| *** cpallares has quit IRC | 00:07 | |
| *** flwang has quit IRC | 00:15 | |
| *** flwang has joined #openstack-zaqar | 00:15 | |
| *** sriram has quit IRC | 00:51 | |
| *** jeffrey4l_ has joined #openstack-zaqar | 01:04 | |
| *** prashanthr_ has joined #openstack-zaqar | 01:04 | |
| *** alcabrera|afk is now known as alcabrera | 01:37 | |
| *** vkmc has quit IRC | 01:41 | |
| *** vkmc has joined #openstack-zaqar | 01:46 | |
| *** jeffrey4l_ is now known as jeffrey4l | 01:53 | |
| openstackgerrit | Jeffrey Zhang proposed a change to openstack/python-zaqarclient: Remove the grace in claim update method https://review.openstack.org/122591 | 02:09 |
|---|---|---|
| *** jeffrey4l has quit IRC | 02:13 | |
| *** jeffrey4l has joined #openstack-zaqar | 02:25 | |
| openstackgerrit | Jeffrey Zhang proposed a change to openstack/python-zaqarclient: Remove the grace in claim update method https://review.openstack.org/122591 | 03:20 |
| *** jeffrey4l has quit IRC | 03:25 | |
| *** flwang has quit IRC | 03:25 | |
| *** flwang has joined #openstack-zaqar | 03:25 | |
| *** jeffrey4l has joined #openstack-zaqar | 03:41 | |
| *** vkmc has quit IRC | 04:31 | |
| *** alcabrera is now known as alcabrera|afk | 04:33 | |
| *** reed has quit IRC | 04:49 | |
| openstackgerrit | Jeffrey Zhang proposed a change to openstack/python-zaqarclient: CLI support for claim CRUD operations https://review.openstack.org/122611 | 05:34 |
| *** jeffrey4l has quit IRC | 05:34 | |
| *** jeffrey4l has joined #openstack-zaqar | 05:35 | |
| *** flwang1 has quit IRC | 06:03 | |
| *** flaper87|afk is now known as flaper87 | 06:36 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/zaqar: Imported Translations from Transifex https://review.openstack.org/122070 | 06:37 |
| jeffrey4l | git log | 06:51 |
| jeffrey4l | oops | 06:51 |
| *** jeffrey4l has quit IRC | 08:01 | |
| *** jeffrey4l has joined #openstack-zaqar | 08:13 | |
| flaper87 | jeffrey4l: there ? | 08:43 |
| flaper87 | jeffrey4l: since you're working on adding CLI support to the client | 08:43 |
| flaper87 | I wanted to ask if you could work on pool and flavors too | 08:43 |
| flaper87 | I think it's important to allow operators to easily create pools and flavors asap | 08:44 |
| flaper87 | I'll cut the client version now | 08:44 |
| flaper87 | we can land the rest of the CLI support later and release a minor | 08:44 |
| openstackgerrit | Flavio Percoco proposed a change to openstack/python-zaqarclient: Add a read-only property for Queues https://review.openstack.org/121478 | 08:51 |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/python-zaqarclient: Updated from global requirements https://review.openstack.org/122670 | 08:51 |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/zaqar: Updated from global requirements https://review.openstack.org/122676 | 08:52 |
| openstackgerrit | A change was merged to openstack/python-zaqarclient: Updated from global requirements https://review.openstack.org/122670 | 09:43 |
| openstackgerrit | Flavio Percoco proposed a change to openstack/zaqar-specs: Define a wire protocol for non-REST APIs https://review.openstack.org/122425 | 09:57 |
| *** prashanthr_1 has joined #openstack-zaqar | 10:01 | |
| *** prashanthr_ has quit IRC | 10:02 | |
| *** flwang1 has joined #openstack-zaqar | 10:07 | |
| *** prashanthr_1 has quit IRC | 10:45 | |
| *** prashanthr_ has joined #openstack-zaqar | 10:46 | |
| *** prashanthr_1 has joined #openstack-zaqar | 10:58 | |
| *** prashanthr_ has quit IRC | 11:00 | |
| *** vkmc has joined #openstack-zaqar | 11:16 | |
| *** vkmc has quit IRC | 11:16 | |
| *** vkmc has joined #openstack-zaqar | 11:16 | |
| vkmc | hellooooooo | 11:18 |
| vkmc | flaper87, around? | 11:23 |
| flaper87 | vkmc: gooooooooooood morning | 11:24 |
| flaper87 | vkmc: I think so | 11:24 |
| flaper87 | :P | 11:24 |
| vkmc | flaper87, :D gooood morning | 11:24 |
| *** prashanthr_ has joined #openstack-zaqar | 11:24 | |
| vkmc | flaper87, all good? | 11:24 |
| vkmc | I started working on adding the client to rally last night | 11:24 |
| flaper87 | vkmc: all good, you? | 11:25 |
| flaper87 | vkmc: cooooool | 11:25 |
| *** prashanthr_1 has quit IRC | 11:25 | |
| flaper87 | vkmc: I just released the client | 11:25 |
| vkmc | flaper87, wee :) great | 11:26 |
| vkmc | I'm doing good too | 11:26 |
| flaper87 | vkmc: any issues so far? | 11:26 |
| vkmc | so I have some questios | 11:26 |
| flaper87 | shoot | 11:26 |
| vkmc | questions* | 11:26 |
| vkmc | heh | 11:26 |
| vkmc | in most of the clients (only to avoid saying in all of them) | 11:27 |
| vkmc | the version is the first parameter received to instantiate the client | 11:27 |
| vkmc | in Zaqar's its not | 11:27 |
| flaper87 | yup | 11:28 |
| flaper87 | but it's a keyword | 11:28 |
| flaper87 | right? | 11:28 |
| vkmc | it is yeah | 11:28 |
| vkmc | makes it sense to change it or we go ahead without that? | 11:28 |
| flaper87 | I don't think we need to change it | 11:29 |
| vkmc | this is the list of clients boris showed to us yesterday https://github.com/stackforge/rally/blob/master/rally/osclients.py | 11:29 |
| flaper87 | I made it a keyword on purpose | 11:29 |
| flaper87 | I really wish people would stop using arguments | 11:29 |
| flaper87 | it makes reading function calls hard if you don't know the API | 11:29 |
| flaper87 | I man, arguments are fine | 11:29 |
| flaper87 | but when calling functions, I wish people would use them as kwargs | 11:30 |
| flaper87 | call_my_func(version=0.1) | 11:30 |
| flaper87 | that's more readable than: | 11:30 |
| flaper87 | call_my_func(0.1) | 11:30 |
| vkmc | totally yeah | 11:30 |
| vkmc | so I checked if there was some standard for clients | 11:31 |
| vkmc | just in case | 11:31 |
| vkmc | and it's not | 11:31 |
| flaper87 | LOL, standard in clients... | 11:31 |
| vkmc | that's why I'm asking why you took that design decision | 11:31 |
| vkmc | there are standards for everything | 11:32 |
| flaper87 | sorry but clients is one of the things in openstack that keeps surprising me everytime | 11:32 |
| flaper87 | yeah, but not in os clients | 11:32 |
| vkmc | cool | 11:32 |
| flaper87 | everyone does things differently | 11:32 |
| flaper87 | which is why we're fully relying on os-common-client | 11:32 |
| vkmc | yeah I assumed so by looking at the code of the other clients | 11:33 |
| vkmc | awesome | 11:33 |
| flaper87 | I hope it'll become THE standard very soon | 11:33 |
| vkmc | yup | 11:33 |
| vkmc | so well, then I go ahead adding it as is :) | 11:34 |
| vkmc | on the other hand... I might have a mentee for the next OPW round | 11:39 |
| vkmc | let me know if you have a project idea in mind | 11:41 |
| vkmc | flaper87, ^ | 11:47 |
| flaper87 | vkmc: w00000000000t | 11:53 |
| flaper87 | that sounds great | 11:53 |
| flaper87 | yeah, I've a couple of ideas, I've been meaning to write them down | 11:53 |
| vkmc | awesome! | 11:53 |
| flaper87 | notmyname: ping when you are around to talk about rings and cool stuff | 11:53 |
