Thursday, 2015-04-30

openstackgerritAishwarya Thangappa proposed openstack/neutron-lbaas: Tempest tests using testscenarios for neutron-lbaas
*** mlavalle has quit IRC00:32
*** hitalia has quit IRC00:47
*** Aish has quit IRC00:51
*** fnaval has quit IRC00:59
*** fnaval has joined #openstack-lbaas01:17
*** rm_you| has joined #openstack-lbaas01:19
*** rm_you has quit IRC01:20
*** bharath has joined #openstack-lbaas01:49
*** SumitNaiksatam has joined #openstack-lbaas01:52
*** bharath has quit IRC01:53
*** SumitNaiksatam has quit IRC02:03
*** SumitNaiksatam has joined #openstack-lbaas03:39
*** Aish has joined #openstack-lbaas04:24
*** openstackgerrit has quit IRC05:22
*** openstackgerrit has joined #openstack-lbaas05:22
*** jschwarz has joined #openstack-lbaas05:35
openstackgerritPhillip Toohill proposed stackforge/octavia: Add devstack plugin for octavia
*** kiran-r has joined #openstack-lbaas05:40
*** kiran-r has quit IRC06:14
*** [1]evgenyf has joined #openstack-lbaas06:33
*** kiran-r has joined #openstack-lbaas06:40
*** kiran-r has quit IRC06:46
*** kiran-r has joined #openstack-lbaas06:55
*** chlong has quit IRC07:42
*** kobis has joined #openstack-lbaas07:49
*** woodster_ has quit IRC07:50
*** kiran-r has quit IRC08:03
*** kobis has quit IRC08:31
*** kobis has joined #openstack-lbaas08:31
*** kiran-r has joined #openstack-lbaas08:56
*** openstackgerrit_ has joined #openstack-lbaas10:16
*** openstackgerrit_ has quit IRC10:31
*** apuimedo has joined #openstack-lbaas10:32
*** kiran-r has quit IRC11:15
*** kiran-r has joined #openstack-lbaas11:37
*** kiran-r has quit IRC11:54
*** kiranr has joined #openstack-lbaas11:54
*** woodster_ has joined #openstack-lbaas11:58
*** kiranr has quit IRC11:59
*** Aish has quit IRC13:35
openstackgerritShang Yong proposed openstack/neutron-lbaas: Fix a pid file name error Closes-bug: 1450474
openstackbug 1450474 in neutron "destory a loadbalancer will not kill haproxy process" [Undecided,New] - Assigned to Brian Shang (brian-shang-dev)13:40
*** ajmiller has joined #openstack-lbaas13:52
*** chlong has joined #openstack-lbaas13:56
*** [1]evgenyf has quit IRC14:25
*** evgenyf has joined #openstack-lbaas14:36
*** fnaval has quit IRC14:42
*** chlong has quit IRC14:53
*** mlavalle has joined #openstack-lbaas14:53
*** fnaval has joined #openstack-lbaas14:55
*** fnaval has quit IRC14:58
*** TrevorV_ has joined #openstack-lbaas15:00
*** fnaval has joined #openstack-lbaas15:02
*** TrevorV_ has quit IRC15:03
*** jorgem has joined #openstack-lbaas15:05
*** jschwarz_ has joined #openstack-lbaas15:07
*** jschwarz has quit IRC15:07
*** apuimedo has quit IRC15:10
openstackgerritPhillip Toohill proposed stackforge/octavia: Update nova_driver to use amp_network
openstackgerritPhillip Toohill proposed stackforge/octavia: Update nova_driver to use amp_network
*** evgenyf has quit IRC15:22
*** kobis has quit IRC15:27
johnsomblogan Are you going to poke adam to +2/+A the controller, or should I poke German?  The oddness is German is a co-author15:27
* blogan pokes rm_work|away, rm_you|15:28
bloganhe's not in the office yet15:28
bloganif you can get german to do it go ahead15:28
johnsomHahahaaha, ok15:28
bloganadam was fine with merging it yesterday too15:28
johnsomYeah.  I'll see if I can find German.  He is out at a bunch of meetings15:29
bloganmust be meeting season15:30
*** rm_work|away is now known as rm_work15:38
openstackgerritMerged stackforge/octavia: Implements Octavia Controller Worker
*** mugu has left #openstack-lbaas15:40
*** jorgem has quit IRC15:41
johnsomCool stuff.  Let me know when the queue consumer is rebased (maybe updated for new bin location), I will kick it through as well.15:48
johnsomI am moving the open bugs in controller worker from our team's stash to launchpad.15:49
rm_workptoohill: commented15:58
rm_workTrevorV: commented on
ptoohillrm_work: There's error handling in the moethod that calls that one16:09
rm_workthen that's the answer i needed16:10
rm_workthat's just one of my "pattern triggers"16:10
rm_worklike seeing nested if statements16:10
ptoohillFair enough, That code was already there, I just updated it to use ID from config16:10
