Friday, 2015-05-29

« Thursday, 2015-05-28 Index Saturday, 2015-05-30 »
woodster_ptoohill: are you just trying to develop with Barbican locally? If so you can run in the standalone mode with a queue00:01
ptoohillwoodster_: I set it up with devstack, I would have to look into that. I was getting errors about user_id none from the queue. Is there somethign known possibly?00:03
ptoohillI would have to look into it more to see if thats the real error or i had something else going on but i could not reg a consumer.00:03
*** bharath has quit IRC00:09
woodster_ptoohill: queues should only be needed for orders, not containers and consumers00:12
ptoohillThen its possible im having other issues but the logs were filled with those errors. Ill dig into it more tonight.00:13
*** mixos has joined #openstack-lbaas00:27
*** mixos is now known as mixos-away00:27
*** mixos-away is now known as mixos00:28
madhu_akmixos, xgerman: sorry, didn't notice your messages.00:33
*** SumitNaiksatam has quit IRC00:52
*** madhu_ak has quit IRC01:11
*** hitalia has quit IRC01:35
*** amotoki has joined #openstack-lbaas01:46
*** amotoki_ has joined #openstack-lbaas01:47
*** bradjones has quit IRC01:52
*** amotoki_ has quit IRC01:52
*** bradjones has joined #openstack-lbaas01:56
*** kiran-r has joined #openstack-lbaas02:20
*** kiran-r has quit IRC02:56
*** kiran-r has joined #openstack-lbaas02:57
*** madhu_ak has joined #openstack-lbaas02:59
*** vivek-ebay has quit IRC02:59
*** kiran-r has quit IRC03:03
*** bharath has joined #openstack-lbaas03:10
*** barra204 has quit IRC03:11
*** barra204 has joined #openstack-lbaas03:13
*** bharath has quit IRC03:14
*** amotoki_ has joined #openstack-lbaas03:36
openstackgerritDoug Wiegley proposed openstack/neutron-lbaas: Pull lbaasv1 tempest tests in-tree, and run v1/v2 tests directly  https://review.openstack.org/18621903:40
*** amotoki_ has quit IRC03:41
*** TrevorV|Home has joined #openstack-lbaas03:46
openstackgerritDoug Wiegley proposed openstack/neutron-lbaas: Pull lbaasv1 tempest tests in-tree, and run v1/v2 tests directly  https://review.openstack.org/18621903:53
*** mixos has left #openstack-lbaas03:54
*** TrevorV_ has joined #openstack-lbaas04:00
*** TrevorV_ has quit IRC04:00
*** TrevorV|Home has quit IRC04:06
*** ajmiller has quit IRC04:12
*** vivek-ebay has joined #openstack-lbaas04:13
*** amotoki has quit IRC04:15
*** amotoki has joined #openstack-lbaas04:37
*** madhu_ak has quit IRC04:55
*** Kunal has joined #openstack-lbaas05:02
*** LouieC has joined #openstack-lbaas05:13
*** LouieC has left #openstack-lbaas05:13
*** kiran-r has joined #openstack-lbaas05:23
*** amotoki_ has joined #openstack-lbaas05:25
*** amotoki_ has quit IRC05:30
*** SumitNaiksatam has joined #openstack-lbaas05:35
*** madhu_ak has joined #openstack-lbaas05:36
*** vivek-ebay has quit IRC05:46
*** kiran-r has quit IRC05:58
openstackgerritDoug Wiegley proposed openstack/neutron-lbaas: Pull lbaasv1 tempest tests in-tree, and run v1/v2 tests directly  https://review.openstack.org/18621906:11
*** Miouge_ has joined #openstack-lbaas06:18
*** Miouge has quit IRC06:20
*** Miouge_ is now known as Miouge06:20
*** madhu_ak has quit IRC06:24
*** amotoki_ has joined #openstack-lbaas06:41
*** amotoki_ has quit IRC06:45
*** woodster_ has quit IRC06:50
*** Miouge_ has joined #openstack-lbaas07:02
*** Miouge has quit IRC07:03
*** Miouge_ is now known as Miouge07:03
*** kiran-r has joined #openstack-lbaas07:04
openstackgerritPhillip Toohill proposed stackforge/octavia: Hooking TLS together  https://review.openstack.org/18554007:09