| flaper87 | (and not exactly this ring http://s.f87.me/XtTLH2 ) | 11:54 |
| vkmc | da ring | 11:54 |
| flaper87 | vkmc: mmh, something stinks here: http://logs.openstack.org/67/120867/3/check/gate-zaqar-python27/6857254/console.html | 12:11 |
| flaper87 | wondering why mongo is not running | 12:11 |
| vkmc | yeah, all fixes are hitting that | 12:11 |
| flaper87 | prashanthr_: congrats ;) | 12:12 |
| prashanthr_ | flaper87: Thank you soo much :) | 12:12 |
| prashanthr_ | Excited to meet all of you | 12:13 |
| flaper87 | prashanthr_: and I'm really exited to have you there | 12:13 |
| prashanthr_ | ha ha :) | 12:13 |
| * vkmc is the snitcher | 12:14 | |
| prashanthr_ | vkmc: No doubt about that. I am very excited about the toffies as well :P | 12:14 |
| flaper87 | I don't care if you drink or not or whether alcohol can cause you an anaphylactic shock, I'm so buying you a beer | 12:14 |
| vkmc | shh flaper87 will hear | 12:14 |
| * vkmc lowers her voice | 12:14 | |
| * prashanthr_ tries to delete the message | 12:15 | |
| flaper87 | *cough* *cough* | 12:15 |
| flaper87 | toooooooooooooooooo late | 12:15 |
| flaper87 | vkmc: prashanthr_ | 12:16 |
| prashanthr_ | Gosshhh, I thought vkmc would have already informed you. | 12:16 |
| flaper87 | data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/2wCEAAkGBxQTEhQUEhQUFRUUGRcVFhcXGBgVFhQYGBUYFxgUFBcYHSkgGBomHBcUITEiJSktLi4uFx8zODMsNygtLisBCgoKDg0OGxAQGywkHyU2LDQsLC0sMiw0LC0sLCwsLCwsLCwsLywsLCwsLCwsLC0sLCwsLCwvLCwsLDQsLCwsLP/AABEIAKoBKQMBIgACEQEDEQH/xAAcAAACAgMBAQAAAAAAAAAAAAAABgUIAwQHAgH/xABQEAABAwICAwoKBQgJAwUAAAABAAIDBBEFIQYSMQcTIkFRVGFxkZMXMjVScoGxs9LTFBYjobIVMzRCYnPB0URTY3WCg+Hw8SWS4iRDZKKj | 12:16 |
| flaper87 | /8QAGgEBAAIDAQAAAAAAAAAAAAAAAAEDAgQFBv/EACwRAAICAQQBAwIFBQAAAAAAAAABAhEDBBIhMVEFE0EicRQVYbHRM4GhwfD/2gAMAwEAAhEDEQA/AIbdG3RMRpsSqYYJwyONwDW7zA6wLGnxnRknMnaUt+FfFect7im+UsO655XrPTb7tiUEA6+FfFect7im+Ujwr4rzlvcU3ylFUmhOISsbJHRzvY8BzXBhIc0i4IPIsvg/xPmNT3ZQEh4V8V5y3uKb5SPCvivOW9xTfKUf4P8AE+Y1PdlHg/xPmNT3ZQEh4V8V5y3uKb5SPCvivOW9xTfKUf4P8T5jU92VD4thE9M/e6iJ8TyA4NeC06pJANjxXB7EA0eFfFect7im+Ujwr4rzlvcU3yklIQD | 12:16 |
| flaper87 | r4V8V5y3uKb5SPCvivOW9xTfKSUhAOvhXxXnLe4pvlI8K+K85b3FN8pJSEA6+FfFect7im+Ujwr4rzlvcU3yklIQDr4V8V5y3uKb5SPCvivOW9xTfKSUhAOvhXxXnLe4pvlI8K+K85b3FN8pJSEA6+FfFect7im+Ujwr4rzlvcU3yklIQDr4V8V5y3uKb5SPCvivOW9xTfKSUhAOvhXxXnLe4pvlI8K+K85b3FN8pJSEA6+FfFect7im+Ujwr4rzlvcU3yklIQDr4V8V5y3uKb5SPCvivOW9xTfKSUhAOvhXxXnLe4pvlI8K+K85b3FN8pJSEA6+FfFect7im+Ujwr4rzlvcU3yklIQDr4V8V5y3uKb5SPCvivOW9xTfKSUhAdV0B3RMRqa+CGacOje | 12:16 |
| flaper87 | Xhzd5gbe0b3DhNjBGYGwru2/O5SqwblflWl65PcvVnFAKzbrnles9Nvu2JQTfuueV6z02+7YlBSB609mlAwtsTpM8NpjZhd58udglOWoqGi7nTNHSXj2py0wxealfhctPI6KT8mU7dZuRsXy3CbcTqH1mPto6p7paaNsc7Kdx4EkjaFsgbbpcXEjjuRxoDjxrJ7B2+S2Ow6zrHqN18bXzEgCSUk5ABzrnoGaf9HdP8Qqa+KnnImgnlbDJSmJm9CNztUtazVu3UGd9vBzvmt3ELYTRzzYdbfZK6opTUWD3wxRE6kLC4HVLgLk8duqwHM310wNjJKCNoLnAj703brJvLQE5k4fSex63KrEpMRweqnrbPmo5YGwVBaGvkEri2SBzgBrhos/lGsFpbq/52g | 12:16 |
| flaper87 | /u+k9j0AjoQhACEIQAhCEAIQhACEIQAhCEAIQhACEIQAhCEAIQhACEIQAhCEAIQhACEIQDZuV+VaXrk9y9WcVY9yvyrS9cnuXqziArNuueV6z02+7YlBN+655XrPTb7tiUEA5bpGzC/7tpvxzL1ukVMsOLySNkIljFK5r2jUIIpYSCBc24l53SNmF/3bTfjmUtp5hUFVVuqmV9Fvcn0VmoJHb80b3DA9xbqaoDbOceFsHqQDDjGOz0sdBUVAihNfE909XT0sYq2m1w1rzldzXR3OR8YiyjquanwoNipp5pzWNgldBVQxyU0kcuYdIL3EgHJ2lZd0TSOgrKSqjp5Hh9PPE+ISFgZI0MFMRS6uZZqsa862ed+gKumuIxSVNA6ORrmx0tGx5BuGuYOG08h | 12:16 |
| flaper87 | HGgGLdcraYa9G2aVr6VwDKeOCKGkDjbWfwDcu1XHM9Oy5UFur/naD+76T2PTHuq41LO2pMeJUUtI5zCymZqmcgObbMRXyddx4ewepLm6v+doP7vpPY9AI6EIQAhC9xxlxs0EnkAufuQHhCkI8FmOeoWj9ohv3E3WrU0zozqvFjt4j94U0yaZhQhCggz0VMZHtYNrjbPi6UyDQ7+2/wDz/wDNRmiP6XF/j925dFDFlFWdLRaWGWDlLyJh0P8A7X/6f+SW6ynMb3MO1pt18h7F1dzFzPSFtqiXr/gklQ1umhiinEjkIQsTmghCEAIQhACEIQAhCEAIQhACEIQAhCEA2blflWl65PcvVnFWPcr8q0vXJ7l6s4gKzbrnles9Nvu2JQTfuueV6z02+7YlBAd | 12:16 |
| flaper87 | Q0p0OrKyLDJKaJr2DD6ZhJlhjOtrSG2rI8E5EZ2t2JfO5jie3eGWyH6RTbSQ0D87ykDrIUnh+6m+OGGJ1FSy7zD9GD36+sY9XVc02dbNbNNuvOjLiygpGl4iDyDLdwgAEQdws9UAAICC8GmJf1MfGf0ml2DafzvFxr7JuZYk0AugjaDmCamlFwQSCLy5izXH/AAnkU6N2CTP/ANHBmLH7SfMAaovw87AkDkvkvL9115veipzrNDDd854Ia9ob4+wNlkH+IoCEfuY4mDYwMBzyNRTA5C5y33kIPrW5uvwuZPRMcAHMoKVrgCHAEa4IDmkg5jaCQpKo3YZHgtfRU5a7WBbrzap1jdwI17G54XXntSbpbpGa6WN5ijhEUTIGMj1tVrI9bVA1iTsdb1ICDQ | 12:16 |
| flaper87 | hCAE7h9gAzJthYDIWtlsSQnHBnB8LDxjgn1bPusrsPZdh7MihtIosmu9R9oTBFCXfm2veNt2tJFrht77DmQMuVaePUTt7e17S1zbEtO0WzseQ2Ksm4yTinyXTppoT0IQtU0yZ0Q/TIut34HLpmouZ6ID/1kXW78Dl0OJkskpaJGRMGoA5zbi7r6xJJ2AC2zaRdZKSirZ1tFlWPC5Pz/pGd4XMtKW2qpOkg/cF0Z9O9puZ2yjXe3VDWg6otqvuDszPXbZbNc80u/Sn+r2I5KStDW5VlwqS8/P2IZCELE5IIWzQ0L5Xasbbnj5AOUniW5XYBNGLlusOMszt1jasXOKdXyZrHNx3JcEUhCFkYAhT2iGDxVMrhM5zWNbrHV2k6wAF9U2GZ4l0TDtzyhd/WO | 12:16 |
| flaper87 | /zbeyMITRx1C7ZUbmlCB4sg/wA7+bFFV2g1BGL/AGvfD5aCjlCFPaS4XDFYwl1r2Ic4P4rgghreQ9qgUIBCEIAQhCAbNyvyrS9cnuXqzirHuV+VaXrk9y9WcQFZt1zyvWem33bEoJv3XPK9Z6bfdsSggBCEIAQhCAEIQgBMlLok5zWuMjRrAHYTtF+hLa6vg7LwRHlYz8IWUVZvaHBDLJqQsxaGt/WkcepoHtJWSkohC98TSSMnAnaTsOxODYEsaSNMczX9HaP9gq2NRdm9m0mPHC4rk26fEZI4jExwDCNUiwNxrl4B1rjIuOyy0q2Yv1i9xcXXuSbk3XiCV0rgyJhc49g6SeIdKacNw2KF7GSPa6dwLgOQDbqD+JzOarzZ8WBcLnwjmTnHH9zn2HaI | 12:16 |
| flaper87 | zym5bvbPOftI/ZbtP3JhbodCxues93KTYeoBOhHEsMsa4k9Xkk/Bo7+TneHYdvFdCR4pLh1EscE9XIUfV0dpGPA2OH35fxWw5vX2rq6XLvhZ6D01p43Xkyylc00s/SX9TfwhdDLR0rn+mDbVJ6Q32WV8+ifUv6S+5CLbwygfPII2DM7TxNHG49CwQQOebMaXHoF107RrBRBELAF5sXnjJ5OoLS1Oo9qPHZztJpvenz0uz3heFMhjDGDrPG48pWy6FbuqsVQ4NFyvNvJOU7vk9IscYxpdCzjWAwvBceA7zhlf0hsKR66kMT9UkHjBGwj+Cdqycvdc7OIciUcblDpTb9UBvXb/AJ+5d3RTyP6ZOzi6/FjityVMl9BamNkjw82Lw0MzIubm4y49i7Rhjmx | 12:16 |
| flaper87 | Aa8oj6HSOaex72rnW5Judurnipn4NNG7IHbM9ueqP2Adp9XKrCmOYbDF+E/dddA5diFXVsRNt/YT++YPv35LeNMa4WbIwnk39j/uDnLrUrZibnej/AInfyXiqppHC94x2n2oCrulIDQATd18szkBtKXF2PdL0A1g+phd9o0FzmZBrgLk6tthXHEIBCEIAQhCAbNyvyrS9cnuXqzirHuV+VaXrk9y9WcQFZt1zyvWem33bEoJv3XPK9Z6bfdsSggBCEIAQhCAEIWSKIuNh/wAIDyxpJsM11bRXhUsJOXBt2Ej+C5zHEG7NqYsMxf7IQPJaLjUcCW2GsCWOI2A59qyjwb2imscm34HepkjjF5HtA/acGqB0rgDmMcNhGR25XUnFRRWaWxx3BeSbX1tfZx | 12:16 |
| flaper87 | XFsuP18aw4xT/YaoJOrxnbn/ym6TT4o6SnlyRlujSJXAcOZDC3extALncbjbaT/BeK6mYwSvAa10mqdc2BDm2DdZ5zDQBcAbTcZ3C+aN1V4YydhFj0EZKVmLQCXW1elcBZJRyNvnyea3NS5NKhkMsbXlrmOzuCLZg2uOg7fWtXEcUZGDcgkbeQf75FHaQ6TBo1RlyNHjHpPIEqCGoqTdsbnDisLMHrORPSroabc7fCMo475ZJVGNvkeLHVbcZcufGpyR+ZUONE5mROfKLZGzRt7VuGYLqafHtTo7Hp8lBSX2NkvSPpbGXVLQ0XLmtt2lNbqkDjWfAMHFXVXB/NsBceQFzrW6TYrLUZFjg5P4NjV1kx1+pMbnOjQjju4X1vGPnf6LNjmjDqZ0k8T5NV2 | 12:16 |
| flaper87 | qRYGQg+KI3N/qxcOuGk7c9ifqKkDGhrRYBbgjFrHMLzX4qbyOfn4NOuKXByuhqnP1hIzVczJ1iHNPS0/wA/vGahsVrdc2GwZD+aedJtEiGvfSZa2bo9gPS3kK5o+4cQ4EEGxByI61bHZOTlE6WHM5QSbt/Jgrp97Y53IMuviSY43zPGpjSGs1nBg2NzPWoZdjS49kLfbOTrc3uZKXSLP7m1QY8MpWta229A7ON1yT2kpoixIja2/wByV9z5v/TqT9yz2JNqd110c0kf0Rjt7e+O+/FpOq4tvbezttdbKRpnUhiDrrzU4i63GuTHdh/+GL8f2+Xul8ZutmSRkf0Ro13tZffibazgLkb2OVS0Bs0nxB28SA53a4X9RVcFYjSkfZPHQVXdQGCEIQgEIQgG | 12:16 |
