Monday, 2014-07-07

« Sunday, 2014-07-06 Index Tuesday, 2014-07-08 »
*** shayneburgess has joined #openstack-trove00:39
*** shayneburgess has quit IRC00:39
*** eghobo has quit IRC01:19
*** nosnos has joined #openstack-trove01:37
openstackgerritA change was merged to openstack/trove: Added route for Admin API to support guest upgrade  https://review.openstack.org/8141001:54
*** miqui_ has quit IRC02:11
*** ViswaV has joined #openstack-trove02:35
*** achampion has quit IRC02:37
*** achampion has joined #openstack-trove02:46
*** ramishra has joined #openstack-trove02:57
*** coolsvap|afk is now known as coolsvap03:05
*** mrsnivvel has joined #openstack-trove03:07
*** yidclare has joined #openstack-trove03:22
*** jeraldv has joined #openstack-trove03:23
*** nosnos has quit IRC03:25
*** ViswaV has quit IRC03:38
*** ViswaV has joined #openstack-trove03:39
*** coolsvap is now known as coolsvap|afk04:00
*** nosnos has joined #openstack-trove04:02
*** yidclare has quit IRC04:02
*** eghobo has joined #openstack-trove04:06
*** robertmyers has quit IRC04:19
*** juantwo has quit IRC04:20
*** robertmyers has joined #openstack-trove04:21
*** robertmyers has quit IRC04:26
*** robertmyers has joined #openstack-trove04:26
*** robertmyers has quit IRC04:31
*** eghobo has quit IRC04:37
*** eghobo has joined #openstack-trove04:40
*** ajc_ has joined #openstack-trove04:52
*** sgotliv has joined #openstack-trove05:25
*** coolsvap|afk is now known as coolsvap05:30
openstackgerritA change was merged to openstack/trove: Add CONTRIBUTING.rst  https://review.openstack.org/10378705:34
*** annashen_ has quit IRC05:36
*** Longgeek has joined #openstack-trove05:45
openstackgerritA change was merged to openstack/trove: Imported Translations from Transifex  https://review.openstack.org/10475705:45
openstackgerritOpenStack Proposal Bot proposed a change to openstack/trove: Imported Translations from Transifex  https://review.openstack.org/10507006:05
*** sgotliv has quit IRC06:10
*** eghobo has quit IRC06:15
*** eghobo has joined #openstack-trove06:15
*** eghobo has quit IRC06:16
*** ramishra has quit IRC06:23
*** ramishra_ has joined #openstack-trove06:28
*** annashen_ has joined #openstack-trove06:42
*** annashen_ has quit IRC06:53
openstackgerritA change was merged to openstack/trove: Logging audit for trove/db module  https://review.openstack.org/10395306:56
*** haomaiwa_ has joined #openstack-trove07:05
*** flaper87|afk is now known as flaper8707:24
*** mariusv has quit IRC08:17
*** mariusv has joined #openstack-trove08:19
*** sgotliv has joined #openstack-trove08:25
*** robertmyers has joined #openstack-trove08:28
*** robertmyers has quit IRC08:32
*** ramishra_ has quit IRC08:33
*** coolsvap is now known as coolsvap|afk08:52
*** ramishra has joined #openstack-trove08:52
*** coolsvap|afk is now known as coolsvap09:04
openstackgerritShalini khandelwal proposed a change to openstack/trove-integration: Initial support for Vertica Datastore in trove  https://review.openstack.org/8980409:27
openstackgerritDenis M. proposed a change to openstack/trove: Mark all running backups as FAILED when instance was deleted  https://review.openstack.org/10283810:02
openstackgerritDenis M. proposed a change to openstack/trove: Update backup state when guest error occures  https://review.openstack.org/10226210:02
*** sgotliv has quit IRC10:22
*** haomaiwa_ has quit IRC10:38
*** haomaiwa_ has joined #openstack-trove10:38
*** demorris has quit IRC10:40
*** boden has joined #openstack-trove10:49
openstackgerritShalini khandelwal proposed a change to openstack/trove: Initial support for Vertica Datastore in trove  https://review.openstack.org/9176110:51
*** haomaiwa_ has quit IRC10:57
*** haomai___ has joined #openstack-trove10:57
*** sgotliv has joined #openstack-trove11:06
*** coolsvap is now known as coolsvap|afk11:08
openstackgerritShashank Gupta proposed a change to openstack/trove: Backup/Restore feature for Vertica Datastore  https://review.openstack.org/9575911:15
