| *** matsuhashi has joined #openstack-ceilometer | 00:12 | |
| *** matsuhashi has quit IRC | 00:31 | |
| *** matsuhashi has joined #openstack-ceilometer | 00:52 | |
| openstackgerrit | Ana Malagon proposed a change to openstack/ceilometer: Added documentation for selectable aggregates https://review.openstack.org/80299 | 01:08 |
|---|---|---|
| *** anamalagon has joined #openstack-ceilometer | 01:11 | |
| *** nosnos has joined #openstack-ceilometer | 01:30 | |
| *** shakamunyi has joined #openstack-ceilometer | 01:37 | |
| *** caynan has quit IRC | 01:44 | |
| openstackgerrit | Jia Dong proposed a change to openstack/ceilometer: Implement meter query by 'counter_volume' field https://review.openstack.org/67384 | 02:09 |
| *** xianghui has joined #openstack-ceilometer | 02:22 | |
| openstackgerrit | gordon chung proposed a change to openstack/pycadf: add sphinx base files https://review.openstack.org/80880 | 02:35 |
| *** matsuhashi has quit IRC | 02:48 | |
| *** flwang has quit IRC | 02:50 | |
| *** matsuhas_ has joined #openstack-ceilometer | 02:52 | |
| openstackgerrit | gordon chung proposed a change to openstack/pycadf: pycadf documentation https://review.openstack.org/65139 | 02:54 |
| openstackgerrit | gordon chung proposed a change to openstack/pycadf: pycadf documentation https://review.openstack.org/65139 | 03:11 |
| *** matsuhas_ has quit IRC | 03:22 | |
| openstackgerrit | gordon chung proposed a change to openstack/pycadf: pycadf documentation https://review.openstack.org/65139 | 03:32 |
| *** flwang has joined #openstack-ceilometer | 03:36 | |
| openstackgerrit | gordon chung proposed a change to openstack/pycadf: pycadf documentation https://review.openstack.org/65139 | 03:42 |
| openstackgerrit | gordon chung proposed a change to openstack/pycadf: pycadf documentation https://review.openstack.org/65139 | 03:57 |
| openstackgerrit | gordon chung proposed a change to openstack/pycadf: pycadf documentation https://review.openstack.org/65139 | 04:15 |
| *** terriyu has joined #openstack-ceilometer | 04:24 | |
| openstackgerrit | gordon chung proposed a change to openstack/pycadf: pycadf documentation https://review.openstack.org/65139 | 04:28 |
| openstackgerrit | A change was merged to openstack/pycadf: pycadf documentation https://review.openstack.org/65139 | 04:34 |
| *** matsuhashi has joined #openstack-ceilometer | 04:43 | |
| openstackgerrit | Piyush Masrani proposed a change to openstack/ceilometer: Documenting hypervisor support for nova meters https://review.openstack.org/79910 | 05:01 |
| *** Akshik has joined #openstack-ceilometer | 05:18 | |
| *** Akshik has quit IRC | 05:26 | |
| *** Akshik has joined #openstack-ceilometer | 05:38 | |
| anamalagon | hey ildikov_ thank you so so much for the help so far - one last thing, I think I did something wrong when I changed the doc string in api/controllers/v2.py (if I look at the ceilometer-docs, the return type under CapabilitiesController looks funny) | 05:41 |
| ildikov_ | anamalagon: hi | 05:42 |
| ildikov_ | anamalagon: there should be a return type part, I will check why it is different from the other endpoints' doc, just one sec | 05:42 |
| *** _nadya_ has joined #openstack-ceilometer | 05:48 | |
| ildikov_ | anamalagon: hm, for the first look, I do not know what can cause the problem, as the only difference between your text at the other doc strings like this that in yours there is no :param: section, but it should not result in this doc format | 05:48 |
| ildikov_ | anamalagon: I will investigate this a bit and then I get back to you with the results, maybe my morning coffee is the only one, what's missing :) | 05:49 |