| flaper87 | zcr8q0vXJ7l6s4qx7lflWl65PcvVnEBWbdc8r1npt92xKCb91zyvWem33bEoIAQhCA+r4hCAFIUsg1bDI8fSo9emOsbjahKdExFCtsRC1isFDUhw6RtCzyPWRabuFYy6A6r7uj4jxt/mEzMrBK2w2EbeVJEdM+QOLGlwaLusL25Lr7h9e6I5Zt4x/EcilS+Df02s2/TPoeNGpbMkZ5rzkMzwgHWA9amzo5WVQ4LDE3iLxYgcoacyevsKktzPHQ6F0bbAh2tsF+Ftun0Skg5rSWlipuTOdlglkZz7CNzOGI687t8ftJcdbPoTJHBA1rt6DCWbcwSDbK44tnGveKN1o3t43NeO0EJI0IiZFJLHFrBssMcmbWsBLHOYXADbfX2ni1VsfNGJMYmd8ab8Y/g | 12:17 |
| flaper87 | kPRmgZJrGRjXZkZi+xOtdVtjYXONgL+s8g6Uu6NDguNrXc49put3TpWVZZNLg3/q/TO2ws9QsveD01PTzk07g2VuT2B20Ea1nsvyZhbzCo3Rwu18QeCdUziM5Ag6jWgWN7tPqIKs1CjSi0nborhOXLvo6FhtYyVt27R4zeMf6dK2nZLnhqHRv12Ehw4/4HlCacIx5s4ANmycbeJ3S3+S8x6h6XLT/AFw5j+33/k38GpU+H2ScstrkmwGZ5AOlJVdSUeJNdJTvaXsNi5uR2m2sCLlptkdh4ll0wqnPljpCHtikGtK4DOZtyDDGb7RbWI4xYcZW9h1C2AF1hrEat7AEMBJaw25L5/6XWgsahj9xvl9L+S9Te/g5Fpbos5riWts8bQNj+kdKS7WKsLXBsg | 12:17 |
| flaper87 | IeAQuZabaNtY107DsIuOW5tc9K6ei1TdQkVZsd/UjtWgbQMPorccEV+vVC5WzErMlinf8AYHEtSVuV94L5JJWiw1gLi5tnlkuq6Aj/AKfRfuYvwhcYqmNkhrGCWFkja2SYMkkEZe0Nmb9mD4x1nAWXVRrEmyl1r/lFtNvYkh3neBCODvv2gj+j8I0+9X8b9m3CusFRFLwfpbacWnpxSbyIMvtuHqGDPedS3j8erx3W7iOKwwzXo6toaKOenO9vczWkp43Mp3HZrOIc3VdxkGy0oa0Pp6Bu+hzt9Zrt+lWOs6tc4l9La73EEHXJ4+hS+gPWlklo39RVfCFYHTWwjf1FV/k2nrKxDPKEIQgEIQgGzcr8q0vXJ7l6s4qx7lflWl65PcvVnEBWbdc8r1npt | 12:17 |
| flaper87 | 92xKCb91zyvWem33bEoIAQhCAEIQgBfV8QgMkUpaQRxKYuS1riCA7MXFri+ZHLxqDVlcIwmCWmhE0MUgEbBZ7GutwRsuMlranUrAk2rstxR3HHsTxhgibDTjVFjrvudZzXG+o7lI2E57NqXy5WH+pWHnP6JD/25dmxZI9GKNniU0Lepjf5LR/NYRXEWXrCzgeDY5JTPD43gW2gnIjkXbNDtNoqllnXY8bQQbHpadhX3Gqqlo2B8jQxpdqizS4lxBNgGjkB7FjfiAkhbJE46jwHNOYuDsyOxT+YyavZx5/5EPEnxZt4zpFTwAullawZ7druhrRmfUuYYhugQQk/QYXOdYtE0xJs0kEtY3aG3a3LLYFpbpjiRCSSc3ewJEXRwy3wUma81TokcRxyomdrS | 12:17 |
| flaper87 | SuJGwXsB1AZJj0e0hnijBc1srTewuGuFjxnt4klplw/8yzqPtK3dMt06spyPgb6fTdn68MreoBw9qzU+klI0uLWPYXuLncB2bjtJ2pUYVkccl0fb8v8AwUJoa3aQwO/X7Q4e0LDLjUO0SC42WvcJNeTewbfK56AsUjlH0ytX12RVHVcI07iyZUPb0Scnp8npdvKpyum1swRY7LG4ty341wklSuC6Qz0/BadaP+rdew9A7W+zoXntb6RFy34eP0+P7eP2Ohh1HFSOlVVQANtgNpXKNLdJDUO1GXETT/3kfrHo5Fl0q0pfP9mwajMtbO5cbXsTye1Kyq0ul9v6pdmWXLfCLT6Bi2HUf7mL8ASdV7lrHySOJlGu6R+UsZF3FzhkYgbE6vHlrcdjd10Nbag | 12:17 |
| flaper87 | pB/YxfgCmCFvLgpOR+Cdt/GqOPY6A8WXJ0BZIdy8RSRSfal0b2vsXwhvBffibciwBt6uldQeSCsVTc7VNgS9Nm3jPrXAZ/Gd1n2qw2lUV2HLiPsVeanx3dZ9qgMxoQhCAQhCAbNyvyrS9cnuXqzirHuV+VaXrk9y9WcQFZt1zyvWem33bEoJv3XPK9Z6bfdsSggBCEIAQhCAEIQgPTNo61Z7R6S8Efot9irACusaH6WzRxMD9V4ttOR7R/Jc71HFLJBV8F2GST5OvhYpEqRacstwmH1G6+v02i80rivT5PBtqcTX0/Dd7h18midpJ1g3/ANuTjOV+jjNl4owPosdrkauV7E2LjYm2V7LWxbTCEsJfFrgWNi3WzvYWBG3NRFVpYJI/s26o2AWAtbi2rb | 12:17 |
| flaper87 | jjyPEoben2V3FSuxe3SPEi9I+xISntI3vfw3OJAOQ5FArtYI7YJGrkdyBMWGO+yb6/aUurPAHcRstzBk2SuiqatUM8ayOKgWSv85et+k85b/4leGUe2MmFVrIxKHhpL22F23ttzDhsINrBQTlgNQf9/wDCxGYquM4RlJq+SXFuv0NolfWlaDpCvDnFYTzeEWQjRiqjd7utYl6e221eVolh33Q/TKA08EbpdTUYxh8UgarQOW6dKHF6VwNqlzuppPsCqzSwSbW3F/v9SmqL6SNkhb1oCxhroL5TTHp1T8K84pizCOBLI4gZAxut2gLiVNitY0W+kEeoLzU4nWOGdY72IBu0j0haGuDzxG20e05rikz7uJ5ST2lSuJxSuPCkL1EFAfEIQgBCEIBs3K/Kt | 12:17 |
| flaper87 | L1ye5erOKse5X5VpeuT3L1ZxAVm3XPK9Z6bfdsSgm/dc8r1npt92xKCAEIQgBCEIAQhCAE5YPLaJnUk1TdJiIZG3aeLsVWWO5GUWM4nXwzpbON/sHtX0Y2PNK1/Zl4M9xKYtN9m7j8X8QWCjqLx35SdmQ5FpfldvSvTa5rsgT2FZKD21RF82YsXfwPWFCKVxBpLcs81FkLYgqRhLs+LbpjwfWtRbMJ4Ktg+TFmcOXrWWDXHKvu+BX7jGjZppg12d9h2ZffxLWe7Mryc0FYrtsMLouvNlsw0l83ZDk4/9FjKaRkkaU+1Y1J1NOCMsiNn+qjSLKlSsyaHLCGARMOV7DiUmyqcP1lBUU/Ab1LP9JUkEq6rdyjsRJUkjxvuUQahfHToDLUgZ8aU6wcMqeqZ | 12:17 |
| flaper87 | 8lAVB4RQGJCEIAQhCAbNyvyrS9cnuXqzirHuV+VaXrk9y9WcQFdt1TBKmTFat8dPO9pe2zmxPc0/Zt2ECxSp9XKzmtT3Mnwq3E8ztY8I7eUrHv7vOd2lAVL+rlZzWp7mT4V8+rtXzWp7mT4Vbbf3ec7tKN/d5zu0oCpX1cq+a1PcyfCj6uVnNanuZPhVtd/d5zu0o393nO7SgKlfVys5rU9zJ8K+fV2r5rU9zJ8Kttv7vOd2lG/u853aUBUn6vVfNajuZPhXr8gVXNqjupPhVrpZ3X8Z3aV4393nO7SgKsfkKqIsaWo7mT4Vj/IFXzWo7qT4Vavf3ec7tK+b+7zndpQkquzR6q5tUdzJ8K3YcFqW7KWp7mX4VaGOd3nO7Sve/u853aUBV04TVc2qe5l | 12:17 |
| flaper87 | +FYX4NUnbSVB/yZL/AIVajf3ec7tK8id3nO7SoIKoSaN1XFTVPcyfCvUej9TzWpP+TJ8Ktbv7vOd2leRO7zndpUoFWRgVTzSo7mT4V9GDVXNajuZPhVp9/d5zu0r4J3XPCd2lTZFFWTgtVzap7mT4V8/ItVzWp7mT4Vaczu853aV9bO7zndpRyJoq0zBakf0apv8AuZPhXt2EVXNqnuZPhVoXzu853aV6393nO7Sq2r5JKsnBqrm1T3MnwrBNo/VH+i1F/wBzJ8KtZv7vOd2lfBO6/jO7SiBV6nwiqDQPo1T3MnwrFJhVZrG1LUdzJ/JWobO7zndpWJ8zs+E7tKzIKsfkmt5tUdzJ8K+fkms5rP3Mn8lanf3ZcJ3aeReDM63jO7T0ICsMmEVRH6NU9z | 12:17 |
| flaper87 | J8K036PVZ/otT3Mnwq2O/u853aV9E7vOd2lAVL+rlZzWp7mT4UfVys5rU9zJ8Ktpv7vOd2lG/u853aUBUv6uVnNanuZPhR9XKzmtT3Mnwq2m/u853aUb+7zndpQFctzTBKmPEqZ8lPOxoMl3Oie1o+yeMyRYKxq+iVxyLj2lSm9jkHYoB//9k= | 12:17 |
| flaper87 | ops | 12:17 |
| flaper87 | not sure what happened there | 12:17 |
| prashanthr_ | flaper87: this is being shown in plain text on my client :( | 12:17 |
| flaper87 | yeah, it's plain text | 12:17 |
| flaper87 | not sure what happened | 12:17 |
| flaper87 | I think I copied something by mistake | 12:17 |
| flaper87 | like a very huge url | 12:17 |
| flaper87 | anyway, sorry about that | 12:18 |
| vkmc | lol | 12:18 |
| prashanthr_ | I think you copied the image and not the URL | 12:18 |
| prashanthr_ | :) | 12:18 |
| flaper87 | right, based on the heather | 12:18 |
| flaper87 | header | 12:18 |
| flaper87 | WTF is wrong with me today? | 12:19 |
| flaper87 | vkmc: re that mongodb thing | 12:19 |
| vkmc | is friday flaper87, you are already thinking about beers | 12:19 |
| vkmc | flaper87, yup, I'm trying to find if there is some change that triggered too or what is going on, did you find something? | 12:19 |
| flaper87 | I'll give it a try locally but I'm afraid something might have changed | 12:19 |
| flaper87 | because just that specific test fails | 12:20 |
| vkmc | s/too/that | 12:20 |
| vkmc | yeah | 12:20 |
| flaper87 | and it uses a weird port which means it will fail to connect | 12:20 |
| flaper87 | thing is, it shouldn't even try to connect | 12:20 |
| *** alcabrera|afk is now known as alcabrera | 12:21 | |
| vkmc | yeah the port is wrong | 12:22 |
| flaper87 | but that's fine, I guess | 12:23 |
| flaper87 | I mean, that test shouldn't even try to connect | 12:23 |
| vkmc | I'm running the tests locally | 12:25 |
| * flaper87 too | 12:26 | |
| flaper87 | ok, it failed | 12:26 |
| flaper87 | at least we know it's not the gate | 12:26 |
| vkmc | yeah | 12:26 |
| vkmc | what is on top of git log? | 12:26 |
| vkmc | are you getting this as well? InvalidDriver: The mongodb driver requires mongodb>=2.2, 2.0.4 found | 12:27 |
| flaper87 | nope | 12:28 |
| flaper87 | I've 2.7.2 installed | 12:28 |
| vkmc | how the heck my env work then | 12:28 |
| *** sphoorti has joined #openstack-zaqar | 12:29 | |
| flaper87 | aaaaaaaaaaaah wait a minute, I think I know what happendd | 12:29 |
| flaper87 | happened | 12:29 |
| flaper87 | it's stevedore's update, probably | 12:29 |
| flaper87 | 2 seconds | 12:30 |
| vkmc | k | 12:30 |
| flaper87 | vkmc: https://github.com/openstack/stevedore/commit/3668de2513053f05cec068d12cc7b9cfb642b298 | 12:34 |
| flaper87 | that broke it | 12:34 |
| vkmc | flaper87, so we don't have a pinned version for stevedore in the requirements | 12:36 |
| flaper87 | right, we don't need it, I'm working on a quick fix | 12:36 |
| vkmc | you are saying that they updated pypi's package and we hit that? | 12:36 |
| flaper87 | yup | 12:36 |