*** coolsvap|afk is now known as coolsvap11:16
*** demorris has joined #openstack-trove11:18
*** ramishra has quit IRC11:43
*** coolsvap is now known as coolsvap|afk11:46
*** demorris has quit IRC11:48
*** juantwo has joined #openstack-trove11:55
*** sgotliv has quit IRC12:04
openstackgerritDenis M. proposed a change to openstack/trove: Mark all running backups as FAILED when instance was deleted  https://review.openstack.org/10283812:10
openstackgerritDenis M. proposed a change to openstack/trove: Update backup state when guest error occures  https://review.openstack.org/10226212:10
*** sgotliv has joined #openstack-trove12:17
*** setho has joined #openstack-trove12:26
*** serverascode has quit IRC12:28
*** coolsvapl has joined #openstack-trove12:30
*** zacksh has quit IRC12:30
*** zacksh has joined #openstack-trove12:30
*** setho2 has quit IRC12:30
*** coolsvap|afk has quit IRC12:30
*** coolsvapl has quit IRC12:31
*** coolsvapl has joined #openstack-trove12:31
*** serverascode has joined #openstack-trove12:31
*** ajc_ has quit IRC12:33
*** radez_g0n3 is now known as radez12:34
*** ramishra has joined #openstack-trove12:36
openstackgerritDenis M. proposed a change to openstack/trove: Add backup size validation at restoring  https://review.openstack.org/8630012:38
*** juantwo has quit IRC12:56
*** juantwo has joined #openstack-trove12:57
*** nosnos has quit IRC13:02
*** tvoran has joined #openstack-trove13:08
*** miqui has joined #openstack-trove13:17
openstackgerritOpenStack Proposal Bot proposed a change to openstack/trove: Updated from global requirements  https://review.openstack.org/10517513:33
*** demorris has joined #openstack-trove13:34
*** jasond` has joined #openstack-trove13:49
jasond`hi, where is the trove API documentation?13:49
denis_makogonjasond`, https://github.com/openstack/trove/tree/master/apidocs/src13:56
jasond`denis_makogon: is there a human-readable version somewhere?13:58
denis_makogonjasond`, actually, don't know, but i'll try to find it for you13:58
jasond`denis_makogon: i can probably just use the rackspace cloud databases doc.  thanks though13:59
denis_makogonjasond`, np13:59
jasond`the link here is stale BTW http://docs.openstack.org/developer/trove/13:59
*** grapex has joined #openstack-trove14:00
denis_makogonjasond`, but there's no API docs =)14:01
*** grapex has quit IRC14:01
*** grapex has joined #openstack-trove14:01
*** grapex_ has joined #openstack-trove14:02
openstackgerritDenis M. proposed a change to openstack/trove: Add backup size validation at restoring  https://review.openstack.org/8630014:03
*** grapex has quit IRC14:06
*** yidclare has joined #openstack-trove14:10
*** robertmyers has joined #openstack-trove14:16
*** yidclare has quit IRC14:18
*** mattgriffin has joined #openstack-trove14:25
*** karimb has joined #openstack-trove14:29
*** Longgeek has quit IRC14:34
*** sgotliv has quit IRC14:35
*** Longgeek has joined #openstack-trove14:36
*** juantwo has quit IRC14:42
*** juantwo has joined #openstack-trove14:42
*** juantwo has quit IRC14:44
*** juantwo has joined #openstack-trove14:45
*** IvanZ has joined #openstack-trove14:58
*** ramishra has quit IRC15:02
*** demorris_ has joined #openstack-trove15:04
*** demorris has quit IRC15:04
*** demorris_ is now known as demorris15:04
openstackgerritDenis M. proposed a change to openstack/trove: Fix Cassandra service status definition  https://review.openstack.org/9942415:12
openstackgerritDenis M. proposed a change to openstack/trove: Fix Cassandra service status definition  https://review.openstack.org/9942415:17
*** karimb has quit IRC15:18
*** ramishra has joined #openstack-trove15:18
*** thedodd has joined #openstack-trove15:21
*** doddstack has joined #openstack-trove15:24
*** thedodd has quit IRC15:26
*** ViswaV_ has joined #openstack-trove15:30
*** ViswaV has quit IRC15:32
*** coolsvapl is now known as coolsvap15:37
*** Longgeek has quit IRC15:39
*** rwsu has joined #openstack-trove15:47
*** shayneburgess has joined #openstack-trove15:49
*** IvanZ has quit IRC15:50
*** IvanZ has joined #openstack-trove15:52
*** IvanZ has quit IRC16:03
openstackgerritDenis M. proposed a change to openstack/trove: Mark all running backups as FAILED when instance was deleted  https://review.openstack.org/10283816:03