| anamalagon | ildikov_: morning coffee is a crucial ingredient :)) | 05:50 |
| ildikov_ | anamalagon: yes, that's absolutely right, my brain just cannot start without it :))) | 05:51 |
| anamalagon | ildikov_: thanks so much for all your help - I really appreciate it :) I'll keep looking through the code on my end, too | 05:53 |
| ildikov_ | anamalagon: I need about 20 mins to get coffee and walk into the office and then I will check this issue and I will share the results with you | 05:54 |
| *** nosnos has quit IRC | 05:54 | |
| *** nosnos has joined #openstack-ceilometer | 05:54 | |
| ildikov_ | anamalagon: cool, then it can be you, who shares the solution with me this time :) | 05:55 |
| *** matsuhashi has quit IRC | 05:56 | |
| *** matsuhashi has joined #openstack-ceilometer | 05:57 | |
| *** matsuhashi has quit IRC | 05:57 | |
| *** matsuhashi has joined #openstack-ceilometer | 06:00 | |
| openstackgerrit | Jenkins proposed a change to openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/76762 | 06:03 |
| *** _nadya_ has quit IRC | 06:05 | |
| *** _nadya_ has joined #openstack-ceilometer | 06:06 | |
| *** matsuhashi has quit IRC | 06:09 | |
| *** ildikov_ has quit IRC | 06:09 | |
| *** matsuhashi has joined #openstack-ceilometer | 06:10 | |
| *** matsuhas_ has joined #openstack-ceilometer | 06:12 | |
| *** matsuhashi has quit IRC | 06:14 | |
| *** _nadya_ has quit IRC | 06:17 | |
| *** Akshik has quit IRC | 06:19 | |
| *** nosnos has quit IRC | 06:24 | |
| *** nosnos has joined #openstack-ceilometer | 06:25 | |
| *** matsuhas_ has quit IRC | 06:25 | |
| *** matsuhashi has joined #openstack-ceilometer | 06:25 | |
| *** Akshik has joined #openstack-ceilometer | 06:37 | |
| *** matsuhashi has quit IRC | 06:39 | |
| *** matsuhashi has joined #openstack-ceilometer | 06:40 | |
| *** ildikov_ has joined #openstack-ceilometer | 06:40 | |
| *** saju_m has joined #openstack-ceilometer | 06:47 | |
| openstackgerrit | liusheng proposed a change to openstack/ceilometer: Fix the return of statistic with getting no sample https://review.openstack.org/80467 | 06:54 |
| openstackgerrit | liusheng proposed a change to openstack/ceilometer: Fix the return of statistic with getting no sample https://review.openstack.org/80467 | 07:01 |
| *** _nadya_ has joined #openstack-ceilometer | 07:05 | |
| *** rongze has joined #openstack-ceilometer | 07:21 | |
| *** _nadya_ has quit IRC | 07:22 | |
| *** rongze has quit IRC | 07:23 | |
| ildikov_ | anamalagon: I have some news for you | 07:35 |
| anamalagon | ildikov_: that | 07:35 |
| anamalagon | that's great! | 07:35 |
| ildikov_ | anamalagon: the good news is that you did not do anything wrong | 07:36 |
| *** _nadya_ has joined #openstack-ceilometer | 07:36 | |
| anamalagon | ildikov_: yay! :D | 07:36 |
| ildikov_ | anamalagon: the bad news is that this behavior seems to be a bug in sphinxcontrib-pecanwsme lib | 07:36 |
| ildikov_ | anamalagon: I would suggest you to leave it as is now and I will try to include this fix into sphinxcontrib | 07:37 |
| openstackgerrit | Zhongyue Luo proposed a change to openstack/ceilometer: Removes use of timeutils.set_time_override https://review.openstack.org/78833 | 07:37 |
| *** _nadya_ has quit IRC | 07:37 | |
| ildikov_ | anamalagon: would this solution be ok with you? | 07:37 |
| anamalagon | ildikov_: works for me! if jenkins works, though, it seems to be failing to build for some reason | 07:39 |
| ildikov_ | anamalagon: Eoghan added a comment to your doc, I think if you fix that, it will be ready to fly | 07:39 |
| ildikov_ | anamalagon: that issue is because of the tempest part, it is not because of your patch | 07:40 |