rm_workwant to comment on the code for posterity?16:11
rm_workthen I can +216:11
ptoohillI did16:11
ptoohillWell, now it's reviewed :P16:11
ptoohillThough, if we ever plan on using get_amphora in other calls those would need error handling or the transform method needs updated. Right now, only one thiung calls it and handles the errors the way it needs to from what I can tell16:14
ptoohillThis patch really was just a way to get around having a networking name in the config. And, since the ID will already live there it was a hop, skip and jump to pull the required info16:16
ptoohillHey ajmiller, for the devstack plugin, we dont have it marked as WIP and am curious in your opinion what, if anything, else it needs?16:19
*** nmagnezi has joined #openstack-lbaas16:19
ajmillerptoohill Mainly it just needs testing and validation with working control plane binaries.16:19
ptoohillI made updates lastnight where it now calls the api built by and a few other tweaks16:19
ajmillerOK, cool.16:19
ajmillerI've been mostly on vacation since last Friday, so am a little behind on stuff.16:20
ptoohillgotcha, Ive only tested the api, creating mgmt network and tempest config population. So, this review will probably hang out in review land for a while I assume?16:20
ptoohillAh, no worries :)16:20
ajmillerYes, it can hang out for a while.  Should it be marked as WIP?16:21
ptoohillI don't think so, from the things ive tested and looking at the other binary start ups it seems to be in good position for review. I was just curious if we needed anything else16:22
ajmillerCool.  I don't think I have any outstanding items to do, other than testing and validation.16:23
ptoohillI guess I do have a question16:24
ptoohillWill the other binaries be installed with
ptoohillIf so, then we would need to change a few things16:25
ajmillerwhich other binaries?16:25
*** Aish has joined #openstack-lbaas16:25
ptoohillthe other workers, everything but api for this question16:25
ptoohillthe consumer, housekeeper etc16:26
ajmillerAh, I'm not sure about that.  I would guess that they would, but that's probably a question for dougwig blogan johnsom xgerman16:26
ptoohill;) fair enough. If so, we should chaenge the 'bin_dir' to use what other projects are using which is:16:27
johnsomI think they probably should be installed with  Not sure why they shouldn't16:27
ptoohillJust curious, as right now the plugin assume it will start from within the package. If we install everything with then we should call those binaries16:28
ptoohillbut, i cant really test those at the moment so its not a big deal16:28
ptoohillNot a big deal right now, was just curious since I was messing with it this AM. We can look into it deeper later.16:29
ajmillerok, great.16:29
openstackgerritTrevor Vardeman proposed stackforge/octavia: Update PUT methods to send update through queue
*** bharath has joined #openstack-lbaas16:51
*** madhu_ak has joined #openstack-lbaas16:55
*** jorgem has joined #openstack-lbaas17:01
*** jorgem has quit IRC17:01
*** rm_work is now known as rm_work|away17:03
*** Youcef has joined #openstack-lbaas17:04
*** hitalia has joined #openstack-lbaas17:13
*** jschwarz_ has quit IRC17:16
*** nmagnezi has quit IRC17:21
*** sbalukoff has quit IRC17:23
*** SumitNaiksatam has quit IRC17:34
*** johnsom_ has joined #openstack-lbaas17:39
*** kbyrne has quit IRC17:47
*** johnsom has quit IRC17:47
*** jkoelker has quit IRC17:47
*** ptoohill has quit IRC17:47
*** jkoelker has joined #openstack-lbaas17:49
*** ptoohill has joined #openstack-lbaas17:49
*** kbyrne has joined #openstack-lbaas17:54
*** SumitNaiksatam has joined #openstack-lbaas17:56
*** johnsom_ is now known as johnsom17:57
openstackgerritMadhusudhan Kandadai proposed openstack/neutron-lbaas: Tempest tests using testscenarios for neutron-lbaas
*** mlavalle has quit IRC18:18
*** mwang2 has joined #openstack-lbaas18:19
*** rm_work|away is now known as rm_work18:23
*** sbalukoff has joined #openstack-lbaas18:28
*** Youcef has quit IRC18:33
*** rm_work is now known as rm_work|away18:35
*** xgerman has joined #openstack-lbaas18:43
*** rm_work|away is now known as rm_work18:48
bloganptoohill, ajmiller: yeah should install those binaries19:25