openstackgerritDenis M. proposed a change to openstack/trove: Update backup state when guest error occures  https://review.openstack.org/10226216:03
*** juantwo has quit IRC16:04
*** juantwo has joined #openstack-trove16:05
*** nehav has joined #openstack-trove16:08
*** doddstack has quit IRC16:16
*** coolsvap is now known as coolsvap|afk16:16
*** cweid has joined #openstack-trove16:18
denis_makogoncores, take a look at https://bugs.launchpad.net/trove/+bug/133866016:20
*** edmund has joined #openstack-trove16:23
*** annashen_ has joined #openstack-trove16:25
*** thedodd has joined #openstack-trove16:34
*** edmund has quit IRC16:35
*** juantwo has quit IRC16:41
*** juantwo has joined #openstack-trove16:43
SlickNikHi folks — it's an hour before the BP meeting, and I don't see anything on the agenda at https://wiki.openstack.org/wiki/Meetings/TroveBPMeeting.17:05
SlickNikSo I'm canceling the meeting for this week to give you back the hour.17:05
denis_makogonSlickNik, take a look at https://bugs.launchpad.net/trove/+bug/133866017:05
SlickNikdenis_makogon: That link seems to say that it affects NFS mounts. Yet we don't use NFS?17:09
denis_makogonSlickNik, nfs was in a discussion, but bug happens in simple devstack installation17:10
denis_makogonSlickNik, i'm only pointing that same thing happens inside dev box17:10
SlickNikdenis_makogon: Does it only affect cassandra? How ofter does it happen?17:11
denis_makogonSlickNik, i was able to reproduce this bug 2-3 time today17:11
SlickNikdenis_makogon: According to that link, the copy succeeded even though rsync errored out. If that's the case, is this even a bug?17:11
denis_makogonSlickNik, yes, because, exit code 24 for rsync also valid17:12
denis_makogonSlickNik, this issue happens with temporary SSTables17:12
SlickNikdenis_makogon: Then perhaps the file is actually going away — and  this has nothing to do with NFS.17:13
denis_makogonSlickNik, yes, nothing about nfs, it's all about rsync, link was like an appendix to bug-report17:14
*** thedodd has quit IRC17:28
*** ramishra has quit IRC17:33
*** yidclare has joined #openstack-trove17:34
*** eghobo has joined #openstack-trove17:35
*** miqui has quit IRC17:45
*** miqui has joined #openstack-trove17:49
openstackgerritMatt Riedemann proposed a change to openstack/trove: Use (# of CPUs) api/conductor workers by default  https://review.openstack.org/10323917:54
amrithdenis_makogon, is this error happening on startup? i.e. when launching a trove instance with volume support?18:02
*** jmontemayor has joined #openstack-trove18:06
*** jmontemayor_ has joined #openstack-trove18:07
espping grapex_18:10
*** jmontemayor has quit IRC18:10
*** eghobo has quit IRC18:11
*** thedodd has joined #openstack-trove18:14
*** eghobo has joined #openstack-trove18:22
*** jmontemayor_ has quit IRC18:22
*** eghobo has quit IRC18:23
grapex_esp: yo18:23
*** eghobo has joined #openstack-trove18:24
esphowdy grapex_ , wanted to ask if you think I should abandon this change and upload a new patch https://review.openstack.org/#/c/85225/18:25
grapex_esp: Sorry... let me review this for a bit so I can remember where it was18:27
grapex_Either way I'm good18:27
grapex_If it's a new patch we lose history18:27
grapex_What would be the upshot? Other than it would seem cleaner18:27
grapex_... I really wish Gerrit would show what changed between what patch and another, without all the other stuff that was checked in18:27
espk, np grapex_ just wanted to make sure I asked ya before I did something wrong…18:28
*** eghobo has quit IRC18:28
espit’s something new I’m trying out18:29
