Tuesday, 2016-02-16

« Monday, 2016-02-15 Index Wednesday, 2016-02-17 »
*** _ducttape_ has quit IRC00:03
*** bana_k has quit IRC00:14
openstackgerritFranklin Naval proposed openstack/octavia: Preparing for Tempest testing  https://review.openstack.org/17219900:20
fnavalplease take a look when anyone has a chance - thanks in advance ^^00:21
openstackgerritFranklin Naval proposed openstack/octavia: Octavia: Basic LoadBalancer Scenario Test  https://review.openstack.org/17219900:22
*** crc32 has quit IRC00:39
*** ducttape_ has joined #openstack-lbaas00:40
*** ducttape_ has quit IRC00:45
*** yamamoto has joined #openstack-lbaas00:56
*** yamamoto has quit IRC01:03
*** neelashah has joined #openstack-lbaas01:24
*** bana_k has joined #openstack-lbaas01:24
*** ducttape_ has joined #openstack-lbaas01:29
*** allan_h has quit IRC01:29
*** kevo has quit IRC01:43
*** ducttape_ has quit IRC01:44
*** ducttape_ has joined #openstack-lbaas01:47
*** yamamoto has joined #openstack-lbaas01:48
*** yamamoto has quit IRC02:00
*** bana_k has quit IRC02:10
*** neelashah has quit IRC02:14
*** armax has joined #openstack-lbaas02:26
*** ducttape_ has quit IRC02:30
*** 18VAAAIFD has joined #openstack-lbaas02:47
openstackgerritStephen Balukoff proposed openstack/octavia: Fix model update flows  https://review.openstack.org/28047802:55
*** sbalukoff has quit IRC03:01
*** armax has quit IRC03:03
*** ducttape_ has joined #openstack-lbaas03:09
*** ducttape_ has quit IRC03:23
*** ducttape_ has joined #openstack-lbaas03:28
*** madhu_ak has joined #openstack-lbaas03:37
*** madhu_ak has quit IRC03:46
*** prabampm has quit IRC03:46
*** amotoki has joined #openstack-lbaas03:56
*** amotoki has quit IRC03:56
*** ducttape_ has quit IRC03:57
*** fnaval has quit IRC04:19
*** fnaval has joined #openstack-lbaas04:30
*** allan_h has joined #openstack-lbaas04:45
*** allan_h has quit IRC05:22
*** sbalukoff has joined #openstack-lbaas05:24
*** fnaval_ has joined #openstack-lbaas05:32
*** fnaval has quit IRC05:35
*** prabampm has joined #openstack-lbaas05:40
*** numans has joined #openstack-lbaas06:17
*** kobis has quit IRC06:43
*** bana_k has joined #openstack-lbaas06:52
*** _cjones_ has joined #openstack-lbaas07:04
*** _cjones_ has quit IRC07:06
*** _cjones_ has joined #openstack-lbaas07:06
*** nmagnezi has joined #openstack-lbaas07:19
*** kobis has joined #openstack-lbaas07:31
*** kiranr_ has joined #openstack-lbaas07:32
*** kiranr_ is now known as kiran-r07:33
*** rcernin has joined #openstack-lbaas07:37
*** bana_k has quit IRC07:37
*** clduser_ has quit IRC07:43
*** clduser has joined #openstack-lbaas07:43
*** eezhova has quit IRC07:44
*** eezhova has joined #openstack-lbaas07:45
*** bedis has quit IRC07:45
*** bedis has joined #openstack-lbaas07:45
*** bana_k has joined #openstack-lbaas07:57
*** fnaval_ has quit IRC07:58
*** fnaval has joined #openstack-lbaas07:58
openstackgerritReedip proposed openstack/neutron-lbaas: Add Timeout on Listeners  https://review.openstack.org/27389608:09
*** bana_k has quit IRC08:23
*** _cjones_ has quit IRC08:29
*** jschwarz has joined #openstack-lbaas08:58
*** kiran-r has quit IRC09:08
*** kiran-r has joined #openstack-lbaas09:09
openstackgerritStephen Balukoff proposed openstack/octavia: Add L7 database structures  https://review.openstack.org/26543009:53
openstackgerritStephen Balukoff proposed openstack/octavia: Update repos for L7 policy / methods  https://review.openstack.org/26552910:02
openstackgerritStephen Balukoff proposed openstack/octavia: Update repos for L7 rules / validations  https://review.openstack.org/27664310:12