ajmillerok, thanks for verifying blogan19:25
bloganjohnsom: i'm rebasing the queue consumer19:26
bloganajmiller: its not setup yet to do that though :)19:27
*** madhu_ak_ has joined #openstack-lbaas19:30
*** madhu_ak has quit IRC19:33
openstackgerritTrevor Vardeman proposed stackforge/octavia: Update PUT methods to send update through queue
*** mwang2 has quit IRC19:43
*** mwang2 has joined #openstack-lbaas19:48
*** hitalia has quit IRC19:51
*** bavanveen has joined #openstack-lbaas19:58
*** madhu_ak has joined #openstack-lbaas20:02
*** madhu_ak_ has quit IRC20:02
*** hitalia has joined #openstack-lbaas20:07
*** mlavalle has joined #openstack-lbaas20:15
*** bavanveen has quit IRC20:32
*** mwang2_ has joined #openstack-lbaas20:44
*** mwang2 has quit IRC20:47
*** madhu_ak has quit IRC20:48
*** madhu_ak has joined #openstack-lbaas20:49
johnsomFYI, SSH driver bug:
openstackLaunchpad bug 1450630 in octavia "SSH driver doesn't update interfaces file correctly" [Critical,New]20:56
*** bharath has quit IRC21:04
*** bharath has joined #openstack-lbaas21:05
*** hitalia has quit IRC21:05
openstackgerritBrandon Logan proposed stackforge/octavia: Implemented Queue Consumer
bloganjohnsom: ^21:11
*** bharath has quit IRC21:11
johnsomI see it21:11
*** bharath has joined #openstack-lbaas21:11
bloganadded the entry point in the setup.cfg too21:12
bloganpbr doesn't like -'s in console_scripts21:13
johnsomCool, I have a fix for the lb update issue I will put up today.21:13
bloganTrevorV: ssh driver bug above21:14
blogandidn't mean to wake you21:14
bloganwell yeah i did21:14
*** bharath has quit IRC21:17
*** hitalia has joined #openstack-lbaas21:17
johnsomblogan german has a question about the cmd directory21:27
bloganneutron and other projects started donig that21:28
johnsomHe put it on the queue consumer in gerrit21:28
johnsomYeah, I don't have the source to point him at21:28
TrevorVjohnsom I'll fix that ASAP, though it is a weird thing to happen.  If you care to know, the "post_network_plug" was called before your interface was "DOWN" so the grep results were empty... Hence the issue you saw.21:32
johnsomAh, interesting.21:33
johnsomNot something I'm doing wrong in the sequencing or something is it?21:33
TrevorVNot sure.21:33
TrevorVIf post_network_plug is called before the neutron interface attach is called, that could present that problem21:33
bloganor if interface attach failed21:34
johnsomThat is the sequence I have in the network subflow21:35
TrevorVYeah that looks correct, assuming the plug networks is completed successfully, as blogan pointed out21:35
bloganah i bet plug vip should go first before plug networks21:36
johnsomI have them in nova and I see eth1 on the amp21:36
TrevorVI was just discussing with blogan, so there should be two separate network subflows21:39
TrevorVOne like so:21:39
TrevorV GetPlumbedNetworks->CalculateDelta->PlugNetworks->AmphoraPostNetworkPlug->PlugVIP->.AmphoraPostVIPPlug21:39
TrevorVha ha21:39
TrevorVhold on21:39
TrevorVFirst flow:21:39
TrevorV GetPlumbedNetworks->CalculateDelta->PlugNetworks->AmphoraPostNetworkPlug21:40
TrevorVSecond flow:21:40
TrevorV GetPlumbedNetworks->PlugVIP->.AmphoraPostVIPPlug21:40
bloganso create lb should only do plug vip and post plug vip21:40
bloganadding a member should do calculate delta, and then do plug network and post plug network21:41
johnsomWhy do they have to be separate?21:41
bloganwell plug vip should only happen on a create lb since we aren't allowing the ability to add multiple vips21:41
bloganand the only time we'll ever need to plug a network is on a member addition21:42
johnsomThis is create lb flow21:42
bloganyeah so create lb flow shouldn't need to do plug network bc the only network it will need to be plugged into is the vip network, which will be taken care of by the plug vip and post vip plug tasks21:43
johnsomI don't think getplumbednetworks needs to be run before plugvip either.  plugvip doesn't care about the networks21:44
johnsomBut it will need the listener network right?21:44
openstackgerritBrandon Logan proposed stackforge/octavia: Implemented Queue Consumer
bloganlistener network?21:48
bloganthere's just the vip network, and the member networks21:48