| anamalagon | ildikov_: ok, I'll fix the paragraph I missed somehow and then resubmit | 07:41 |
| ildikov_ | anamalagon: cool, hopefully Jenkins will be nice that time :) | 07:42 |
| anamalagon | ildikov_: yeah :) | 07:42 |
| ildikov_ | anamalagon: if not, you have the chance to start a recheck, you have to push the Review button and add a text as recheck bug bug_number, I think there is already an example on your patch added by me :) | 07:43 |
| ildikov_ | anamalagon: you have to recheck it without any score so you have to choose 0 | 07:43 |
| ildikov_ | anamalagon: hopefully it will not be needed, but it's always good to know that you have options :) | 07:44 |
| anamalagon | ildikov_: heh, yes - thanks, I'll try that if the build fails | 07:44 |
| openstackgerrit | Yang Yu proposed a change to openstack/ceilometer: Make sure use IPv6 sockets for ceilometer in IPv6 environment https://review.openstack.org/67786 | 07:45 |
| ildikov_ | anamalagon: many times you can see an Elastic recheck comment on your patch after a failed tempest check, it usually gives hint that which bug number you should include | 07:45 |
| openstackgerrit | Yang Yu proposed a change to openstack/ceilometer: Make sure use IPv6 sockets for ceilometer in IPv6 environment https://review.openstack.org/67786 | 07:48 |
| openstackgerrit | Ana Malagon proposed a change to openstack/ceilometer: Added documentation for selectable aggregates https://review.openstack.org/80299 | 07:52 |
| *** Akshik has quit IRC | 07:56 | |
| *** nijaba has joined #openstack-ceilometer | 08:02 | |
| *** nijaba has quit IRC | 08:02 | |
| *** nijaba has joined #openstack-ceilometer | 08:02 | |
| *** Akshik has joined #openstack-ceilometer | 08:04 | |
| *** nosnos has quit IRC | 08:13 | |
| *** nosnos has joined #openstack-ceilometer | 08:14 | |
| *** matsuhashi has quit IRC | 08:17 | |
| *** matsuhashi has joined #openstack-ceilometer | 08:17 | |
| *** Alexei_987 has quit IRC | 08:21 | |
| *** openstack has quit IRC | 08:21 | |
| *** openstack has joined #openstack-ceilometer | 08:29 | |
| openstackgerrit | Ildiko Vancsa proposed a change to openstack/ceilometer: Fix bug in get_capabilities behavior in DB drivers https://review.openstack.org/80852 | 08:30 |
| *** openstackstatus has joined #openstack-ceilometer | 08:30 | |
| *** nosnos has quit IRC | 08:34 | |
| *** flwang has quit IRC | 08:40 | |
| *** eglynn has joined #openstack-ceilometer | 08:44 | |
| *** flwang has joined #openstack-ceilometer | 08:44 | |
| *** flwang has quit IRC | 08:48 | |
| *** nacim has joined #openstack-ceilometer | 08:49 | |
| *** eglynn has quit IRC | 08:53 | |
| *** matsuhas_ has quit IRC | 08:56 | |
| *** eglynn has joined #openstack-ceilometer | 08:59 | |
| *** Akshik has joined #openstack-ceilometer | 09:06 | |
| *** matsuhashi has joined #openstack-ceilometer | 09:06 | |
| anamalagon | eglynn: hi :) | 09:10 |
| eglynn | anamalagon: happy St. Patchrick's Day! :) | 09:11 |
| anamalagon | eglynn: just wanted to say thank you very much for all the help and your patience on the patch - it looks like the build was successful :D | 09:11 |
| anamalagon | eglynn: happy St. Patrick's day to you!! | 09:11 |
| eglynn | anamalagon: np! we should be able to get that puppy landed this morning, thanks for your tenacity! | 09:11 |
| eglynn | anamalagon: ... BTW if you need any help with your OPW application, just shout | 09:12 |
| anamalagon | eglynn: :) I made many mistakes but thanks to you and ildikov_ it came through alright | 09:12 |
| *** yassine has joined #openstack-ceilometer | 09:13 | |
| eglynn | anamalagon: ... don't think of them as "mistakes" ... I prefer "learning opportunities" ;) | 09:13 |