*** numans has quit IRC10:18
*** numans has joined #openstack-lbaas10:18
openstackgerritStephen Balukoff proposed openstack/octavia: Add L7 api - policies  https://review.openstack.org/26569010:21
openstackgerritStephen Balukoff proposed openstack/octavia: Add L7 api - rules  https://review.openstack.org/27771810:28
openstackgerritStephen Balukoff proposed openstack/octavia: Add L7 controller worker flows and tasks  https://review.openstack.org/27776810:36
*** prabampm has quit IRC10:37
openstackgerritStephen Balukoff proposed openstack/octavia: Add L7 jinja template updates  https://review.openstack.org/27822310:42
openstackgerritStephen Balukoff proposed openstack/octavia: Add L7 documentation  https://review.openstack.org/27883010:45
*** johnsom has quit IRC11:00
openstackgerritStephen Balukoff proposed openstack/octavia: Update data model conversion to be more complete  https://review.openstack.org/28016611:01
openstackgerritStephen Balukoff proposed openstack/octavia: Fix model update flows  https://review.openstack.org/28047811:03
*** prabampm has joined #openstack-lbaas11:09
*** 18VAAAIFD has quit IRC11:40
*** pcaruana has joined #openstack-lbaas12:18
*** yamamoto has joined #openstack-lbaas12:30
*** rtheis has joined #openstack-lbaas12:31
*** ducttape_ has joined #openstack-lbaas13:12
*** doug-fish has joined #openstack-lbaas13:14
*** neelashah has joined #openstack-lbaas13:18
*** neelashah has quit IRC13:24
*** ducttape_ has quit IRC13:25
*** doug-fish has quit IRC13:34
*** doug-fish has joined #openstack-lbaas14:02
*** neelashah has joined #openstack-lbaas14:05
*** kiran-r has quit IRC14:08
*** neelashah has quit IRC14:21
*** localloop127 has joined #openstack-lbaas14:40
*** neelashah has joined #openstack-lbaas14:43
*** nmagnezi has quit IRC14:55
*** ducttape_ has joined #openstack-lbaas14:57
*** ducttape_ has quit IRC14:57
*** ducttape_ has joined #openstack-lbaas14:57
*** numans has quit IRC15:15
*** ajmiller has joined #openstack-lbaas15:39
*** johnsom has joined #openstack-lbaas15:48
*** nmagnezi has joined #openstack-lbaas15:49
*** minwang2 has joined #openstack-lbaas15:50
*** minwang2 has quit IRC15:52
*** minwang2 has joined #openstack-lbaas15:55
*** jschwarz has quit IRC15:57
*** Bjoern has joined #openstack-lbaas16:05
*** Aish has joined #openstack-lbaas16:13
*** pcaruana has quit IRC16:18
*** jwarendt has joined #openstack-lbaas16:21
*** rcernin has quit IRC16:37
*** armax has joined #openstack-lbaas16:38
*** kobis has quit IRC16:45
*** ajmiller has quit IRC16:47
*** fnaval has quit IRC16:47
*** allan_h has joined #openstack-lbaas16:59
*** madhu_ak has joined #openstack-lbaas17:00
*** localloop127 has quit IRC17:04
*** fnaval has joined #openstack-lbaas17:04
*** madhu_ak has quit IRC17:05
*** kobis has joined #openstack-lbaas17:06
*** rcernin has joined #openstack-lbaas17:09
*** madhu_ak has joined #openstack-lbaas17:15
*** jschwarz has joined #openstack-lbaas17:20
*** Aish has quit IRC17:29
*** minwang2 has quit IRC17:30
*** _cjones_ has joined #openstack-lbaas17:35
*** localloop127 has joined #openstack-lbaas17:35
*** localloop127 has quit IRC17:39
*** Aish has joined #openstack-lbaas17:40
*** ajmiller has joined #openstack-lbaas17:46
*** kevo has joined #openstack-lbaas17:50
johnsomFolks, I have created an etherpad for the LBaaS related summit talks open for vote.  Please check that I didn't miss any: https://etherpad.openstack.org/p/Austin-LBaaS-talks17:51
*** bana_k has joined #openstack-lbaas17:56
*** localloop127 has joined #openstack-lbaas17:57
*** yamamoto has quit IRC18:05
*** minwang2 has joined #openstack-lbaas18:06
*** allan_h has quit IRC18:08
*** yamamoto has joined #openstack-lbaas18:10
*** PK has joined #openstack-lbaas18:11
Bjoernthanks for the etherpad18:17