*** Kunal has quit IRC07:14
*** Miouge has quit IRC07:25
*** Miouge has joined #openstack-lbaas07:33
*** apuimedo has joined #openstack-lbaas08:19
*** apuimedo has quit IRC08:27
*** ByPasS has quit IRC08:44
*** Miouge has quit IRC08:58
*** Miouge has joined #openstack-lbaas09:01
*** Miouge has quit IRC09:01
*** amotoki_ has joined #openstack-lbaas10:17
*** amotoki_ has quit IRC10:22
*** apuimedo has joined #openstack-lbaas10:29
*** apuimedo has quit IRC10:38
*** amotoki has quit IRC10:53
*** apuimedo has joined #openstack-lbaas10:55
*** Miouge has joined #openstack-lbaas10:59
*** amotoki has joined #openstack-lbaas11:49
*** Miouge has quit IRC11:52
*** amotoki has quit IRC11:54
*** kiran-r has quit IRC12:05
*** kiran-r has joined #openstack-lbaas12:05
*** kiran-r has quit IRC12:16
*** amotoki has joined #openstack-lbaas12:49
*** woodster_ has joined #openstack-lbaas12:50
*** apuimedo has quit IRC12:52
*** amotoki has quit IRC12:54
*** apuimedo has joined #openstack-lbaas13:09
*** mixos has joined #openstack-lbaas13:49
*** jhova has joined #openstack-lbaas13:50
*** mixos has quit IRC13:56
*** apuimedo has quit IRC14:00
*** apuimedo has joined #openstack-lbaas14:16
*** amotoki has joined #openstack-lbaas14:38
*** amotoki has quit IRC14:43
*** vivek-ebay has joined #openstack-lbaas14:46
*** ajmiller has joined #openstack-lbaas14:48
*** amotoki has joined #openstack-lbaas14:52
*** TrevorV|Home has joined #openstack-lbaas15:04
*** apuimedo has quit IRC15:07
*** apuimedo has joined #openstack-lbaas15:24
TrevorV|HomeI think devstack is broken again...15:35
TrevorV|HomeWhy this always happen when I'm in a time crunch?15:35
*** apuimedo has quit IRC15:38
ptoohillTrevorV|Home: what is it doing?15:43
TrevorV|HomeFailing.15:43
TrevorV|HomeI'll gist you15:43
TrevorV|Homeptoohill, https://gist.github.com/interminator/2fc03730b31de01573cd15:43
ptoohilldo you see other 500's by chance?15:44
ptoohillcheck the git status of glance/other projects. If you have a bunch of req files not commited youre in bad state. Blow away repos and start fresh :/15:45
ptoohillits been doing that more often for me lately'15:45
TrevorV|HomeI did that ptoohill15:45
ptoohillwell i dont know if that error is in/from glance itself or like octavia trying to find image15:46
TrevorV|HomeI did it from just a "git pull" on devstack originally, then I removed the virtualenv on the directory I was stacking in, and then I nuked the /opt/stack/* stuffs and tried again15:46
TrevorV|HomeAll failed with same error15:46
ptoohillpossible the image didnt get built/uploaded15:46
*** mixos has joined #openstack-lbaas15:46
*** mixos is now known as mixos-away15:46
ptoohillhmm15:46
TrevorV|HomeBut the "cirros" one should have regardless, right?15:47
ptoohilloctavia uses special one15:47
ptoohillwould need more context to know if this is 'global' error or specific to our image15:47
TrevorV|HomeYes, I know, but I don't see any trace of it dying specifically in octavia, and it should at least show ONE image, when it shows none15:47
ptoohilltrue15:48
*** mixos-away is now known as mixos15:48
ptoohilli just know when i was seeing the unrecognized args (for multiple projects) i wiped /opt/stack got latest master devstack and re ran. everything worked after that. but its done it a few times now15:49
ptoohilli dont know what causes this :/15:50
TrevorV|HomeI've done that as a "standard debug step" every time I run into errors and come running to others for help ha ha.15:50
ptoohillI havnt had to debug beyond that, i wont be of much help then15:51
ptoohillsorry15:51
johnsomI saw that they were moving the cirros image location to here: http://download.cirros-cloud.net/  Not sure what you have in your localrc15:51