grapex_It would benefit me personally if you uploaded a new patch esp but I don't know if that would be a good precedent, so I'm good either way. SlickNik would have a good answer there for what the official stance is.18:29
espk, I’ll check with Slick :) thx!18:29
*** ViswaV_ has quit IRC18:32
*** miqui has quit IRC18:32
*** sgotliv has joined #openstack-trove18:32
*** ViswaV has joined #openstack-trove18:33
*** amcrn has joined #openstack-trove18:34
*** rueben has joined #openstack-trove18:38
*** radez is now known as radez_g0n318:39
*** miqui has joined #openstack-trove18:41
*** nehav has left #openstack-trove18:42
*** cweid has quit IRC18:46
*** cweid has joined #openstack-trove18:48
openstackgerritA change was merged to openstack/trove-integration: Fix signature for launching int-tests  https://review.openstack.org/10320118:49
*** rueben has quit IRC18:53
*** eghobo has joined #openstack-trove18:56
*** eghobo has quit IRC18:57
*** eghobo has joined #openstack-trove18:58
SlickNikesp: I'm fine either way. Feel free to do  whatever makes your workflow easier. If you do decide to start a new review, just make sure that you put in a code review comment with a link to the last review, though. So it's there for posterity.19:00
*** radez_g0n3 is now known as radez19:00
espk, thx SlickNik I may upload a clean patch before abandoning the other to see if it helps.19:01
espSlickNik: I’m madly debugging oslo.messagin atm19:02
espgetting closer19:02
*** ViswaV_ has joined #openstack-trove19:04
*** ViswaV has quit IRC19:05
grapex_esp: Let me know if you make a new patch19:06
grapex_as I'm commenting on the old one. :/19:06
grapex_But I'm good either way19:06
grapex_I like parts of Gerrit, but I wish it would focus more on the conversation like GitHub does19:06
grapex_Its so painful to dig through old patch sets and figure out what the conversation was...19:07
grapex_esp: One thing maybe have a discussion about here is there are a lot of subclasses in this pull request that don't do anything.19:08
grapex_esp: In a language with a compiler having interfaces is a good thing as it allows the compiler to pin down exactly what the code needs- so if you get the signature of a method wrong, the compiler tells you. But there is no compiler, so the best some of these ABC classes will do is tell you the class is wrong as it's being instantiated, which might be after you've spent ten minutes setting up a VM, making an instance,19:09
grapex_ and trying to run an upgrade.19:09
grapex_esp: So my opinion is that the ABC classes and all these subclasses should go away, and be replaced with the simple "driver" pattern used everywhere, were some random class is instantiated using ImportClass.19:11
grapex_SlickNik vipul amcrn: I'm curious if you all see it that way too or have other opinions here.19:11
*** yidclare has quit IRC19:12
*** michael-yu has joined #openstack-trove19:13
amcrngrapex_: agreed. i generally avoid subclassing even in java unless absolutely necessary, especially in large codebases with dozens of active developers.19:15
*** denis_makogon has quit IRC19:33
*** denis_makogon_ is now known as denis_makogon19:33
*** dmakogon_ has joined #openstack-trove19:33
denis_makogonamrith, yes, it happens when instance was provisioned with volume19:33
amrithok, I know the problem, will send you info.19:34
amrithI have a fix if you want me to submit it19:34
amrithfeel free to assign bug to me.19:34
amrithdenis_makogon, ^^19:34
denis_makogonamrith, of course, feel free to fix it19:34
amrithok, will reassign bug to myself if that's OK with you. thx19:35
amrithpl configm19:35
amrithconfirm19:35
amriththx19:35
denis_makogonamrith, i guess that the simplest fix is to extend exec with appropriate attribute 'check_exit_code' for processutils19:35
denis_makogonamrith, i'm fine that you taking it19:35
amriththanks, will send you code for review19:35
denis_makogonamrith, how do you plan to fix it?19:35
amrithlet me get you a link, one second19:36