*** neelashah has quit IRC18:34
*** jschwarz has quit IRC18:37
*** ajmiller has quit IRC18:44
*** ajmiller_ has joined #openstack-lbaas18:44
*** kobis has quit IRC18:48
*** yamamoto has quit IRC18:49
*** nmagnezi has quit IRC18:50
*** neelashah has joined #openstack-lbaas19:00
*** Bjoern has quit IRC19:01
*** allan_h has joined #openstack-lbaas19:21
*** allan_h has quit IRC19:22
*** kobis has joined #openstack-lbaas19:24
*** allan_h has joined #openstack-lbaas19:24
*** ajmiller has joined #openstack-lbaas19:27
*** SumitNaiksatam has joined #openstack-lbaas19:29
*** ajmiller_ has quit IRC19:30
*** yamamoto has joined #openstack-lbaas19:50
fnavalsbalukoff: this PR should have a working end-to-end scenario test for an octavia LB:   https://review.openstack.org/#/c/172199/19:51
fnavalplease test it out / review when you get a chance. thanks in advance19:52
*** yamamoto has quit IRC19:56
sbalukoffAwesome!19:57
*** neelashah has quit IRC20:03
sbalukoffblogan: Thanks for your comment on my patch from last night. I've responded, eh!20:03
blogansbalukoff: okay!20:03
openstackgerritFranklin Naval proposed openstack/octavia: Octavia: Basic LoadBalancer Scenario Test  https://review.openstack.org/17219920:05
*** neelashah has joined #openstack-lbaas20:05
*** PK has quit IRC20:09
*** kevo has quit IRC20:13
openstackgerritFranklin Naval proposed openstack/octavia: Octavia: Basic LoadBalancer Scenario Test  https://review.openstack.org/17219920:14
*** PK has joined #openstack-lbaas20:15
*** PK has quit IRC20:37
openstackgerritFranklin Naval proposed openstack/octavia: Octavia: Basic LoadBalancer Scenario Test  https://review.openstack.org/17219920:39
openstackgerritFranklin Naval proposed openstack/octavia: Octavia: Basic LoadBalancer Scenario Test  https://review.openstack.org/17219920:40
*** neelashah has quit IRC21:20
rm_worksbalukoff: after looking through some of this L7 stuff for about the 10th time, I'm starting to wonder if this isn't another case of "make sure there's no glaring syntax errors, and merge the whole wad ASAP and fix resulting bugs afterwards" >_<21:21
sbalukoffrm_work: I tried to add enough unit and functional tests for that to be the case.21:21
sbalukoffBut!21:21
sbalukoffI did find a bug this morning I've yet to root out.21:22
rm_worktrying to wrap my head around the new graph model21:22
sbalukoffGive me a chance to get that patched, and I'd be ready for the whole chain to merge.21:22
rm_workkk21:22
sbalukoffLet me know if there's anything I can answer about the graph generation / traversal code.21:22
rm_workI had just finished explaining to trevorv why i wouldn't necessarily expect updating default_pool_id to actually update default_pool, and there you go fixing it to actually work :P21:22
sbalukoffI also added a couple tests for that to make sure I wasn't totally crazy in how it should work.21:23
sbalukoffHa!21:23
sbalukoffThere's also a session persistence update in that patch that I'm surprise hadn't manifested before.21:23
rm_workbut that solves the dilemma i had about copy/pasting db-query code for Pool objects into the Listener repo21:23
rm_worksince you can't easily access one repo from another21:24
sbalukoffOnce I understood how the model updates worked, I was surprised it didn't come up before.21:24
sbalukoffAre you talking about the find_in_graph method?21:24
rm_workno, it was just when updating the listener he needed to set the default_pool in a spot and only had the ID21:25
rm_workand was having to pull the pool object from the DB as a workaround21:25