johnsomI think 0.3.0 is the last image in launchpad15:52
ptoohillits complaining about bad args though. And when i get this its for multiple projects15:52
ptoohillhopefully its just an image location issue for him15:53
TrevorV|Homejohnsom, thanks for the heads up15:53
TrevorV|HomeI have the 0.3.0 specifically in my localrc though, so it probably shouldn't fail there15:53
ptoohillthe error, to me, means the clients/project are out of sync somehow15:54
ptoohillopenstack image list: error: unrecognized arguments: --property status=active15:54
ptoohillmay need to specifically wipe the installed clients since they arent installed seperately from devstack15:55
ptoohillare*15:55
TrevorV|Homedoes that not happen when I run "./clean.sh"?15:55
ptoohillnot sure, i dont think so15:55
TrevorV|Homewell since I've never seen "openstack image list" as a command in general until this morning, I'd say its probably installed via devstack15:56
ptoohilldevstack does install it yes, but its not installed in /opt/stack15:56
ptoohillthough, barbican is15:56
ptoohilland i pull them down if im working on one specifically15:56
ptoohillTheres a place in the docs for devstack regarding the installtion of clients15:57
ptoohilleh, neutronclient is there too. unless i pulled that one. cant remember. but the others are not15:58
*** vivek-eb_ has joined #openstack-lbaas16:02
*** vivek-ebay has quit IRC16:04
TrevorV|HomeI nuked all the clients16:09
TrevorV|HomeI'm re-stacking now16:09
TrevorV|HomeSee if that helps16:09
*** vivek-eb_ has quit IRC16:14
*** mixos is now known as mixos-away16:23
*** mixos-away is now known as mixos16:23
*** amotoki has quit IRC16:24
openstackgerritDoug Wiegley proposed openstack/neutron-lbaas: Pull lbaasv1 tempest tests in-tree, use installed neutron for v1/v2 job  https://review.openstack.org/18621916:28
openstackgerritDoug Wiegley proposed openstack/neutron-lbaas: Pull lbaasv1 tempest tests in-tree, use installed neutron for v1/v2 job  https://review.openstack.org/18621916:29
*** kiran-r has joined #openstack-lbaas16:30
*** madhu_ak has joined #openstack-lbaas16:33
*** vivek-ebay has joined #openstack-lbaas16:36
*** mlavalle has joined #openstack-lbaas16:53
*** TrevorV|Home has quit IRC16:56
*** hitalia has joined #openstack-lbaas16:58
*** TrevorV|Home has joined #openstack-lbaas16:58
*** SumitNaiksatam has quit IRC17:19
*** SumitNaiksatam has joined #openstack-lbaas17:20
*** mixos is now known as mixos-away17:25
*** barra204 has quit IRC17:26
TrevorV|HomeSorry, for anyone who followed the conversation earlier, after I manually removed all the openstack related python clients, the stack worked successfully, but I may have found a bug in the "./unstack.sh" process17:27
*** kiranr has joined #openstack-lbaas17:31
*** SumitNaiksatam has quit IRC17:31
*** SumitNaiksatam has joined #openstack-lbaas17:33
*** kiran-r has quit IRC17:35
*** kiranr has quit IRC17:41
*** kiran-r has joined #openstack-lbaas17:42
*** kiran-r has quit IRC17:46
rm_workheh, unstack17:55
rm_worksomething i've never actually bothered running :P17:55
*** bharath has joined #openstack-lbaas17:56
johnsomYeah, I have moved on to just using clean.sh17:57
rm_worki still just re-image and stack17:58
rm_workguaranteed clean :P17:58
*** mixos-away is now known as mixos17:59
*** SumitNaiksatam has quit IRC18:05
*** SumitNaiksatam has joined #openstack-lbaas18:06
*** barra204 has joined #openstack-lbaas18:08
*** kiran-r has joined #openstack-lbaas18:11
*** kiran-r has quit IRC18:11
*** barra204 has quit IRC18:15
*** barra204 has joined #openstack-lbaas18:20
openstackgerritAishwarya Thangappa proposed openstack/neutron-lbaas: Tempest tests for Listener using testscenarios.  https://review.openstack.org/17981818:20