denis_makogonamrith, ok19:37
denis_makogonamrith, tomorrow i'll submit new patch for cassandra backup/restore, take a look at it when you'll have time19:37
amrithok, thanks19:37
amrithso in datastore/cassandra/manager.py there is a call to device.migrate_data()19:38
amrithat the point when this happens, cassandra is up and running19:38
amriththis is not good19:38
amrithso after selc.app.install_if_needed(packages)19:38
amrithand before attempting to migrate the data19:38
amriththe database should be stopped19:38
amriththat's it19:38
denis_makogonamcrn, it'll increase time for provisioning19:39
denis_makogonamcrn, and will set datastore to SHUTDOWN state19:39
denis_makogonamcrn, sorry =)19:39
*** amcrn is now known as ghost_of_amcrn19:39
ghost_of_amcrn;)19:39
denis_makogonamrith, so, it's not quite good, i think19:39
amrithno, no, speak with ghost_of_amcrn  ;)19:39
ghost_of_amcrnno worries denis_makogon :)19:39
amrithtake a look at the code once I submit19:40
denis_makogonamrith, there's a fix for cassandra status definition, https://review.openstack.org/#/c/99424/3, so i think you should try your code with it19:41
denis_makogonamrith, why exit code extension doesn't work for you?19:42
amrithbecause that's not a fix19:43
denis_makogonamrith, since rsync remembers temporary SST files19:43
denis_makogonamrith, why is that?19:43
denis_makogonamrith, according to tons of topics related to rsync - exit code 24 is also valid19:44
amrithbecause the problem is that you are attempting to rsync a filesystem while it is being changed.19:44
amrithand rsync is not atomic19:44
amrithin the case of the topic you found, that may have been acceptable19:44
amrithI will put details in the bug and in the commit message, feel free to review and provide input.19:45
denis_makogonamrith, actually i'm not happy with the way how does datastore application and status works, since guest tries to define status of datastore when packages are not even installed19:46
denis_makogonamrith, so, that's why possibly, cassandra instance will be marked as SHUTDOWN until everything will be ready (FS migrations, restoring, etc.)19:47
denis_makogonamrith, what do you think?19:48
amrithWhat do I think about what? that you don't like something (your first comment) or why cassandra may be marked SHUTDOWN.19:50
amrithAbout the latter, I don't know19:50
amrithmaybe it will19:50
amrithmaybe it won't19:50
amrithI don't know for sure19:50
amrithbut, looking at the code19:50
amrithmongodb and mysql already do this (what I'm suggesting)19:50
amrithand redis doesn't appear to have this notion at all19:51
amrithso, let me make the change and then you can comment19:51
amrithok?19:51
denis_makogonamrith, ok19:51
denis_makogoni wonder if it's possible to split up deployment process with status definition19:52
denis_makogonso, guest datastore application will do deployment first, then it'll start to define actual status19:52
denis_makogonprepare call can/might be splitted into two different parts19:53
denis_makogonsomething like guest_api().deploy(...) and only then call prepare19:54
*** michael-yu has quit IRC19:54
*** yidclare has joined #openstack-trove20:04
ghost_of_amcrngrapex_ esp: regarding "Partially implements guest agent upgrade strategy" (i.e. https://review.openstack.org/#/c/85225/), did we consider baking the version into metadata and/or a file within the package itself? this would avoid the necessity in passing the instance_version in the payload.20:07
*** ghost_of_amcrn is now known as amcrn20:07
*** boden has quit IRC20:10
espamcrn: that’s probably a good idea.  we can totally drop the instance version from the payload20:13
espamcrn: one of the things grapex_ pointed out was that we should allow for alternative ways to retrieve the version within trove20:13
grapex_ghost_of_amcrn esp: Yep- originally the implementation got it from pip20:14
espthe reference imple in this patch uses package_resource20:14
espbut I can see other projects use pbr20:14