rm_workto set it on the listener21:25
rm_workanywho, I'm going to check out the absolute last patch in that chain and just hammer it a bit with some manual testing, once you're ready21:25
sbalukoffThere's a trick to that method: The object you're looking for needs to actually be connected to the graph somewhere. Thankfully, since everything (right now) ultimately has a link to a load balancer object... the node you're looking for will almost certainly be in the same load balancer tree, so you'll find it if it's there.21:26
johnsomMy two cents, I would really like one of the cores to check this stuff out and do a basic smoke test before we merge it all and try to work out the bugs.21:26
rm_workand if it all looks good, I'm going to recommend we +2/+A everything from the end of the chain to the start, so it all merges as a big wad21:26
rm_workjohnsom: ^^21:26
rm_workjohnsom: basically that's what i said i would do :P21:26
johnsomWe have four open bugs with what has merged so far and I worry about gate breakage slowing down other merges21:26
rm_work<rm_work>anywho, I'm going to check out the absolute last patch in that chain and just hammer it a bit with some manual testing, once you're ready21:26
rm_workjohnsom: sound like a plan?21:27
sbalukoffrm_work: Ok, again, I'm still working on fixing that bug I found today. Want me to ping you once I have the fix for it committed?21:27
rm_worksbalukoff: yes21:27
sbalukoffOk, will do.21:27
rm_workthat's what I meant by "once you're ready"21:27
rm_workmeanwhile, it's about to be sandwich time21:27
sbalukoffOk!21:27
rm_worksandwich time = best time21:27
johnsomrm_work Cool.  I am just a bit concerned because I couldn't even build an lb with some of the patches last wekk.21:27
sbalukoffHeh!21:27
*** kevo has joined #openstack-lbaas21:27
rm_workhmmm21:28
rm_worki don't remember merging anything that i didn't at least get a basic build working with21:28
johnsomsbalukoff rm_work Ping me too, I will also do a checkout and try it out.  If it seems cool, I will be in favor of a mass checkin21:28
rm_workthough i was kinda out of the game for the end of last week21:28
sbalukoffjohnsom: Will do! and thanks!21:28
ptoohillsbalukoff: ping21:29
sbalukoffptoohill: Pong!21:29
johnsomsbalukoff Not to bring up a rough subject, but are the models working now such that we don't need to re-order?  I saw you committed some changes, but lost my IRC log over the weekend21:29
ptoohillah, youre here, i was scrolled up :) Can you 'share' l7 policies between listeners?21:29
ptoohillim looking now to verify, but youre here and can probably help me out quicker :)21:30
sbalukoffjohnsom: with this patch, they should be: https://review.openstack.org/#/c/280478/  (which has a parent patch as well)21:30
johnsomsbalukoff Awesome, thanks!21:31
sbalukoffjohnsom: I still think we should revisit the subject how to do reverts at some point, especially because you're going to see duplicated algorithms in the L7 model code. But I don't think it's important enough to switch bikes mid-ride just before the feature freeze deadline.21:31
johnsomrm_work BTW, the sonar gate isn't voting21:31
rm_workurgh21:31
rm_workwhat happened....21:31
rm_workOH GOD THE BUILD QUEUE21:32
johnsomNot sure.  Maybe jenkins isn't triggering on openstack events?21:32
ptoohillsbalukoff: ^ messaged above without tag21:32
rm_workexecutors died21:32
sbalukoffptoohill: No, L7 policies are associated with exactly one listener.21:32
rm_workwtf21:32
ptoohillah perfect! Thank you sbalukoff21:32
ptoohill!21:32
rm_workAUGH DRIVE SPACE21:32
rm_workneed to clean up builds faster21:32
rm_workshit21:32
johnsomopps21:32
sbalukoffD'oh!21:32