| anamalagon | eglynn: ah, right yess...I had many learning opportunities | 09:13 |
| eglynn | anamalagon: FYI I'm travelling today and will be off-site at internal meetings all week, so bouncing on- and off-grid a bit | 09:13 |
| eglynn | anamalagon: ... so email probably best way of getting in touch in needed (eglynn@redhat.com) | 09:14 |
| *** nosnos has joined #openstack-ceilometer | 09:14 | |
| anamalagon | eglynn: ok, sounds good. | 09:14 |
| *** eglynn has quit IRC | 09:18 | |
| ildikov_ | anamalagon: np, so you're patch is on the gate now, hopefully it will get through without any problems :) | 09:53 |
| *** nosnos_ has joined #openstack-ceilometer | 09:55 | |
| *** nosnos has quit IRC | 09:58 | |
| *** Akshik_ has joined #openstack-ceilometer | 09:59 | |
| *** Akshik has quit IRC | 10:03 | |
| *** nosnos_ has quit IRC | 10:10 | |
| *** nosnos has joined #openstack-ceilometer | 10:12 | |
| *** saju_m has quit IRC | 10:17 | |
| *** anamalagon has quit IRC | 10:26 | |
| *** saju_m has joined #openstack-ceilometer | 10:33 | |
| *** matsuhashi has quit IRC | 10:38 | |
| *** nosnos has quit IRC | 10:46 | |
| *** nosnos has joined #openstack-ceilometer | 10:47 | |
| *** matsuhashi has joined #openstack-ceilometer | 10:54 | |
| *** matsuhashi has quit IRC | 11:05 | |
| *** openstackgerrit has quit IRC | 11:10 | |
| *** openstackgerrit has joined #openstack-ceilometer | 11:10 | |
| *** matsuhashi has joined #openstack-ceilometer | 11:13 | |
| *** xianghui has quit IRC | 11:13 | |
| *** saju_m has quit IRC | 11:15 | |
| openstackgerrit | Ildiko Vancsa proposed a change to openstack/ceilometer: Fix bug in get_capabilities behavior in DB drivers https://review.openstack.org/80852 | 11:23 |
| *** matsuhashi has quit IRC | 11:31 | |
| *** chuck_ has joined #openstack-ceilometer | 11:31 | |
| *** chuck_ is now known as zul | 11:34 | |
| *** zul has joined #openstack-ceilometer | 11:34 | |
| *** saju_m has joined #openstack-ceilometer | 11:37 | |
| *** Akshik_ has quit IRC | 11:38 | |
| *** Akshik has joined #openstack-ceilometer | 11:38 | |
| *** sdake has quit IRC | 11:49 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Rationalize get_resources for mongodb https://review.openstack.org/80105 | 12:05 |
| openstackgerrit | A change was merged to openstack/ceilometer: Added documentation for selectable aggregates https://review.openstack.org/80299 | 12:06 |
| *** promulo_ has quit IRC | 12:14 | |
| *** yfujioka has quit IRC | 12:26 | |
| *** matsuhashi has joined #openstack-ceilometer | 12:27 | |
| *** matsuhashi has quit IRC | 12:37 | |
| *** nosnos_ has joined #openstack-ceilometer | 12:37 | |
| *** flwang has joined #openstack-ceilometer | 12:37 | |
| *** nosnos has quit IRC | 12:37 | |
| *** matsuhashi has joined #openstack-ceilometer | 12:37 | |
| *** yassine has quit IRC | 12:42 | |
| *** matsuhashi has quit IRC | 12:42 | |
| *** yassine has joined #openstack-ceilometer | 12:42 | |
| *** Akshik has quit IRC | 12:47 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: DBDeadlock exception in sql backend https://review.openstack.org/80461 | 12:51 |
| *** promulo has joined #openstack-ceilometer | 12:52 | |
| *** Akhil has joined #openstack-ceilometer | 13:00 | |
| *** gordc has joined #openstack-ceilometer | 13:02 | |
| *** matsuhashi has joined #openstack-ceilometer | 13:03 | |
| *** parkerr_ has quit IRC | 13:04 | |
| *** matsuhashi has quit IRC | 13:05 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: 028 migration script incorrectly skips over section https://review.openstack.org/80455 | 13:05 |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: improve performance of resource-list in sql https://review.openstack.org/80343 | 13:05 |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: DBDeadlock exception in sql backend https://review.openstack.org/80461 | 13:05 |