| flaper87 | latest stevedore is 1.0.0 | 12:36 |
| vkmc | yeah, in our requirements we specify stevedore>=0.14 | 12:37 |
| vkmc | that commit belongs to 1.0.0 | 12:38 |
| vkmc | we could ask for <1.0.0 in the requirements, for an easy fix, and then check how to upgrade to 1.0.0 | 12:39 |
| vkmc | or what do you have in mind? | 12:39 |
| openstackgerrit | Flavio Percoco proposed a change to openstack/zaqar: Catch Exception instead of RuntimeError for driver https://review.openstack.org/122718 | 12:41 |
| flaper87 | vkmc: ^ | 12:41 |
| flaper87 | no need to pin, lets just update our code | 12:41 |
| vkmc | :D :D | 12:41 |
| vkmc | you rule | 12:41 |
| vkmc | let me try that in my updated env | 12:42 |
| vkmc | and we can ninja approve | 12:42 |
| *** jchai has joined #openstack-zaqar | 12:43 | |
| sphoorti | Hello All :) | 12:43 |
| vkmc | flaper87, this is the OPW applicant I told you about ;) | 12:44 |
| flaper87 | sphoorti: hellooooooooooooooooooooooooooooooooooooooooooooooooooooooooo | 12:44 |
| flaper87 | welcome to the craziest channel ever | 12:44 |
| flaper87 | with the craziest people ever | 12:44 |
| sphoorti | hello flaper87 | 12:44 |
| flaper87 | in this, well, crazy world | 12:44 |
| sphoorti | :D | 12:44 |
| * flaper87 introduces himself | 12:44 | |
| sphoorti | feels the right place to me :P | 12:45 |
| flaper87 | sphoorti: hey, I'm Flavio the robot | 12:45 |
| flaper87 | I've bad memory and most of the things I say make no sense | 12:45 |
| vkmc | that is when he needs a reboot | 12:45 |
| * sphoorti loves robots :p | 12:45 | |
| vkmc | you can trigger that job with #action reboot flaper87 | 12:46 |
| sphoorti | hahaha | 12:46 |
| vkmc | so... as I mentioned, flaper87 has some ideas for this OPW round | 12:47 |
| vkmc | but the first thing we have to do | 12:47 |
| vkmc | is get that first contribution | 12:47 |
| vkmc | you already have experience with that procedure though :) | 12:47 |
| sphoorti | yes ofcourse! | 12:47 |
| vkmc | it's a matter of finding a bug | 12:47 |
| vkmc | awesome | 12:48 |
| vkmc | I'm aware you already have your Zaqar env up | 12:48 |
| vkmc | hopefully with an updated mongo version, not like me | 12:48 |
| sphoorti | yes I have setup the environment | 12:48 |
| vkmc | +2 | 12:48 |
| sphoorti | but i set it up two weeks back | 12:48 |
| sphoorti | should I make any changes now ? | 12:48 |
| vkmc | no, I think it's ok | 12:49 |
| vkmc | simply update your master branch regularly to keep up with the updates | 12:49 |
| sphoorti | okay I shall do that | 12:49 |
| vkmc | kewl | 12:50 |
| openstackgerrit | Flavio Percoco proposed a change to openstack/zaqar: Catch Exception instead of RuntimeError for driver https://review.openstack.org/122718 | 12:50 |
| vkmc | flaper87, no i18n? | 12:51 |
| flaper87 | vkmc: not for non-user facing exceptions | 12:52 |
| flaper87 | I think | 12:52 |
| flaper87 | I'm not sure, really :P | 12:52 |
| flaper87 | I think that's what we've done in other places | 12:52 |
| vkmc | ok :) | 12:52 |
| flaper87 | sphoorti: ^ see? bad memory | 12:52 |
| sphoorti | haha | 12:53 |
| * sphoorti is looking at the review | 12:53 | |
| vkmc | I'm in the i18n team as well, so if I see some inconsistency I'll make the required updates | 12:53 |
| sphoorti | trying to grasp | 12:53 |
| *** alcabrera is now known as alcabrera|afk | 12:53 | |
| flaper87 | vkmc: does that review look good to you? | 12:55 |
| flaper87 | s/review/patch | 12:55 |
| vkmc | sphoorti, I think I got a bug for you https://bugs.launchpad.net/zaqar/+bug/1369536 | 12:56 |
| vkmc | flaper87, I can read between lines, I'm reviewing it :) | 12:56 |
| * flaper87 swears he didn't implied anything else | 12:57 | |
| * flaper87 silently goes away | 12:57 | |
| vkmc | sphoorti, feel free to go through the bug list and get one by yourself though :) | 12:57 |
| sphoorti | I am looking currently at the bug you showed me | 12:57 |
| *** jchai is now known as jchai_afk | 12:58 | |
| vkmc | sphoorti, sure thing | 12:59 |
| vkmc | flaper87, with your patch tests are running fine in my env now, +2 .. +A? | 13:07 |
| vkmc | and as you said, no i18n for non-user facing exceptions | 13:08 |
| *** sriram1 has joined #openstack-zaqar | 13:10 | |
| *** jchai_afk is now known as jchai | 13:11 | |
| flaper87 | vkmc: https://review.openstack.org/#/c/121478/ | 13:17 |
| flaper87 | :D | 13:17 |
| openstackgerrit | Flavio Percoco proposed a change to openstack/zaqar: Enable ZAQAR_TEST_SLOW for py27 https://review.openstack.org/122732 | 13:21 |
| *** mpanetta has joined #openstack-zaqar | 13:23 | |
| *** earnThis has joined #openstack-zaqar | 13:24 | |
| flaper87 | did you guys know about this? http://www.vmdoh.com/blog/openstack-marconi-queue-backend-drupal | 13:24 |
| flaper87 | I had no idea | 13:24 |
| vkmc | no :o | 13:26 |
| vkmc | so cool! | 13:26 |
| mpanetta | Wow! | 13:26 |
| mpanetta | Someone needs to forward that on to the list :P | 13:26 |
| mpanetta | It is almost a year old, wow. | 13:27 |
| *** malini has joined #openstack-zaqar | 13:28 | |
| vkmc | flaper87, another question wrt the client... https://github.com/openstack/python-zaqarclient/blob/master/zaqarclient/queues/client.py#L23 | 13:28 |
| vkmc | conf is a json string, right? | 13:28 |
| vkmc | is there some schema for it or something? | 13:29 |
| vkmc | damn you pep8 | 13:53 |
| *** mpanetta has quit IRC | 13:54 | |
| *** ametts has joined #openstack-zaqar | 13:54 | |
| *** mpanetta has joined #openstack-zaqar | 13:54 | |
| *** alcabrera|afk is now known as alcabrera | 14:01 | |
| *** openstackgerrit has quit IRC | 14:02 | |
| *** flwang1 has quit IRC | 14:10 | |
| *** jeffrey4l has quit IRC | 14:10 | |
| *** jeffrey4l has joined #openstack-zaqar | 14:17 | |
| *** prashanthr_1 has joined #openstack-zaqar | 14:24 | |
| *** prashanthr_ has quit IRC | 14:25 | |
| vkmc | hey guys, can someone review this patch? https://review.openstack.org/#/c/122718/ | 14:32 |
| vkmc | it's blocking the py27 gate | 14:32 |
| *** flwang1 has joined #openstack-zaqar | 14:47 | |
| *** flwang has quit IRC | 14:48 | |
| *** AAzza_afk has quit IRC | 14:48 | |
| notmyname | flaper87: I'm online now | 14:51 |
| *** AAzza_afk has joined #openstack-zaqar | 14:53 | |
| *** AAzza_afk is now known as AAzza | 14:54 | |
| notmyname | flaper87: I believe you wanted to talk about rings | 15:00 |
| mpanetta | is flaper87 getting married? :P | 15:01 |
| notmyname | maybe he's going on and adventure with some friends to throw one in a volcano | 15:04 |
| *** reed has joined #openstack-zaqar | 15:06 | |
| * flaper87 back | 15:10 | |
| flaper87 | sorry, call | 15:10 |
| vkmc | thanks malini, alcabrera | 15:10 |
| flaper87 | mpanetta: married? | 15:11 |
| * flaper87 just died | 15:11 | |
| mpanetta | notmyname: Ahh! | 15:11 |
| flaper87 | notmyname: have 5 imns? | 15:11 |
| flaper87 | notmyname: http://lists.openstack.org/pipermail/openstack-dev/2014-September/046489.html | 15:11 |
| mpanetta | flaper87: Yeah, rings and all. All though I like the adventure idea better | 15:12 |