rm_worki'm on it21:32
fnavaldoes anyone know how to get around a tox error that looks like this: Double requirement given: requests-mock>=0.7.0 (from -r /opt/stack/octavia/test-requirements.txt (line 11)) (already in requests-mock>=0.7.0 (from -r /opt/stack/octavia/test-requirements.txt (line 5)), name='requests-mock')21:33
rm_workoh jeeze it's actually logs21:34
sbalukoffIsn't test-requirements.txt just additive to requirements.txt?21:34
sbalukoffIs that mock requirment already listed elsewhere?21:34
sbalukofffnaval ^^21:34
rm_worklooks like ine 5 and 1121:34
rm_workbad merge prolly21:34
fnavalyes, i thought so too, but removing it causes other errors to happen21:35
rm_work*bad rebase merge from a thing21:35
rm_workremove it from line 1121:35
rm_workorder can matter21:35
fnavalhmm. ok, i'll try doing that21:35
fnavalah yes ok21:36
fnavalit was duped21:36
rm_workhmm jenkins logs are FULL of this shit:21:39
rm_work        [DNSQuestion@1708491063 type: TYPE_IGNORE index 0, class: CLASS_UNKNOWN index 0, name: ]21:39
rm_work        [DNSQuestion@1932278557 type: TYPE_IGNORE index 0, class: CLASS_UNKNOWN index 0, name: ]21:39
rm_worklike21:39
rm_worki can't even scroll through all of it21:39
rm_workit's all the same21:39
johnsomHow strange21:40
openstackgerritFranklin Naval proposed openstack/octavia: Octavia: Basic LoadBalancer Scenario Test  https://review.openstack.org/17219921:40
rm_workhttps://gist.github.com/rm-you/e8b13a408318318af7ab21:41
rm_workis this a DoS?21:42
sbalukoffSomeone really hates our analysis of Octavia code.21:42
rm_workall the same IP21:43
sbalukoffOr rather, it's probably a bot trying a DNS exploit.21:43
johnsomReally suspect.  Why is java doing dns incoming?21:43
sbalukoffMaybe change the local firewall to block everything but SSH and HTTPS?21:43
rm_workyeah this is super weird21:44
johnsomYeah, I would check the firewall settings, apt-get upgrade the packages21:45
johnsomIt's a russian IP, so, yeah...21:46
rm_worklulz21:46
rm_workhttps://issues.jenkins-ci.org/browse/JENKINS-1016021:50
johnsomInteresting21:51
*** neelashah has joined #openstack-lbaas21:52
rm_worknow busy trying to figure out why deleting the logs didn't free the disk space O_o21:52
rm_worknm21:53
rm_worki know why21:53
johnsomTry running sync21:53
rm_workjenkins still has the file handles open21:53
rm_workjust had to restart it21:53
rm_workand going to disable the dns shit21:54
johnsomYeah, we don't need it21:54
johnsomIf you want me to nmap the box to make sure extra ports aren't open, let me know21:54
*** sballe has left #openstack-lbaas21:58
rm_workif you want to kick that, sure21:59
rm_workjust restarted with that shit disabled21:59
madhu_akfnaval: just wondering, https://review.openstack.org/172199 is working locally? or still in WIP ?22:02
fnavalmadhu_ak: it should work - just figuring out why it's failing upstream22:02
fnavalflake8 issues apparently22:03
madhu_akfnaval: testing out locally22:03
fnavali'm rebuilding a new devstack with that change and going to test it out22:03
madhu_aksure fnaval22:04
fnavalpulling in refs/changes/99/172199/3622:04
rm_workso, should I try to re-kick our sonar job on a ton of patches? :P22:08
johnsomUp to you.  It's not easy the way the jenkins UI is setup22:09
fnavalmadhu_ak: i managed to reduce about 50% of the proposed code. but it's still about 8K lines22:09
rm_workeh just did it i think22:10
rm_workwasn't too bad22:10
madhu_akfnaval: Just tested them. Able to run the tests successfully22:12
rm_worki'll have to keep an eye on jenkins to make sure that doesn't repeat22:13