| *** matsuhashi has joined #openstack-ceilometer | 13:05 | |
| *** matsuhashi has quit IRC | 13:10 | |
| *** nosnos_ has quit IRC | 13:12 | |
| *** ryanpetrello has joined #openstack-ceilometer | 13:14 | |
| *** julim has joined #openstack-ceilometer | 13:22 | |
| *** sayalilunkad has joined #openstack-ceilometer | 13:31 | |
| ildikov_ | gordc: hi. are you around? | 13:39 |
| gordc | ildikov_: yep. | 13:42 |
| ildikov_ | gordc: I have a newbie core question :) | 13:43 |
| gordc | ildikov_: sure. | 13:43 |
| ildikov_ | gordc: I've just reached the VMware bugfixes review and I like the patch, the only "issue" is that it does not have a bug report and it refers to the vmware-vcenter-server BP | 13:44 |
| ildikov_ | gordc: is it a blocking issue in the approval or is it ok for I-rc1 as is? | 13:45 |
| gordc | ildikov_: hmm.. i feel like i'm going to give you the wrong answer here.lol | 13:46 |
| ildikov_ | gordc: should I ask maybe jd__ about this? :) | 13:48 |
| gordc | ildikov_: i don't know if we formally need a specific bug number before letting it in... eglynn, jd__, are the release gods. they'll have an answer. | 13:48 |
| gordc | ildikov_: yeah, best to ask them. i'm not a good role model to follow. | 13:48 |
| ildikov_ | gordc: k, will ask them then :) | 13:49 |
| gordc | ildikov_: or you can plus two and let some one else approve it. | 13:49 |
| gordc | or plus 1...i use plus 1 when i'm not sure of things usually. | 13:49 |
| ildikov_ | gordc: hm, good point :) | 13:49 |
| gordc | ildikov_: regarding your comments here: https://review.openstack.org/#/c/80662 | 13:50 |
| ildikov_ | gordc: anyway, I will try to ask and then I will not have to wait for the "judgement" of the release gods :) | 13:51 |
| ildikov_ | gordc: yep, listening | 13:51 |
| gordc | do you think we should be adding the staticmethod if the function def is def inspect_memory_usage(self, instance): seems kinda of strange to me but i'm not sure what's right. | 13:51 |
| jd__ | that doesn't really matter | 13:52 |
| jd__ | if it's cleary a bug fix we're smart enough to know it even if there's no bug number | 13:53 |
| jd__ | let's not be buttoned-down | 13:53 |
| ildikov_ | jd__: ok, cool, thanks I was not sure about the official process | 13:54 |
| ildikov_ | jd__: another question quickly | 13:54 |
| ildikov_ | jd__: I added a comment to gordc's patch: https://review.openstack.org/#/c/80662 | 13:55 |
| ildikov_ | jd__: about the staticmethod comment of yours from here: https://review.openstack.org/#/c/74702/14/ceilometer/compute/virt/inspector.py | 13:55 |
| jd__ | what/where's your comment | 13:56 |
| jd__ | ? | 13:56 |
| ildikov_ | gordc: so I just included jd__ into this question, as I can argue for and against it too | 13:56 |
| *** sayalilunkad has quit IRC | 13:56 | |
| ildikov_ | jd__: my comment was to fix that @staticmethod comment too in that patch | 13:56 |
| gordc | cool | 13:56 |
| * jd__ is having a hard time to follow | 13:57 | |
| gordc | jd__: should i add @staticmethod to the functions in 'abstract' inspector class here: https://review.openstack.org/#/c/74702/14/ceilometer/compute/virt/inspector.py | 13:58 |
| ildikov_ | gordc, jd__: so the question is that should we add the @staticmethod decorator to the methods in the Inspector class in ceilometer/compute/virt/inspector.py | 13:58 |
| ildikov_ | gordc, jd__: for me it seems like the situation of the storage functions in the Connection class in base.py | 13:58 |