| flaper87 | If you can read the paragraph were I explain pools | 15:12 |
| notmyname | flaper87: for you? I'll do you one better and give you six minutes ;-) | 15:12 |
| flaper87 | notmyname: AWESOME! | 15:12 |
| alcabrera | rings - distributed support networks to keep relationships/nodes healthy | 15:12 |
| vkmc | bbl o/ | 15:12 |
| flaper87 | we have pools and we choose the pool based on a weight | 15:12 |
| mpanetta | alcabrera: Or something to rule them all | 15:12 |
| flaper87 | I was wondering wether we could use some of the very clever distribution algorithms swift uses to distribute queues between those pools | 15:13 |
| alcabrera | mpanetta: :P | 15:13 |
| mpanetta | hehe | 15:13 |
| notmyname | flaper87: depends on what you're trying to achieve | 15:14 |
| flaper87 | As of now, I really don't want to get to the 'message distribution' thing. I think that will bring way more issues into the project than we want (or are able to) hand;e | 15:14 |
| flaper87 | (but let me know if I'm wrong and I should consider message distribution) | 15:14 |
| *** openstackgerrit has joined #openstack-zaqar | 15:14 | |
| flaper87 | notmyname: I don't think the current wighted distribution algorithm is fair and it puts lots of responsibilities into the operators hands | 15:14 |
| flaper87 | my goal would be to make zaqar fully responsible for distributing queues between the available pools | 15:15 |
| *** sriram1 has quit IRC | 15:15 | |
| flaper87 | (and in the future re-balancing if necessary) | 15:15 |
| flaper87 | I was thinking that we could probably use a consistent hash to do so | 15:16 |
| mpanetta | So it really is one ring to rule them all (all the pools that is) | 15:16 |
| flaper87 | but although I understand how the consistent hash thing works, I've no experience with it | 15:16 |
| notmyname | describe to me more of what the pools provide to zawr | 15:16 |
| notmyname | zaqar | 15:16 |
| flaper87 | pools are basically shards. In other words, they are storage cluster | 15:16 |
| flaper87 | For example, if we had support for swift, a pool would be a swift cluster | 15:17 |
| flaper87 | Pools are isolated from each other | 15:17 |
| * sebasmagri was going thru that thread right now | 15:17 | |
| flaper87 | sebasmagri: let me know your thoughts | 15:18 |
| *** sriram1 has joined #openstack-zaqar | 15:18 | |
| notmyname | flaper87: 1:1 with the storage implementation? eg could I have 2 swift clusters in one pool or swift + etc in one pool? | 15:18 |
| flaper87 | notmyname: 1:1 | 15:18 |
| flaper87 | you can have 2 swift pools | 15:18 |
| flaper87 | but you can't have 2 swift clusters in 1 pool | 15:19 |
| notmyname | also, what problems do pools solve for zaqar or the deployer? why did you do it that way? | 15:19 |
| * notmyname is trying to understand | 15:19 | |
| notmyname | although I'm a huge fan of solving the problem of queues in openstack, I've never looked at the marconi/zaqar code. mea culpa | 15:19 |
| flaper87 | notmyname: couple of things. 1. scaling: it allows deployers to scale out their storage layer when it reaches its limits. | 15:20 |
| * sebasmagri flaper87: your 'distributed storage' idea makes sense for me. I like to think about this as thinking about RAID... | 15:20 | |
| flaper87 | 2. The ability of supporting different storage types in 1 deployment (redis + mongo + ?) | 15:20 |
| sebasmagri | ok my keyboard was going crazy | 15:20 |
| flaper87 | s/of supporting/to support/ | 15:21 |
| flaper87 | it's friday.... | 15:21 |
| flaper87 | sebasmagri: lol | 15:21 |
| flaper87 | However, I think the scaling part of pools is still a bit limited | 15:21 |
| flaper87 | notmyname: ah and 3. data isolation | 15:21 |
| notmyname | flaper87: is a pool exposed to the end user at all? | 15:22 |
| sebasmagri | flaper87: another crazy idea I've had is using something like a Bittorrent/Kademlia DHT for storage :) | 15:22 |
| flaper87 | notmyname: no, operators thing only | 15:22 |
| notmyname | ok | 15:22 |
| notmyname | flaper87: why is the scaling story bad with pools today? | 15:22 |
| flaper87 | notmyname: because it requires the operator to put a weight on their pools. Basically, the current algorithm is asking the operator to tell zaqar how to balance the data | 15:23 |
| flaper87 | not sure whether that's really bad or not | 15:23 |
| notmyname | flaper87: what use cases does "I'm deploying zaqar and I want more than one backend implementation" solve? eg who is wanting to deploy both redis and mongo in a single deployemnt? | 15:23 |
| flaper87 | but I started thinking that we could be more clever on that side | 15:23 |
| *** sphoorti has quit IRC | 15:24 | |
| flaper87 | notmyname: this comes with no research and just assumptions from the ideas we had: I think a small percentage of operators may want to provide different storage capabilities for different user needs (like throughput, durability, etc) | 15:24 |
| flaper87 | while pools are hidden from the end user, flavors are not. This means a user could say: "Hey create this queue in a store with support for high throughput" | 15:25 |
| flaper87 | The user doesn't know whether that goes into redis, mongo or whatever | 15:25 |
| notmyname | sounds like swift's storage policies | 15:25 |
| flaper87 | there you go, great minds think alike | 15:26 |
| * flaper87 puts his ego back in jail | 15:26 | |
| notmyname | lol | 15:26 |
| flaper87 | sebasmagri: that's what I call a "porn idea" | 15:26 |
| sebasmagri | flaper87: I was not thinking about porn at all | 15:26 |
| sebasmagri | xD | 15:26 |
| notmyname | so does a user ask for a queue and then use that or is the user putting messages into zaqar with some "tag" identifying the queue/stream/flow | 15:26 |
| notmyname | eg "PUT /v42/my_tenant/my_flow_id/foo" | 15:28 |
| flaper87 | notmyname: the user may/may not ask for a queue. If the user just puts the message into queue X and queue X doesn't exist, then zaqar will create it in the default pool (queues are lazy in zaqar). However, if the user has more specific requirements like needing a not-persistet, highthroughput queue then the user can create the queue requesting it to be store in that storage | 15:28 |
| flaper87 | notmyname: right | 15:28 |
| notmyname | hmm | 15:28 |
| flaper87 | note that the user doesn't say: "put this in redis" | 15:29 |
| flaper87 | the user says: "put this somewhere that is capable of doing x and y" | 15:29 |
| flaper87 | those flavors need to be created by the operator so the user can choose just between the flavor | 15:29 |
| flaper87 | ... the flavors that exist | 15:30 |
| flaper87 | That said, I'm a bit worried about the current balancing algorithm giving to much responsibility to the operator | 15:30 |
| flaper87 | If the wrong weight is chosen for a pool, the operator might end up with an unbalanced distribution | 15:31 |
| *** kgriffs|afk has quit IRC | 15:31 | |
| notmyname | eg put 60% in this backend ant 40% in that one? | 15:31 |
| flaper87 | right | 15:31 |