grapex_In my experience relying on the packaging mechanism isn't as sure-fire as somehow baking it into the code you're deploying20:14
amcrnesp: could we standardize on an idea like: whether it's pip or <whatever>, in the root of the package there must be a file called "version.txt" (or somewhere in the code as grapex suggests above)20:14
grapex_Though that's with Sneaky Pete which is a binary. Still I'd love it if baking it into the code itself was possible20:15
espamcrn +  grapex_: yeah we can figure something out.  not sure I will get to it any time soon though some maybe we can do a follow up if this patch is good enough.20:16
amcrngrapex_: as an aside, i think the next iteration of your guest should be named Predictable Pete; Sneaky is somewhat derogatory.20:16
espI’m getting close on the oslo.messaging stuff20:16
amcrnor Pleasant Pete20:16
grapex_amcrn: You can say that about a dozen more times20:17
esplol, yeah be called sneaky doesn’t really build trust20:17
amcrnesp: thing is, i'm not sure we can do that in another patch, because it'd be considered a breaking change to reject instance_version20:17
grapex_amcrn: Its one of the worst names for anything ever20:17
grapex_hub_cap named it after a bottle of beer in a paper bag20:18
*** rueben has joined #openstack-trove20:18
grapex_which is kind of awesome, but nobody knows that20:18
espamcrn: k, i’ll try get through my tasks and circle back on the version.txt thing or embedded version “string”20:18
grapex_It wasn't until later when we tragically found out that the word "Sneaky" has a negative connotation in English20:18
amcrngrapex_: do you think to ease some of esp's pain that we could merge it and break the contract by maybe having the upgrade feature turned off by default with a comment saying "we reserve the right to break this, it's a wip"20:19
amcrnbecause i have a feeling this version-in-the-package thing might be a decent amount of work/thought20:19
espgrapex_: at least it was a short name, it could have been “Puff the Magic Dragon” or something hideous like that20:20
amcrngrapex_: lol, the backstory on that is hilarious20:20
grapex_amcrn: Sure20:20
grapex_Iff we do agree we can change it20:20
espgrapex_: I’ll sign my name in Vipul’s blood20:20
grapex_amcrn: I really like that idea, I feel like too often we tie our own hands around here to the first draft of ideas or code20:20
grapex_esp: Ha ha! Good joke esp20:21
* grapex_ picks up phone to call FBI20:21
amcrnperfect-o, so once that abc stuff gets removed, it's good to go20:21
espgrapex_: he uses my blood all the time..20:21
amcrnesp: is that how vipul gains his powers?20:21
grapex_esp: It's good to know you two are close.20:21
espamcrn: I like to think so20:21
* amcrn calls up WB to kick-start a new series20:22
* esp dreams up an episode of Buffy the Vampire slayer feature esp and vipul20:24
grapex_amcrn: Finally a show that combines pale hipster vampires with pale hipster software developers20:24
esplol20:25
grapex_The product managers should be werewolves20:25
amcrnPerson A: "I want to suck your blood!".........    Person B: "VC Firm XYZ already did that in our last round of funding!"    <Insert Laugh Track>20:25
grapex_I guess upper management would be demons or something20:25
openstackgerritA change was merged to openstack/trove: Imported Translations from Transifex  https://review.openstack.org/10507020:25
vipulhaha natural enemies!20:26
grapex_amcrn: lol!20:26
espooh, our product dude does have were wolve like eye brows20:26
amcrnlol20:26
grapex_esp: lol20:26
grapex_I know his pain20:26
esphaha20:26
espI would send a pic but I don’t want to start us down that road20:27
grapex_One time I saw Timothy Dalton in public and he yelled out "what's up Bert!"20:27
openstackgerritPeter Stachowski proposed a change to openstack/trove: Fixed '--version' for trove processes/utilities  https://review.openstack.org/10526620:28