| ildikov_ | gordc, jd__: it is another question that how "nice" is to have @staticmethods in the base class, which are then used as member functions | 13:59 |
| jd__ | did everyone read http://julien.danjou.info/blog/2013/guide-python-static-class-abstract-methods already? :) | 14:00 |
| jd__ | the answer is yes, to be perfect gordc you should add @staticmethod | 14:00 |
| gordc | jd__: no plugging your personal website. :) | 14:00 |
| jd__ | you can't have both static and abstract in Python 2 | 14:00 |
| jd__ | but yes in theory the base class should be both static and abstract as the implementation is static | 14:00 |
| jd__ | that's just a Python 2 shortcoming | 14:01 |
| ildikov_ | jd__: guess not, I will give it a try ;) | 14:01 |
| gordc | jd__: you didn't add staticmethod to your first code example under abstract methods section. i guess that's the scenario we're running into now. | 14:03 |
| *** rwsu has joined #openstack-ceilometer | 14:07 | |
| ildikov_ | jd__: so as we do not have a constraint in python that the parameter list should be the same for the abstract method and the method in the child class so we just use the fact that the abstract method does not have an implementation, therefore there is no usage of the self param so we can mark it as static, am I more or less correct? | 14:08 |
| *** nsaje has quit IRC | 14:11 | |
| *** zul has quit IRC | 14:11 | |
| *** nsaje has joined #openstack-ceilometer | 14:12 | |
| *** zul has joined #openstack-ceilometer | 14:13 | |
| nprivalova | jd__, hi! I have a dream... | 14:18 |
| ildikov_ | jd__: ok, after reading the guide, you've sent, I think I can see the big picture now, thanks :) | 14:18 |
| nprivalova | jd__: to run tempest on mongo because all other backends do not allow us to do it :( | 14:18 |
| nprivalova | jd__: have you tried to push new-mongo-support to infra? | 14:19 |
| openstackgerrit | Alexander Bochkarev proposed a change to openstack/ceilometer: Samples from net-poolster use different resource-format https://review.openstack.org/80979 | 14:19 |
| jd__ | nprivalova: lol I definitely liked the first part of your messages, but I think we should do something so you get better dreams :)) | 14:23 |
| * jd__ images nprivalova dreaming about tempest and Mongo at night | 14:23 | |
| jd__ | s/images/imagines/ | 14:24 |
| jd__ | nprivalova: the thing is that we need MongoDB 2.4 on Ubuntu, which is in cloud-archive, but we can't use cloud archive on CI nodes because it pulls a new version of libvirt which is bugged with Nova AFAIK | 14:24 |
| jd__ | I can find the bug number if you want | 14:24 |
| nprivalova | jd__: It's almost the truth:) because we cannot put any test in tempest now | 14:24 |
| openstackgerrit | Alexander Bochkarev proposed a change to openstack/ceilometer: Samples from net-pollster use different resource-format https://review.openstack.org/80979 | 14:25 |
| *** tomixxx5 has joined #openstack-ceilometer | 14:25 | |
| tomixxx5 | do i have to install the telemetry agent on each compute node? http://docs.openstack.org/havana/install-guide/install/apt/content/ceilometer-install-nova.html | 14:26 |
| nprivalova | jd__: oh, ok. I'm preparing a new change-request. it's rather ugly... to fetch deb-packet without adding repos | 14:26 |
| tomixxx5 | or can i exclusevily install the agent on the controller node? | 14:26 |
| tomixxx5 | hi btw :-) | 14:26 |
| jd__ | nprivalova: is that going to be a change on openstack-infra-config? | 14:26 |
| nprivalova | jd__: yep | 14:26 |
| jd__ | nprivalova: I'm prett sure they're going to kick you out, don't you think? :( | 14:26 |
| jd__ | though I recognize it's better to try than to do nothing :) | 14:27 |