| *** kgriffs|afk has joined #openstack-zaqar | 15:32 | |
| *** kgriffs|afk is now known as kgriffs | 15:33 | |
| * notmyname is thinking | 15:34 | |
| flaper87 | notmyname: I guess my first question is: Do you think it is fair to ask the operator to do that? (after all, you also need to configure the ring in swift) | 15:34 |
| flaper87 | another question could be: Do you think the weight algorithm is sane or we should totally work on something different? | 15:35 |
| * flaper87 hires someone to massage notmyname brain to help him think | 15:35 | |
| notmyname | thoughts (unformed and in no particular order): I think the current plan is not quite the right layer of abstraction. I think if you use swift you'll need to get some changes into swift itself (I have ideas). I think multiple storage backends is an ok idea but a lot of extra complexity for zaqar right up front before the whole thing is proven yet | 15:38 |
| notmyname | let me "talk out loud" as a potential deployer who wants to provide zaqar | 15:39 |
| * flaper87 listens | 15:39 | |
| * flaper87 reads* | 15:39 | |
| notmyname | I've got to deploy one backend (eg redis). maybe I already have that running somewhere. I need to scale and manage that for zaqar. I guess I'd set the weight to 100%? | 15:39 |
| notmyname | so now when the end-user comes along, he will thwo messages at the system and get messages out | 15:40 |
| notmyname | so everything seems ok so far | 15:40 |
| notmyname | but if I want to have different messages with different perf, then I need to add a different backend (maybe--let's assume yes here) | 15:41 |
| notmyname | and now the deployer balances across both with some config setting, but the user is also "balancing" by choosing what they want. those 2 things seem at odds | 15:41 |
| flaper87 | ah yeah wait | 15:41 |
| flaper87 | so, the operator would balance between pools of the same type, whereas the end user can choose between types of pools | 15:42 |
| notmyname | so there's a higher-level abstraction than pools? it's (pool, type)? | 15:43 |
| flaper87 | yes, flavor=pool-type | 15:43 |
| * flaper87 is bad at giving names | 15:43 | |
| * flaper87 will use a place holder for his kids so they can choose their own name later on | 15:44 | |
| notmyname | heh | 15:44 |
| flaper87 | lets assume we don't have flavors | 15:45 |
| flaper87 | lets say we just have pools | 15:45 |
| flaper87 | and the user has no idea | 15:45 |
| flaper87 | the user can simply put and get messages out of zaqar | 15:45 |
| notmyname | ok :-) | 15:45 |
| notmyname | right | 15:45 |
| flaper87 | With pools, the operator can add several clusters of storage to zaqar | 15:45 |
| *** malini|afk has quit IRC | 15:46 | |
| *** kgriffs has quit IRC | 15:46 | |
| *** ekarlso has quit IRC | 15:46 | |
| flaper87 | Zaqar will chose what pool to use for queue X when the queue is created | 15:46 |
| notmyname | ok, so why do I (deployer) want to balance across 2 pools that have the same semantics? ie with different weights | 15:46 |
| flaper87 | and it will do that using a "weight-based" algorithm | 15:46 |
| *** malini|afk has joined #openstack-zaqar | 15:46 | |
| *** kgriffs|afk has joined #openstack-zaqar | 15:46 | |
| *** ekarlso has joined #openstack-zaqar | 15:47 | |
| flaper87 | notmyname: scaling. Say you have a mongodb cluster that has 3 nodes. Suddenly that cluster is not able to scale more than that. You can add another mongodb cluster and tell zaqar to start putting queues there | 15:47 |
| *** kgriffs|afk is now known as kgriffs | 15:47 | |
| notmyname | flaper87: by queue or by message? | 15:47 |
| flaper87 | notmyname: queue | 15:47 |
| notmyname | ah ok | 15:47 |
| notmyname | so that's going to make things lumpy | 15:47 |
| notmyname | eg the swift queue is going to be a _lot_ busier than other queues | 15:48 |
| flaper87 | http://www.urbandictionary.com/define.php?term=lumpy <- I assume you didn't mean #1 | 15:48 |
| notmyname | eg the heat queue | 15:48 |
| notmyname | wow | 15:48 |
| flaper87 | :D | 15:48 |
| notmyname | you're almost (but not quite) at SQS semantics, and if you balanced by message, not queue, you'd be all the way there | 15:49 |
| * notmyname likes sqs semantics | 15:49 | |
| flaper87 | right | 15:49 |
| *** sriram1 is now known as srriam | 15:49 | |
| *** srriam is now known as sriram1 | 15:49 | |
| notmyname | ok, here's what I'd expect or want to see | 15:50 |
| flaper87 | I'm not sure whether this is the right time to bring `messages distribution` in for zaqar but I'm open to it | 15:50 |
| flaper87 | As in, I'm a bit scared to do so because it definitely brings more issues in and I wonder if we have enough time to handle them all right now | 15:51 |
| flaper87 | although, with a good plan forward, that could definitely happen | 15:51 |
| notmyname | ...thinking... | 15:51 |
| flaper87 | in other words, I'm relying on the storage to do message distribution | 15:51 |
| flaper87 | mongodb already distributes messages in its cluster | 15:52 |
| flaper87 | s/I'm/Zaqar/ | 15:52 |
| * flaper87 is not a software, just a robot | 15:52 | |
| flaper87 | what Zaqar does is distributing queues across pools | 15:53 |
| flaper87 | then the pool (storage cluster, i.e: mongodb/redis) will take care of distributing the message | 15:53 |
| notmyname | I think that might be harder :-) | 15:53 |
| notmyname | ok, here's what I'd do (as some random guy on the internet who isn't actually contributing code) | 15:54 |
| * flaper87 has to jump into a call in 5 mins but he will be back | 15:55 | |
| notmyname | distribute messages not queues across the available storage that supports the requested SLA | 15:55 |
| notmyname | if a user needs and a deployer provides something with FIFO, then that would be a different SLA with a different pool | 15:56 |
| notmyname | but it's zaqar that provides and guarantees that, not the storage imlementation | 15:56 |
| notmyname | eg as a user I shouldn't know or care or be able to figure out that you are using sql vs swift vs redis for the storage pool | 15:56 |
| flaper87 | right, I don't the user can do that right now (unless the flavor name is "redis") | 15:57 |
| flaper87 | :D | 15:57 |
| notmyname | sure | 15:57 |
| notmyname | a whiteboard would be a lot easier than IRC ;-) | 15:58 |
| flaper87 | notmyname: I was thinking just that | 15:58 |
| flaper87 | really | 15:58 |
| flaper87 | so, lets do this | 15:58 |
| flaper87 | I'll draw this down and uploaded it somewhere | 15:58 |
| flaper87 | (or just use some drawing software) | 15:59 |
| flaper87 | so we can refer to the same diagram | 15:59 |
| notmyname | sure | 15:59 |
| notmyname | a queue is a super high-level thing. and queues will be radically different. I actually don't care what a "queue" is. it's effectively just a tag that I can get a set of messages on | 15:59 |
| notmyname | and I suspect you'll have a hard time balancing queues. which will lead to deployer pain | 16:00 |