bloganyou mean listener as in octavia listener? or haproxy listener?21:48
johnsomI agree, vip and member21:49
bloganbut yeah getplumbednetworks wouldn't need to be called either21:49
bloganfor the vip21:49
bloganwell the create lb flow21:49
johnsomcreate lb should setup the vip network, right?21:50
xgermanblogan - hi21:51
bloganjohnsom: yes21:52
bloganxgerman: hi21:52
xgermanyeah, I am cool with cmd (need to read up) but are we naming the commands then .py?21:52
xgermanor like we started with bin without the .py suffix?21:52
johnsomblogan When I am done with this lb update patch I will look at those network methods and see if my understanding of what each of them do is wrong21:52
dougwiganother neutron process change, that didn't hit in time for the octavia meeting.  any existing octavia or lbaas core especially, take note:
johnsomAre you tapping your fingers together and laughing dougwig?21:54
dougwigi think i'm missing the reference?  :)21:55
bloganxgerman: bc pbr will automatically read the setup.cfg and it expects .py files to create the console_scripts21:55
xgermanok, got it21:55
*** bharath has joined #openstack-lbaas21:56
xgermanI have been in meetings the last three days — so my head hurts21:56
bloganxgerman: my meetings have increased as well, thankfully i haven't been unfortunate enough to be in as many as you21:57
xgermanwell, three days all-day in a row is killing me...21:58
johnsomxgerman Better you than me...  Grin21:58
xgerman(+ I need to live in the airport like dougwig)21:58
openstackgerritMerged stackforge/octavia: Implemented Queue Consumer
xgermanjohnsom Happy to take that bullet...21:59
johnsomI have never been able to master the sleeping with eyes open thing...21:59
dougwigjohnsom: necessary skill for college.22:00
johnsomI just sat in the back.  The classes I didn't care about were large....22:00
bloganor even better, learn to pass the class without attending the class22:00
johnsomblogan +122:01
bloganalways pissed me off when professors made attendance part of the grade22:01
rm_workthey were lazy and didn't want to test on everything they taught -- or else, they wanted to teach stuff you didn't actually need to know, and therefore wouldn't make sense to test on, but they REALLY wanted you to know it22:10
rm_workmostly the latter is professors who just want to tell stories and force people to listen22:11
johnsomStill wonder why I needed to know the name of the hormone that makes tobacoo plants grow until they die as an engineering student....22:13
johnsomTrue midterm question22:14
*** sbalukoff has quit IRC22:15
dougwigjohnsom: do you still remember the name?22:15
johnsomNo, frankly I don't think I ever knew it22:15
johnsomI did end up passing the class though...22:16
dougwigjohnsom: darn, we could've used it for our next naming argument.22:16
bloganshould have named amphora that, it woudl have been much more obvious22:21
rm_workI hate all of you, but mostly blogan22:22
bloganeveryone mostly hates me22:22
bloganmy wife hated me so much she married me22:23
dougwigwe have to marry you if we hate you enough?  oh my.22:26
*** chlong has joined #openstack-lbaas22:36
blogandoug logan, got a nice ring to it22:38
bloganbut you wouldn't get a nice rnig22:38
openstackgerritMichael Johnson proposed stackforge/octavia: Fixes load balancer create flow issue
*** clev is now known as clev-away22:48
*** jorgem has joined #openstack-lbaas22:51
*** jorgem has quit IRC22:58
openstackgerritmin wang proposed openstack/neutron-lbaas: Admin_state_up tempest test for health monitor
*** woodster_ has quit IRC23:10
*** mlavalle has quit IRC23:11
*** clev-away is now known as clev23:17
openstackgerritMichael Johnson proposed stackforge/octavia: Fixes load balancer create flow issue
*** woodster_ has joined #openstack-lbaas23:22
*** TrevorV|Home has joined #openstack-lbaas23:26
*** rm_work is now known as rm_work|away23:27
*** rm_work|away is now known as rm_work23:27
*** rm_work is now known as rm_work|away23:29
*** clev is now known as clev-away23:38
*** TrevorV|Home has quit IRC23:43
*** SumitNaiksatam has quit IRC23:54

Generated by 2.14.0 by Marius Gedminas - find it at!