TrevorV|Homejohnsom, I was running this:  "./unstack.sh && ./clean.sh" and it still didn't clean up everything that was causing a conflict18:22
*** crc32 has joined #openstack-lbaas18:24
*** barra204_ has joined #openstack-lbaas18:27
*** crc32 has quit IRC18:30
*** crc32 has joined #openstack-lbaas18:30
*** barra204 has quit IRC18:31
ptoohillTrevorV|Home: clean.sh calls unstack18:34
TrevorV|Home....srsly.... I'm dumb18:34
ptoohill:P18:34
*** crc32 has quit IRC18:35
*** crc32 has joined #openstack-lbaas18:40
*** bharath has quit IRC18:49
*** barra204_ is now known as shakamunyi18:52
*** madhu_ak has quit IRC19:02
*** crc32 has quit IRC19:10
*** ajmiller has quit IRC19:12
*** ajmiller has joined #openstack-lbaas19:13
openstackgerritSungjin Yook proposed openstack/neutron-lbaas: Add a new lbaas agent Scheduler, LeastPoolAgentScheduler  https://review.openstack.org/18160919:15
*** madhu_ak has joined #openstack-lbaas19:16
openstackgerritSungjin Yook proposed openstack/neutron-lbaas: Add a new lbaas agent Scheduler, LeastPoolAgentScheduler  https://review.openstack.org/18160919:17
*** madhu_ak has quit IRC19:42
ajmillerTrevorV|Home if clean.sh is leaving something behind that is causing a conflict, we should look into it and get it fixed.  It should (and used to) work.19:50
TrevorV|HomeWell ajmiller it doesn't remove a single client.19:51
TrevorV|HomeIf it INSTALLS the clients with ./stack.sh, then I'd argue it should remove the clients when you do a clean.sh19:51
TrevorV|HomeWhich apparently isn't happening19:52
ajmillerYes, I agree that is what it should do.19:52
ajmillerBTW, I have found/fixed a couple of bugs in devstack clean.sh that were a result of the plugin decomposition.  Would not surprise me if there are more.19:53
TrevorV|HomeYeah, well, at this point I don't really ever expect devstack to be completely function at any given time.  It "should" be, but I run into issues all the time19:54
TrevorV|Homefunctional***19:54
ajmillerLOL19:55
ajmillerYeah, it can be a bit fragile.  They've been doing a ton of refactoring, and I believe they have some more changes coming19:56
*** SumitNaiksatam has quit IRC19:59
rm_worki will repeat the (probably unhelpful) advice I give all the time: nuke it from orbit19:59
rm_workwhich requires setting it up in such a way that you can do that20:00
rm_workIE, not local dev :(20:00
TrevorV|Homerm_work, not interested in NOT local dev20:03
*** madhu_ak has joined #openstack-lbaas20:03
*** SumitNaiksatam has joined #openstack-lbaas20:04
*** openstack has joined #openstack-lbaas20:05
johnsomIt's the only way to be sure....20:05
ptoohillany thoughts on the consumers?20:05
TrevorV|HomeIts also NOT an appropriate solution for a fragile environment.20:05
rm_workptoohill: ugh, i think we just have to fake it for Octavia, and just do a register call in the driver20:07
rm_workptoohill: though, lets make failing that call a hard-stop error20:07
rm_workerr, no-continue20:07
rm_workyou know what i mean20:07
*** SumitNaiksatam has quit IRC20:11
ptoohillyea20:11
*** SumitNaiksatam has joined #openstack-lbaas20:11
ptoohillThis means we are getting the barbican data several times then..20:12
rm_workwell, only the container20:12
rm_workwhich is just one single call20:12
rm_worknot too bad20:12
ptoohill3 times..20:12
rm_workit isn't actually pulling any of the secrets20:12
rm_workor, it should not20:13
ptoohillit does...20:13