| notmyname | eg if swfit tried to balance containers rather than objects, we'd have a really hard time spreading it out or taking advantages of failure domains | 16:00 |
| notmyname | since one container can have 1 billion objects and another can have 1 | 16:01 |
| flaper87 | notmyname: re queues being a tag: http://blog.flaper87.com/post/people-dont-like-to-queue-up/ | 16:01 |
| notmyname | same with zaqar. I'd expect a medium swift cluster to easily put many thousand messages per second into a queue. I suspect the heat queue wouldn't be so busy | 16:01 |
| flaper87 | notmyname: awesome advice, thanks a lot. | 16:01 |
| notmyname | flaper87: I feel like there's a ton more to figure out :-) | 16:02 |
| notmyname | flaper87: have fun on your call :-) | 16:02 |
| notmyname | flaper87: glancing over that blog post, the concept of topics seems sane :-) | 16:04 |
| openstackgerrit | Jeffrey Zhang proposed a change to openstack/zaqar: Fix the msg encode/decode error https://review.openstack.org/122340 | 16:12 |
| *** alcabrera is now known as alcabrera|afk | 16:24 | |
| *** jchai is now known as jchai_afk | 16:57 | |
| * flaper87 back | 17:01 | |
| flaper87 | notmyname: yeah, definitely a lot more to figure out but you already gave a big and useful heads up. Now we have to figure out how to make that happen :P | 17:02 |
| flaper87 | kk, I gotta take my head off of work for a bit | 17:02 |
| *** jchai_afk is now known as jchai | 17:04 | |
| *** alcabrera|afk is now known as alcabrera | 17:24 | |
| *** sphoorti has joined #openstack-zaqar | 17:33 | |
| *** reed has quit IRC | 17:58 | |
| *** openstackgerrit has quit IRC | 18:01 | |
| *** openstackgerrit has joined #openstack-zaqar | 18:02 | |
| sphoorti | flaper87: are you around ? | 18:03 |
| notmyname | sphoorti: he took off about an hour ago | 18:08 |
| sphoorti | thanks a lot notmyname | 18:08 |
| sphoorti | notmyname: I had some querries about a bug | 18:09 |
| sphoorti | could you help me with it ? | 18:09 |
| notmyname | unlikely. I lurk here (and occasionally chat) because I want to use zaqar someday. I've never looked at the code | 18:09 |
| notmyname | (yet) | 18:10 |
| sphoorti | sure | 18:11 |
| sphoorti | not an issue | 18:11 |
| vkmc | hey sphoorti, sup? | 18:23 |
| vkmc | 4 hours in a bank gosh | 18:23 |
| sphoorti | hey vkmc ! | 18:23 |
| sphoorti | no online transactions ? | 18:23 |
| sphoorti | 4 hours is tooo much | 18:23 |
| vkmc | yeah, I didn't have lunch yet | 18:23 |
| sphoorti | awwwww :( | 18:24 |
| sphoorti | go grab some lunch | 18:24 |
| vkmc | yeah, I'm waiting for the water for my noodles to boil :p | 18:24 |
| sphoorti | yum yum :p | 18:24 |
| vkmc | so, how can I help you with that bug? | 18:26 |
| sphoorti | vkmc: I didnt get a complete idea of the bug from the description apart from the fact that unit test needs to be written | 18:32 |
| sphoorti | https://bugs.launchpad.net/zaqar/+bug/1369536 | 18:32 |
| vkmc | well... there is not much to add to that | 18:33 |
| vkmc | we have added some features to the API v1.1 | 18:34 |
| vkmc | (Zaqar has two versions currently, v1.0 and v1.1) | 18:34 |
| vkmc | and those unit tests are missing | 18:34 |
| vkmc | in order to add those, you should go through the repo structure and understerstand how the tests are build | 18:35 |
| sphoorti | in both versions or just v1.1 ? | 18:35 |
| sphoorti | yep I am doing that | 18:35 |
| vkmc | AFAIK, just v1.1 | 18:35 |
| vkmc | you can base your work in other tests | 18:36 |
| vkmc | hmm I remember AAzza wrote a great doc about testing | 18:37 |
| vkmc | but it's not online yet | 18:37 |
| sphoorti | okay | 18:37 |
| vkmc | to run the tests, simply run 'tox -e py27' | 18:38 |
| vkmc | or the environment you want :) | 18:38 |
| vkmc | can you go ahead from here? or is there something else you are blocked with? | 18:39 |
| sphoorti | I will try | 18:39 |
| sphoorti | thank you so much | 18:40 |
| vkmc | kewl :D | 18:45 |
| vkmc | noodles time! (4pm lol) | 18:45 |
| sphoorti | its 1230 am here :D | 18:47 |
| sphoorti | enjoy your meal | 18:47 |
| *** sphoorti has quit IRC | 19:05 | |
| *** jchai is now known as jchai_afk | 19:11 | |
| * vkmc lurks | 19:25 | |
| *** jchai_afk is now known as jchai | 19:26 | |
| * flaper87 lurks | 19:30 | |
| flaper87 | not for long | 19:30 |
| flaper87 | vkmc: how's the rally thing going? | 19:30 |
| flaper87 | ah, I just read you were stuck in the bank | 19:31 |
| flaper87 | sorry to hear that | 19:31 |
| *** malini has quit IRC | 19:31 | |
| *** malini|afk is now known as malini | 19:31 | |
| vkmc | flaper87, hey :) I asked you about the conf in the client | 19:32 |
| flaper87 | That was this morning, right? | 19:33 |
| vkmc | how can I manage that in a conf file? | 19:33 |
| vkmc | it was yeah | 19:33 |
| flaper87 | mmh, why a conf file? | 19:33 |
| flaper87 | why do you need that? | 19:33 |
| * flaper87 confused | 19:33 | |
| flaper87 | mmh, you mean to pass the conf object to the client instance? | 19:34 |
| vkmc | to instantiate the client | 19:34 |
| vkmc | exactly | 19:34 |
| flaper87 | something cool about zaqarclient is that it doesn't depend on oslo.config | 19:34 |
| vkmc | doing it in a conf file seemed the best option... but I can wire it | 19:34 |
| flaper87 | I say it is cool because it doesn't force any config library into the user | 19:34 |
| flaper87 | I don't think you need a config file | 19:35 |
| vkmc | cool | 19:35 |
| flaper87 | you just need to pass a dictionary to zaqarclient | 19:35 |
| flaper87 | there's a conf keyword which expects a dict | 19:35 |
| vkmc | exactly | 19:35 |
| flaper87 | that's it | 19:35 |
| flaper87 | you don't need anything else (AFAIK) | 19:35 |
| flaper87 | :P | 19:35 |
| flaper87 | vkmc: for example: https://github.com/openstack/python-zaqarclient/blob/master/examples/keystone_auth.py#L30 | 19:36 |
| vkmc | what I'm saying is that IIRC the signature for clients in Rally just receive the version | 19:36 |
| vkmc | I'm thinking on a decent way to pass the conf dict | 19:36 |
| flaper87 | let me check rally's clients code | 19:37 |
| flaper87 | 2s | 19:37 |
| flaper87 | ok, here | 19:37 |
| flaper87 | https://github.com/stackforge/rally/blob/master/rally/osclients.py#L166 | 19:37 |
| flaper87 | it gets the image_url | 19:37 |
| flaper87 | you'll get zaqar_url | 19:37 |
| flaper87 | which you can pass in the url keyword | 19:37 |
| flaper87 | Client(url=zaqar_url, conf=zaqar_conf) | 19:38 |
| flaper87 | in zaqar_conf you need to specify the auth token | 19:38 |
| * flaper87 is sure as hell that there's something missing in zaqarclient | 19:38 | |
| flaper87 | so, the dict would look like this | 19:39 |
| vkmc | we are in the same page so far | 19:39 |
| flaper87 | vkmc: http://paste.openstack.org/show/113436/ | 19:39 |
| flaper87 | s/'zaqar'/the_real_token/ | 19:40 |
| flaper87 | and that's it | 19:40 |
| vkmc | ok, so you are suggesting me to wire it to the function | 19:40 |
| flaper87 | you don't need anything else | 19:40 |