*** michael-yu has joined #openstack-trove20:28
denis_makogonamrith, ping20:41
denis_makogonamrith, do you really think that your comments are real reason for -1? https://review.openstack.org/#/c/97510/1320:42
*** rueben has quit IRC20:43
denis_makogonamrith, there's no difference (only 9 sec), so, it's like whishlist things, so, are there any significant reasons to hold this patchset?20:45
*** denis_makogon has quit IRC20:47
*** miqui has quit IRC20:49
*** juantwo has quit IRC21:00
*** rueben has joined #openstack-trove21:02
openstackgerritA change was merged to openstack/trove: Add guestagent API calls for replication  https://review.openstack.org/9190421:05
*** flaper87 is now known as flaper87|afk21:17
openstackgerritDan Nguyen proposed a change to openstack/trove: Updates RPC API to use oslo.messaging  https://review.openstack.org/9448421:23
*** sgotliv has quit IRC21:23
*** radez is now known as radez_g0n321:26
*** karimb has joined #openstack-trove21:30
espSlickNik: ping21:30
SlickNikesp: what's up?21:31
espSlickNik: hey dude, was wondering if you prefer a separate bp for trove-int regarding the oslo work21:33
espI need to add oslo.messaging to guest agent in trove-int21:33
esphttps://blueprints.launchpad.net/trove/+spec/rpc-versioning - that’s the one for trove21:35
SlickNikesp: Just use the same BP21:35
SlickNik(in the patchset you submit for the change to trove-integration)21:36
espthx SlickNik21:36
SlickNiknp, that should simplify things.21:36
*** jasond` has quit IRC21:36
openstackgerritDan Nguyen proposed a change to openstack/trove-integration: Add oslo.messaging to guestagent image  https://review.openstack.org/10528221:39
*** michael-yu has quit IRC21:43
*** shayneburgess has left #openstack-trove21:56
*** michael-yu has joined #openstack-trove22:00
*** rueben has quit IRC22:11
openstackgerritDan Nguyen proposed a change to openstack/trove: Updates RPC API to use oslo.messaging  https://review.openstack.org/9448422:15
*** robertmyers has quit IRC22:16
*** yidclare has quit IRC22:18
*** yidclare has joined #openstack-trove22:19
*** juantwo has joined #openstack-trove22:24
*** karimb has quit IRC22:24
*** juantwo has quit IRC22:30
*** juantwo has joined #openstack-trove22:31
*** yidclare has quit IRC22:32
*** yidclare has joined #openstack-trove22:32
*** yidclare has quit IRC22:33
*** thedodd has quit IRC22:38
openstackgerritSimon Chang proposed a change to openstack/trove: Adjusted audit logging for trove.instance module  https://review.openstack.org/10529422:39
*** demorris has quit IRC22:51
*** mattgriffin has quit IRC23:00
*** tvoran has quit IRC23:00
*** grapex_ has quit IRC23:18
amrithdmakogon_, you there?23:39
amrithdenis, you asked about why I -1'ed https://review.openstack.org/#/c/97510/1323:48
amrithCurrently, int-tests runs for about <> minutes. Adding another 18s to it may appear insignificant but realize that this is 18s for each person who runs int-tests, 18s on each iteration through gate and so on. It may appear like an insignificant amount of time but if each test just added a second here and a second there, pretty soon you are seeing a significant increase in the time for int-tests as a whole.23:48
amrithSecond, in an earlier version of the code you had 20s and I asked why. You reduced it to 1s. Now you've increased it to 10s. What I'd like to know is this; if someone tells me that I have to wait an extra 18s each time I run int-tests, I'd like to know that it is for a good reason. If you were, say, waiting 10s and then checking to make sure that the delta between the timestamps was 10s, that's one thing. But, we're just23:48
amrith sitting there for 10s for no good reason.23:48
amrithSo, maybe you think this is a wishlist, that's fine. But is there a reason for the 10s; that's the question. Why not 1?23:48
*** michael-yu has quit IRC23:55
*** michael-yu has joined #openstack-trove23:58
« Sunday, 2014-07-06 Index Tuesday, 2014-07-08 »

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