rm_workwhy?20:13
ptoohillit validates the data on the first go20:13
ptoohillthen driver would be second, that shouldnt need to20:13
rm_worki thought validate was manual20:13
ptoohillthen third to use the data20:13
ptoohillmanual?20:13
rm_workyou had to do .validate()20:13
rm_workor something20:13
rm_workit lazy-loads the data20:13
rm_workso unless you actually try to fetch the key/cert/whatever20:14
rm_workit doesn't fetch them20:14
ptoohillyea? It loads the certificate data and uses the tls_utils to verify certificates20:14
rm_workbut only if you tell it to20:14
ptoohillwhat would we be validating if we didnt validate the certs20:14
rm_worknot just on the first get20:14
ptoohillyea...20:14
rm_workso, neutron-lbaas validates the stuff as part of the user's API call20:14
ptoohillwould be nice if barbican did the validation20:14
ptoohillyes20:14
rm_work(which does load everything)20:14
ptoohillyes20:14
rm_workthen it passes it to the driver20:14
ptoohillcorrect20:14
rm_workthe driver should just be able to do the register call20:15
ptoohillnot the data, just the ref20:15
rm_workwhich will only be one single http hit20:15
ptoohilland then octavia will also load everything20:15
rm_workptoohill: err, it should be validating the data on the first neutron-lbaas hit20:15
ptoohillthis is 8 calls for one process20:15
rm_workptoohill: ok, but it's only one more than current20:15
rm_workis what i'm saying20:15
ptoohillfair enough20:15
rm_workright now neutron-lbaas does check_only (container, 2-4 pieces of data, total of 3-5 calls to barbican), then octavia does a full pull to set up the amp (same 3-5)20:16
rm_workso really now it's 6-1020:16
rm_workif we add a register call in the octavia driver, it SHOULD be 7-1120:17
rm_workwhich is ... yeah, not ideal20:17
rm_workbut the other option is passing only the data, not the container-id20:17
rm_workso octavia doesn't hit barbican at all20:17
ptoohillwhat about having the needed data passed to octavia somehow?20:17
rm_workwhich puts us back to having duplicate customer data in octavia20:17
rm_workbecause then we have to store everything that comes in20:18
rm_workon our own account20:18
ptoohilldo we HAVE to20:18
rm_workyes20:18
rm_workbecause we no longer have a barbican-id to work with20:18
ptoohilltrue20:18
rm_workwe have just the raw cert/pk data20:18
rm_workso it doesn't really solve the problem :/20:18
ptoohillyea, hrmph20:18
rm_workbut, this is the amount of calls i already cleared with them20:19
rm_work(the barbican team)20:19
ptoohillok, ill pull check out of octavia and work on driver20:19
rm_workit isn't pretty, but... yeah, the only way to fix it would be if barbican could validate20:19
rm_workwhich would be great, but they don't want to do at the moment :(20:19
ptoohillpeople were confused when i brought that up20:20
ptoohillit should be something they do...20:20
openstackgerritSungjin Yook proposed openstack/neutron-lbaas: Add a new lbaas agent Scheduler, LeastPoolAgentScheduler  https://review.openstack.org/18160920:20
rm_workptoohill: i have asked before20:21
rm_workit seems straight forward20:21
rm_workbut there are a couple of issues20:21
ptoohillI would have thought so20:21
rm_workone, it adds a lot of load to their API calls if there is validation at that layer20:21
ptoohillbut they are the secret store20:21
ptoohillhow hard are they going to be hit for one20:22
rm_workand second, it may not even be POSSIBLE, because if they use transport key encryption, their API may not even be able to see the secrets20:22
ptoohillfor two.. that should be their job20:22
ptoohilloh :/20:22
rm_workyeah20:22
