21:44
*** TravT has quit IRC21:45
*** lhcheng has quit IRC21:46
*** colinmcnamara has quit IRC21:48
*** exploreshaifali has quit IRC21:48
*** lhcheng has joined #openstack-horizon21:49
*** colinmcnamara has joined #openstack-horizon21:51
*** ayoung has quit IRC21:54
*** jpomero_ is now known as jpomero21:54
*** peristeri has quit IRC21:54
*** ybathia has joined #openstack-horizon21:55
*** lhcheng_ has joined #openstack-horizon21:57
*** rebelagentm has quit IRC21:59
*** openstackgerrit has quit IRC22:03
*** openstackgerrit has joined #openstack-horizon22:05
*** lhcheng_ has quit IRC22:05
*** lhcheng has quit IRC22:05
*** lhcheng_ has joined #openstack-horizon22:05
*** lhcheng has joined #openstack-horizon22:05
*** lhcheng has quit IRC22:05
*** colinmcnamara has quit IRC22:10
*** e0ne has joined #openstack-horizon22:11
*** e0ne has quit IRC22:11
*** mikedillion has quit IRC22:12
*** amerine has quit IRC22:12
*** alexpilotti has quit IRC22:20
*** e0ne has joined #openstack-horizon22:26
*** mikedillion has joined #openstack-horizon22:29
*** e0ne has quit IRC22:36
*** clu_ has quit IRC22:36
*** sigmavirus24 is now known as sigmavirus24_awa22:36
*** ericpeterson has quit IRC22:39
esphello, anyone run into a django css compressor issue in devstack before?22:49
espFilterError: /bin/sh: 1: django_pyscss.compressor.DjangoScssFilter: not found22:50
*** rwsu has joined #openstack-horizon22:52
*** mikedillion has quit IRC22:58
*** EmilyW has quit IRC23:06
*** tqtran_ has quit IRC23:12
*** ArcTanSusan has quit IRC23:16
*** colinmcnamara has joined #openstack-horizon23:17
*** ArcTanSusan has joined #openstack-horizon23:19
*** Dragomir has joined #openstack-horizon23:21
*** sbfox has quit IRC23:25
*** colinmcnamara has quit IRC23:32
*** colinmcnamara has joined #openstack-horizon23:35
*** ArcTanSusan has quit IRC23:45
*** colinmcnamara has quit IRC23:45
*** ArcTanSusan has joined #openstack-horizon23:51
*** david-lyle has joined #openstack-horizon23:52
*** bpokorny has quit IRC23:52
*** david-lyle has quit IRC23:52
*** colinmcnamara has joined #openstack-horizon23:53

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