openstackgerritmin wang proposed openstack/octavia: Implements: blueprint anti-affinity server group  https://review.openstack.org/27234422:15
bloganjohnsom, rm_work, sbalukoff: do yall have anymore information about the intermitten scenario job failures? like do they relate to the data model discussions?22:16
rm_workblogan: i believe they do22:16
rm_workand i hope they will all disappear after the model changes go in22:16
rm_workbut we'll have to see22:16
bloganbut the model changes are still failing the scenario tests22:16
rm_workor at least, the underlying cause of weirdness will be gone, and anything left will be REAL bugs22:16
rm_workit's possible there are actual bugs exposed by the model fixes22:16
fnavalhmm.. now i'm getting a weird sonar error - I don't think it's related to my changes22:17
rm_workfnaval: err22:17
rm_workwhat22:17
fnavalrm_work: http://sonar.octavia.io/archive/octavia/036870b64088f6326fa689591cec8529f9f6811a/245/console_log.txt.gz22:17
rm_workuhhhh22:18
rm_workthat looks like a bug possibly:22:18
rm_workdebug = <oslo_config.cfg.GroupAttr object at 0x7fd2509012d0>22:18
rm_workin the config22:18
johnsomblogan Not sure.22:19
fnavaldoes that backslash supposed to be there?22:19
rm_workfnaval: let the sonar queue clear through22:19
rm_workand then i'll rerun22:19
fnavalok22:19
rm_workbut it looks like the debug value isn't being passed correctly22:19
fnavalmadhu_ak: it looks like we're good with that changeset: refs/changes/99/172199/3622:20
rm_worklike something is passing in an OBJECT and not the actual value22:20
madhu_akyep. reviewing them on gerrit and see if I can find anything unusual :)22:20
rm_workfnaval: and it's only on yours...22:20
johnsomblogan Could be one or more of these causing issues too:https://bugs.launchpad.net/octavia/+bugs?field.tag=sharedpools22:21
fnavalyou can use this script to build a devstak to tst also: https://gist.github.com/fnaval/e3640d8e713804aa8ffb22:21
rm_workother changes are passing fine22:21
fnavalrm_work: ??  hmmm I dont touch the unit test stuff22:21
rm_workyeah weird, i'll re-kick that check job in a bit22:21
rm_workprolly 30m or so to clear the current queue22:22
rm_workoh i forgot to sandwich >_<22:22
fnavalbut i think i have testtools in test-requirements.txt22:22
*** rtheis has quit IRC22:26
fnavalstrange that those failures weren't hitting before22:27
fnavalrm_work: hmm.. I think that the other PRs which have Sonar run against it are giving false successes22:29
fnavalex: http://sonar.octavia.io/archive/octavia/8e05ec5ac2320b5c641d4c43b8adebdacd150075/252/console_log.txt.gz22:29
fnavali think that should have failed22:30
*** localloop127 has quit IRC22:31
fnavalmaybe the introduction of some test libraries in my PR is now correctly reporting the failures?  (e.g, fixtures and testscenarios)22:32
rm_workhmm22:33
fnavalor maybe i picked up a bug when i pulled down master and rebased22:33
fnavali'll try rebasing again22:33
madhu_akI could see the same error like sonar reporting. ValueError: u'00:00:00:00:00:00' does not appear to be an IPv4 or IPv6 address22:34
madhu_akregardless of tox -e cover, rest looks okay to me.22:35
fnavalcool. thanks for reviewing it madhu22:36
sbalukoffblogan: I was hoping the model changes would fix things too, but apparently they didn't (though this time at least it doesn't appear they've made the problem worse.)22:36
fnavali got the scenario test to pass locally - now have to figure out why the CI gates are failing22:38
openstackgerritFranklin Naval proposed openstack/octavia: Octavia: Basic LoadBalancer Scenario Test  https://review.openstack.org/17219922:39
fnavalhmmm erroring again on sonar22:50