| nprivalova | jd__: maybe... but I'm ready to fight :) | 14:27 |
| nprivalova | jd__: will upload soon, let you know | 14:28 |
| *** malini has joined #openstack-ceilometer | 14:29 | |
| jd__ | hehe | 14:31 |
| *** julim_ has joined #openstack-ceilometer | 14:33 | |
| *** julim has quit IRC | 14:36 | |
| luis_ | this patch has been out there for a while in case any of you guys have a chance: https://review.openstack.org/#/c/70824/ | 14:49 |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: DBDeadlock exception in sql backend https://review.openstack.org/80461 | 14:53 |
| *** saju_m has quit IRC | 15:03 | |
| *** tburnes has joined #openstack-ceilometer | 15:07 | |
| Alexei_9871 | Hello all! I'm now working on a concept of refactoring ceilometer models to achieve better storage performance but I fail to understand some concepts behinds some of our models and I cannot find proper documentation for them. I try to refer to http://docs.openstack.org/developer/ceilometer/measurements.html but I cannot find how user/project and resources are related to meters | 15:08 |
| gordc | luis_: you need to rebase | 15:09 |
| gordc | Alexei_9871: i'm pretty sure they're not. | 15:09 |
| Alexei_9871 | gordc: I such case I simply fail to understand how it's supposed to work | 15:10 |
| gordc | Alexei_9871: they seem related to v1 api which use to have user/project specific apis | 15:10 |
| gordc | Alexei_9871: i'm applying my own clean up to sql backend: https://review.openstack.org/#/c/80461/ | 15:10 |
| Alexei_9871 | gordc: yeah :) that's one of the reasons I've started my own search | 15:10 |
| gordc | Alexei_9871: i don't think we need project/user/source models at all... nothing seems to be using those relationships... just creating overhead | 15:11 |
| Alexei_9871 | gordc: cool let's drop them :) | 15:11 |
| Alexei_9871 | gordc: so If I understand correctly ceilometer is basically 1 big table? | 15:13 |
| gordc | Alexei_9871: that's my hope unless there's someone using them. those models seem to be tied to old v1 api... | 15:13 |
| gordc | Alexei_9871: that's how i envision it... or one table for meter and one for events. | 15:13 |
| Alexei_9871 | gordc: Ok it seems that I need to start by revisiting our API spec :) | 15:14 |
| gordc | Alexei_9871: i don't know if it applies to all backends but i think each backend can customise from the 'single table' concept to improve performance however it can | 15:14 |
| *** yassine has quit IRC | 15:14 | |
| Alexei_9871 | gordc: generally I'm against supporting multiple backends :) | 15:15 |
| malini | hello !! I am looking for some info on how ceilometer is set up to run in devstack - especially w.r.t if/how it uses mongo db. Can somebody here help? | 15:15 |
| Alexei_9871 | gordc: ceilometer is the most loaded and write-heavy openstack cervice | 15:15 |
| Alexei_9871 | service* | 15:15 |
| gordc | Alexei_9871: yeah, take a look. be good to hear your input... from my view the backends don't match v2 api purpose | 15:15 |
| Alexei_9871 | gordc: and for high load apps you customize backends to your needs not needs to backends | 15:15 |
| gordc | Alexei_9871: you'll find resistence on not supporting multiple backends... everyone has a preference there. | 15:16 |
| Alexei_9871 | gordc: My IMHO that it's a matter of performance | 15:16 |
| Alexei_9871 | gordc: backends cannot provide similar performance | 15:16 |
| gordc | malini: by default it uses sql backend you need to add an opt to localrc to enable mongo | 15:16 |
| *** Akhil has quit IRC | 15:16 | |
| Alexei_9871 | gordc: and the best should be choosen | 15:16 |
| gordc | Alexei_9871: you're going to have a massive debate on what is the 'best'. :) | 15:17 |