rm_work:(20:22
ptoohillwell dern20:22
rm_workso the KMIP/HSM device would need to somehow support doing validation internally20:22
rm_workwhich... i don't think is the case20:23
rm_worksince from the perspective of those devices, the secrets are just blobs20:23
rm_workIIRC20:23
ptoohillI was questioned at the summit... 'what good is a keystore if they dont validate for you'20:23
rm_workexcepting a couple of special cases, but barbican has to be generic enough to work with any backend20:23
rm_workwhat good is a keystore if it isn't secure :P20:23
ptoohill><20:23
rm_workand validating the data en-route is basically saying it isn't secure20:23
ptoohillso we make the data insucre by validating it?20:24
rm_workbecause you are by definition a 3rd-party intercepting the secret data :P20:24
ptoohillsame as the backends then20:24
rm_workbarbican isn't actually the keystore20:24
ptoohillshould we never validate20:24
rm_workthe HSM is a keystore20:24
ptoohilltrue, good point20:24
rm_workbarbican is just an API for the keystore :(20:24
rm_workptoohill: it's possible we SHOULDN'T20:24
rm_workjust fuck it20:25
rm_workand send the thing through20:25
ptoohill:P20:25
rm_workif haproxy fails... that is the user's problem for specifying a bad config20:25
rm_workif it can surface that the error is a bad ssl config -- and we can then pass that on to the user -- then perfect20:25
rm_workif that isn't possible (it might not be) then it's a little less clear cut20:25
ptoohillthough, thats aync and they wont know unless they query some sort of logs20:26
ptoohillor events20:26
rm_workright20:26
rm_workthey'd have to wait while the LB is in PENDING_CREATE20:26
rm_workand then query until it gos ACTIVE20:26
rm_workand when it goes to ERROR20:26
ptoohillyea20:26
rm_workwe'd need to have some detail20:26
ptoohill:/20:26
ptoohillHRMMMM20:26
ptoohillI suppose for now, the extra calls arent too bad. Nor do i think they ever will be. But think this could be better20:27
woodster_just noting that Barbican is starting to do some validation based on secret type. I don't think we'd be opposed to do so for the cert type actually, but probably need help to pull that off :)20:27
ptoohillthat does sound like a fun problem ;)20:28
rm_workwoodster_: how much of the backlog did you read through? :)20:28
*** openstack has joined #openstack-lbaas20:29
ptoohillyea, would happen 'once' (7-11 calls) per update/create of listener.20:30
woodster_but ideally the validation happens when the customer puts the cert in in the first place (or barbican generates it)...then no need to validate thereafter since secrets are immutable...correct? Just making sure I'm understanding 'ideal' here20:32
ptoohillThat would be the ideal situation woodster_20:32
ptoohillThen for our case we would just need to reg a consumer, then gather needed data in backend20:32
rm_workwoodster_: yep20:34
woodster_sounds cool. For transport key of a cert-typed secret, I'd expect them to do some validation, but need to feed that back to contribs20:34
rm_workwoodster_: i thought you guys couldn't intercept the data in your API because of transport-key encryption support20:34
rm_workah20:34
*** SumitNaiksatam has quit IRC20:50
*** SumitNaiksatam has joined #openstack-lbaas20:53
*** openstackgerrit has quit IRC20:59
*** openstackgerrit has joined #openstack-lbaas21:00
*** jorgem has joined #openstack-lbaas21:14
*** shakamunyi has quit IRC21:26
*** jhova has quit IRC21:29
*** openstack has joined #openstack-lbaas21:31
madhu_akdougwig, blogan : anyone around?21:42
bloganmadhu_ak: here21:42
madhu_ak(thumbsup)21:42