| flaper87 | yup, that's what other clients do as well | 19:40 |
| vkmc | that is what I was asking :p | 19:40 |
| flaper87 | https://github.com/stackforge/rally/blob/master/rally/osclients.py#L166-L174 | 19:40 |
| flaper87 | that's wiring to me | 19:40 |
| flaper87 | :P | 19:40 |
| flaper87 | I mean, it's passing the expected arguments | 19:41 |
| vkmc | check that for the other params the CONF file is being used | 19:41 |
| flaper87 | vkmc: right | 19:41 |
| flaper87 | just FYI, you're testing the 'pass the token instead of auth arguments' thing | 19:42 |
| flaper87 | :P | 19:42 |
| flaper87 | so, expect something to be broken | 19:42 |
| vkmc | ok ok | 19:42 |
| vkmc | I was making it more complicated than that | 19:44 |
| vkmc | thanks for clearing out | 19:44 |
| * vkmc fails to KISS | 19:44 | |
| flaper87 | vkmc: to be fair, the client could use some documentation | 19:47 |
| flaper87 | and I keep saying that I'll write it but .... | 19:47 |
| vkmc | we can do doc-rock-day for the client soon | 19:47 |
| vkmc | let's focus on the most important features now though | 19:47 |
| vkmc | I'll help with that when it comes the time | 19:49 |
| vkmc | in the meantime | 19:49 |
| vkmc | I keep poking you (sorry abt that) | 19:49 |
| flaper87 | haahha, nah, don't worry | 19:52 |
| flaper87 | keep doing it | 19:52 |
| flaper87 | if I don't reply is because I'm working on having an actually life | 19:52 |
| flaper87 | but I keep failing at it | 19:52 |
| flaper87 | :P | 19:52 |
| vkmc | it's ok :) | 19:53 |
| vkmc | btw | 19:55 |
| vkmc | (run away or I'll keep asking things) | 19:55 |
| vkmc | client's API_NAME is 'queues' | 19:55 |
| vkmc | sorry | 19:55 |
| vkmc | 'queuing' | 19:55 |
| vkmc | we should rename to messaging | 19:56 |
| flaper87 | yeah, we gotta do that | 19:56 |
| flaper87 | in devstack mainly, that's the only place | 19:56 |
| flaper87 | .. where we are using it | 19:56 |
| vkmc | we are registering to the keystone catalog with that too | 19:56 |
| flaper87 | yeah, I meant to say, that's the only place where we're actually registering to keystone | 19:57 |
| flaper87 | :D | 19:57 |
| vkmc | oh ok | 19:57 |
| vkmc | well I'll change it before submitting it to rally | 19:57 |
| vkmc | I mean, for this call https://github.com/openstack/python-zaqarclient/blob/master/examples/keystone_auth.py#L30 | 19:58 |
| vkmc | darn wrong link | 19:58 |
| vkmc | https://github.com/stackforge/rally/blob/master/rally/osclients.py#L166 | 19:58 |
| vkmc | here | 19:58 |
| vkmc | service_type='queuing' | 19:58 |
| vkmc | flaper87, ^ | 20:00 |
| vkmc | oh and also, this is failing to merge because of the bug its trying to fix https://review.openstack.org/#/c/122718/ | 20:03 |
| *** jchai is now known as jchai_afk | 20:03 | |
| *** jchai_afk is now known as jchai | 20:05 | |
| vkmc | alcabrera, do you know how to overcome this? ^ | 20:14 |
| alcabrera | o.o | 20:15 |
| * alcabrera reads for context | 20:15 | |
| vkmc | a stevedore update is making out py27 gate to break | 20:16 |
| vkmc | that is the fix | 20:16 |
| -openstackstatus- NOTICE: Gerrit will be offline from 20:30 to 20:50 UTC for project renames | 20:16 | |
| *** ChanServ changes topic to "Gerrit will be offline from 20:30 to 20:50 UTC for project renames" | 20:16 | |
| vkmc | at least, it works in our environments | 20:16 |
| alcabrera | I've no idea how to overcome this issue. :/ | 20:23 |
| vkmc | no problem Alej, thanks for checking | 20:24 |
| alcabrera | good luck! :) | 20:24 |
| vkmc | :) | 20:24 |
| openstackgerrit | Victoria MartÃnez de la Cruz proposed a change to openstack/python-zaqarclient: Updates API name https://review.openstack.org/122856 | 20:27 |
| *** flwang has joined #openstack-zaqar | 20:29 | |
| *** alcabrera is now known as alcabrera|afk | 20:32 | |
| *** malini1 has joined #openstack-zaqar | 20:40 | |
| *** malini1 has quit IRC | 20:40 | |
| *** malini1 has joined #openstack-zaqar | 20:41 | |
| * flaper87 back | 20:44 | |
| flaper87 | vkmc: FYI, reverify no bug doesn't exist | 20:45 |
| flaper87 | reverify requires a bug | 20:45 |
| vkmc | oh noes | 20:46 |
| flaper87 | I just rechecked it | 20:46 |
| flaper87 | lets see what happens | 20:46 |
| flaper87 | did you upload a new patch set? | 20:46 |
| vkmc | yeah, pep8 error | 20:46 |
| vkmc | I swear reverify no bug was a thing | 20:46 |
| flaper87 | it was | 20:46 |
| flaper87 | like 2 cycles ago | 20:46 |
| flaper87 | :D | 20:46 |
| vkmc | yeah but | 20:47 |
| vkmc | now... it still triggers the jobs | 20:47 |
| flaper87 | does it? | 20:47 |
| flaper87 | nope, it doesn't | 20:47 |
| vkmc | ok, good to know | 20:48 |
| flaper87 | wait, I'm going to rebase again | 20:49 |
| malini | there was/is a reverify no bug | 20:49 |
| malini | AFAIK reverify is for gate job & recheck is for check jobs | 20:49 |
| vkmc | yeah I had the same | 20:49 |
| flaper87 | malini: yes but reverify requries a bug number | 20:49 |
| vkmc | now flaper87 ruined it | 20:49 |
| flaper87 | :P | 20:49 |
| *** malini1 has left #openstack-zaqar | 20:50 | |
| malini | there was really a reverify no bug | 20:50 |
| malini | flaper87 is in the minority :-P | 20:50 |
| vkmc | hahaha | 20:51 |
| flaper87 | malini: vkmc https://wiki.openstack.org/wiki/GerritJenkinsGit#Test_Failures | 20:51 |
| flaper87 | *cough* *cough* | 20:51 |
| vkmc | he is right http://openstack.10931.n7.nabble.com/Retiring-quot-reverify-no-bug-quot-td26343.html | 20:52 |
| openstackgerrit | Flavio Percoco proposed a change to openstack/zaqar: Catch Exception instead of RuntimeError for driver https://review.openstack.org/122718 | 20:52 |
| *** ChanServ changes topic to "OpenStack Queuing and Notification Service || Smile :D || Meetings every Monday @ 21:00 UTC/15:00 UTC || Wiki: https://wiki.openstack.org/wiki/Zaqar || Paste: http://paste.openstack.org/ || Send messages and make some noise :D" | 20:52 | |
| -openstackstatus- NOTICE: Gerrit is back online | 20:52 | |
| malini | atleast it was around till December 11, 2013..saving grace | 20:53 |
| vkmc | haha yeah | 20:53 |
| vkmc | and now I think it does the same as recheck no bug | 20:53 |
| vkmc | anyhow | 20:53 |
| vkmc | thanks for rebasing flaper87 | 20:54 |
| vkmc | hope it doesn't hit with anything now | 20:54 |
| *** openstackgerrit has quit IRC | 20:55 | |
| *** openstackgerrit has joined #openstack-zaqar | 20:55 | |
| *** reed has joined #openstack-zaqar | 21:07 | |
| *** sriram1 has quit IRC | 21:12 | |
| *** jchai has quit IRC | 21:14 | |
| *** mpanetta has quit IRC | 21:40 | |
| vkmc | so it begins... https://review.openstack.org/#/c/122876/ | 22:01 |
| vkmc | and the update in DevStack https://review.openstack.org/#/c/122878/ | 22:10 |
| *** earnThis has quit IRC | 22:11 | |
| *** ametts has quit IRC | 22:12 | |
| flaper87 | vkmc: thanks :) | 22:39 |
| *** flaper87 is now known as flaper87|afk | 22:42 | |
| vkmc | have a good weekend flaper87|afk o/ | 22:59 |
| *** echevemaster has joined #openstack-zaqar | 23:12 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!