fnavalhttp://logs.openstack.org/99/172199/30/check/gate-octavia-pep8/99e35bf/console.html22:52
fnavalwrong link; right link: http://sonar.octavia.io/archive/octavia/a42d611b6f042f1f972e619a27dd1f31d4920a5d/266/console_log.txt.gz22:53
fnavalah ok, got what you meant now rm_work with debug = <oslo_config.cfg.GroupAttr object at 0x7f8aef8f0590>22:55
*** Bjoern has joined #openstack-lbaas22:55
blogansbalukoff: i commented on your review about trying to just move to sqlalchemy models so we don't have to keep going down the path of duplicating what it does22:55
blogansbalukoff: but obviously not right now22:55
bloganwhich is what you probably were tlakinga bout too22:55
sbalukoffblogan: Yep.23:13
*** kobis has quit IRC23:16
*** rcernin has quit IRC23:18
*** kev0 has joined #openstack-lbaas23:28
*** ducttape_ has quit IRC23:31
*** markvan_ has joined #openstack-lbaas23:31
*** bedis_ has joined #openstack-lbaas23:32
*** itsuugo_ has joined #openstack-lbaas23:35
rm_worksbalukoff: did you finish your fixes and i missed it, or still working?23:36
sbalukoffStill working on it.23:36
*** cgross- has joined #openstack-lbaas23:36
sbalukoff(I'm taking my time-- trying to make sure I don't introduce any new brokenness.)23:36
*** kevo has quit IRC23:37
*** bedis has quit IRC23:37
*** itsuugo has quit IRC23:37
*** markvan has quit IRC23:37
*** cgross has quit IRC23:37
*** mdavidson has quit IRC23:37
*** itsuugo_ is now known as itsuugo23:37
sbalukoffThis might be something that I end up committing late tonight.23:37
rm_workok no worries23:37
rm_workjust wanted to make sure i wasn't slacking23:37
sbalukoffAlso, I have been turning my manual scenario tests into something I can run more easily (not quite scripted, but more copy-pastable)23:38
sbalukoffso that I can make sure changes don't introduce new brokenness.23:38
sbalukoffHave I mentioned lately how happy I am that fnaval is working on native tempest tests for Octavia? ;)23:38
rm_workyes23:39
rm_worksbalukoff: https://review.openstack.org/#/c/265430/28/octavia/db/models.py23:39
rm_workso23:39
rm_worki am not sure you fully answered my question on line 24223:40
rm_work"Is this intended to prevent changes from being saved in the class scope variable, or is it just a shortcut?"23:40
sbalukoffYou're asking why it's named with an underscore?23:41
rm_workno23:41
rm_workwhy did you even do "_listeners = self._default_listeners"23:41
*** neelashah has quit IRC23:41
fnavalsbalukoff:  it's up for review - just submitting one more patch23:41
rm_workwas it to save 13 characters from self._default_listeners ?23:41
openstackgerritFranklin Naval proposed openstack/octavia: Octavia: Basic LoadBalancer Scenario Test  https://review.openstack.org/17219923:42
fnavalthere you go23:42
sbalukoffOh! No-- I wanted to make the distinction between listeners that reference that pool via their default_listener_id, and l7policies that reference that pool via their redirect_pool_id.23:42
johnsomsbalukoff https://review.openstack.org/#/c/277768/323:42
*** Bjoern has quit IRC23:42
sbalukofffnaval: Thanks!23:42
rm_worksbalukoff: so it was just renaming it for readability?23:42
fnavalbut, i have a feeling it might fail again on sonar again23:42
sbalukoffrm_work: I don't think so. But maybe I'm not following what you're implying?23:43
*** mdavidson has joined #openstack-lbaas23:43
fnavalbut it works if you just cherry pick that patch23:43
rm_worksbalukoff: what do you gain by using _listeners instead of self._default_listeners in the code below? (this is a leading question)23:43
rm_work(I have a few answers in mind, want to know which one it is)23:43
rm_workone valid answer is: it is 13 less characters23:44