openstackgerritTrevor Vardeman proposed stackforge/octavia: Update PUT methods to send update through queue  https://review.openstack.org/17098921:43
madhu_akthere is a patch for flavor-framework (https://review.openstack.org/#/c/139758) and upon migrating the flavor version onto neutron master branch, when I create a flavor with right attributes, I end up getting a problem.21:46
madhu_akblogan ^^21:46
madhu_akhere is what I have tried so far: https://gist.github.com/akmadhusudhan/b712c7a7a9590fd2d75b21:47
bloganyou're rebasing it onto neutron master?21:47
madhu_akyep..21:47
madhu_akrebasing is already done21:47
madhu_akand the unit tests seems passing for now...right now, I plan to write neutron api tests for flavor21:48
bloganhmm tenant_id is obviously not being passed in21:48
madhu_akwell, I did try specifying tenant_id in the body, it didn't help :(21:49
bloganwhat kind of test is this?21:49
blogan"unit" or tempest?21:49
blogani.e. doesn't require keystone21:49
madhu_akneutron api tests21:49
*** TrevorV|Home has quit IRC21:49
*** bharath has joined #openstack-lbaas21:50
bloganso thats probably tempest21:50
blogantox -e api?21:50
madhu_akyes21:51
bloganyou have devstack running?21:51
madhu_akyep21:51
bloganhmm21:51
bloganwell it should grab the tenant_id from the token21:51
bloganobviously _verify_attributes is expecting tenant_id to be in there21:51
blogango up the stack and try to pinpoint where the tenant_id should be injected21:52
madhu_akI suspect populate_tenant_id seems broken ...I am *unsure* about this21:52
bloganyeah this is just a troubleshooting job21:53
bloganfiguring out where it broke and why21:53
madhu_akaahh21:53
*** bharath has quit IRC21:54
madhu_akhow you meant this line -- go up the stack and pinpoint21:54
madhu_aksomehow, I could not think of debugging it in pycharm21:55
blogani mean the stack trace there21:55
bloganyeah i don't know what hte problem is so debugging woudl be what i would do21:55
bloganyou end up learning a lot more as you do it too21:56
madhu_akhmm, thats the right thing..by the way, if I dont need to use verify_attributes, I am able to do POST,PUT,DELETE the flavors though21:57
madhu_akIn that case, it is not validating the tenant_id?21:58
bloganprobably21:59
bloganthough with admin its not really doing much validation21:59
madhu_akyeah, one has to be be an admin, when creating flavor..22:00
madhu_akby the way, Isnt the correct way when an admin wants to don something (assuming he knows what needs to do)22:01
madhu_aklike admin dont need to validate...22:01
*** ptoohill_ has joined #openstack-lbaas22:10
bloganyou mean give admin total control?22:14
blogandon't validate the data at all?22:14
xgermanwell, I was objecting to not validating the admin input —22:14
xgermanso let’s not have made_ak go against the party line ;-)22:14
bloganxgerman: you mean madhu_ak has to be part of the HP hive mind?22:15
xgermanyep22:15
madhu_ak:)22:15
bloganresistance is futile22:16
*** jorgem has quit IRC22:29
openstackgerritSungjin Yook proposed openstack/neutron-lbaas: Add a new lbaas agent Scheduler, LeastPoolAgentScheduler  https://review.openstack.org/18160922:29
*** jorgem has joined #openstack-lbaas22:40
*** jorgem has quit IRC22:58
mixosblogan: resistance is futile ==> my favorite line from Star Trek23:03
*** SumitNaiksatam has quit IRC23:04
*** madhu_ak has quit IRC23:10
*** mixos has quit IRC23:39
*** mixos has joined #openstack-lbaas23:41
*** mixos has left #openstack-lbaas23:46
*** ajmiller has quit IRC23:54
« Thursday, 2015-05-28 Index Saturday, 2015-05-30 »

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