sbalukoffpool.listeners should be a unique list of listeners that reference the pool, whether that be directly through a listener.default_pool or indirectly via a listener.l7policy[n].redirect_pool23:44
rm_workok but23:44
rm_work_l_ids = [l.id for l in _listeners]23:44
rm_workis functionally identical to23:44
*** Bjoern has joined #openstack-lbaas23:45
rm_work_l_ids = [l.id for l in self._default_listeners]23:45
fnavalbbl23:45
rm_workI just want to know why line 242 exists23:46
rm_worksbalukoff: ^^23:46
rm_worka perfectly valid answer is "I didn't want to type the extra 13 characters later"23:47
rm_workbut I want to know if there's more23:47
*** Bjoern has quit IRC23:47
sbalukoffI'm appending to _listeners on line 248. I can't do that with _default_listeners23:48
rm_workerg ok that's what i was afraid of23:49
rm_workit's not doing what you think it's doing23:49
sbalukoffOr at least, I'm pretty sure I can't, since _default_listeners is an SQLAlchemy relationship.23:49
rm_workyou ARE appending to self._default_listeners23:49
rm_workthat's what I meant by " If it's the former, this doesn't work."23:50
sbalukoffAah-- Ok, so I'm getting a reference there where I need to be copying the whole list?23:50
rm_workyes23:50
rm_workuse copy.copy()/deepcopy()23:50
sbalukoffOk.23:50
openstackgerritMerged openstack/octavia: Stop logging amphora cert generation in debug  https://review.openstack.org/27933423:51
sbalukoffGood catch.23:51
rm_workprobably just copy.copy()23:51
rm_workshould be enough23:51
rm_workthere's other tricks that I think work too --23:51
rm_workyou can redefine it with a comprehension23:51
sbalukoffCould also just do:  _listeners = [l for l in _default_listeners]23:51
rm_workor use .extend() on a blank list23:51
rm_workyes23:51
rm_workthat works23:52
rm_workbut copy.copy() is *for* this :)23:52
sbalukoffOk, please comment and -1, eh!23:52
rm_worktthat said, i don't care which one you do23:52
rm_workif copy.copy works i think it's clearer23:52
rm_workand possibly more efficient? though i doubt that matters23:52
sbalukoffIs copy.copy a builtin or do I have to import it?23:52
rm_workimport, but it's standard lib23:53
sbalukoffOk.23:53
sbalukoffI'll check it out.23:53
blogancan't just do _listeners = self._default_listeners[:]23:54
blogan?23:54
bloganthats normally how you make a copy of a list in python, as long as you don't need a deep copy23:54
blogansbalukoff: ^^23:54
rm_workblogan: that works too23:54
sbalukoffblogan: I'll do that.23:54
rm_worklike i said there's tricks23:54
rm_worki prefer clarity, but23:55
rm_worki'm willing to +2 anything that's functional for that23:55
rm_worki just imagine someone walking through the code in a year being like "why the hell is this happening"23:55
bloganwelp the whole [:] trick is pretty standard list splicing23:55
sbalukoffYep. It's a good catch.23:55
rm_workcopy.copy() makes it explicitly clear that you're copying :)23:55
rm_worki wonder what it does for lists on the backend23:56
*** yuanying has quit IRC23:56
bloganwhat backend?23:57
sbalukoffthe sqlalchemy back-end.23:57
rm_workno i mean like23:58
rm_workin the C code for the copy lib23:58
rm_workbut anyway23:58
rm_work"Shallow copies of dictionaries can be made using dict.copy(), and of lists by assigning a slice of the entire list, for example, copied_list = original_list[:]."23:58
rm_workfrom the copy docs23:58
rm_workhttps://docs.python.org/2/library/copy.html23:58
rm_workso yeah23:58
rm_workprobably it just DOES that23:58
*** yuanying has joined #openstack-lbaas23:58
« Monday, 2016-02-15 Index Wednesday, 2016-02-17 »

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