| malini | gordc: is there any job running in the gate that uses mongodb as the ceilometer backend? | 15:17 |
| Alexei_9871 | malini: unittests run against a real mongodb | 15:17 |
| Alexei_9871 | gordc: well right now for example SQL backend uses JOINS massively and this us surely not a big-data solution | 15:18 |
| Alexei_9871 | gordc: and if we drop join support we end up using mongo :) | 15:19 |
| malini | gordc, Alexei_9871: do you know if the tests run with journal disabled? (To give some background, I am working on adding gate jobs for marconi using mongo as the backend. But the db is really slow to startup, making us think journaling could be an issue) | 15:20 |
| *** saju_m has joined #openstack-ceilometer | 15:20 | |
| Alexei_9871 | malini: mongod --maxConns 32 --nojournal --noprealloc --smallfiles --quiet --noauth --port ${MONGO_PORT} --dbpath "${MONGO_DATA}" --bind_ip localhost --config /dev/null &>${MONGO_DATA}/out & | 15:21 |
| Alexei_9871 | that's the option it uses | 15:21 |
| Alexei_9871 | malini: and it uses /tmp as data dir | 15:21 |
| Alexei_9871 | malini: you can speed up tests greatly if you move /tmp on tmpfs | 15:21 |
| malini | Alexei_9871: aah cool & you also have nojournal | 15:21 |
| Alexei_9871 | basically you can check setup-test-env.sh in ceilometer | 15:22 |
| Alexei_9871 | malini: it's kind a provision script for ceilometer tests ^ | 15:23 |
| malini | thanks Alexei_9871!! tht really helps! I will chk that out | 15:23 |
| *** Alexei_9871 is now known as Alexei_987 | 15:24 | |
| gordc | sorry got pulled into a call. | 15:27 |
| gordc | malini: i don't believe gate runs against mongo. just postgres and mysql | 15:27 |
| malini | gordc: That is good info..I also wanted to know if we'll impact ceilometer by turning journal off..Looks like we won't :) | 15:28 |
| gordc | malini: something to do with cloud-archive not having right mongo version... if you set CEILOMETER_BACKEND opt in localrc of your devstack to mongo that should work for you. | 15:29 |
| gordc | malini: not familiar with journaling so i couldn't tell you. | 15:29 |
| *** yassine has joined #openstack-ceilometer | 15:29 | |
| malini | gordc:hmm..did you mean you had trouble getting ceilometer run with mongo in gate? | 15:29 |
| malini | tht is the issue we have with marconi | 15:30 |
| gordc | Alexei_987: yeah we have a lot of joins. if we get rid of all these side tables i think it'll be better... that said i'd think cassandra or hbase would be a better backend solution for us but again, different preferences/use cases. | 15:30 |
| *** shakamunyi has quit IRC | 15:31 | |
| gordc | malini: uh... i think cloud-archive only pulls in mongo 2.2 ? and we need 2.4+? i don't remember the exact details. | 15:31 |
| malini | gordc: from the gate logs, it was at mongo 2.0.4 | 15:31 |
| malini | gordc: & what is cloud-archive ? :) | 15:32 |
| gordc | malini: oh... even worse.lol | 15:32 |
| gordc | malini: some magically term i hear thrown around. :) | 15:32 |
| malini | :D | 15:33 |
| gordc | malini: i assume it's the source repo the gate machines pull from... | 15:33 |
| malini | ok..But I have a few ideas on what to try next.Thanks for your help! | 15:33 |
| gordc | malini: np. good luck | 15:34 |
| malini | thanks gordc! | 15:34 |
| openstackgerrit | Luis A. Garcia proposed a change to openstack/ceilometer: Install global lazy _() https://review.openstack.org/70824 | 15:35 |
| luis_ | gordc, done thanks | 15:36 |
| *** annegentle has joined #openstack-ceilometer | 15:46 | |
| *** Ruetobas has quit IRC | 16:01 | |
| *** yassine has quit IRC | 16:03 | |
| nprivalova | jd__, gordc, Alexei_987, ildikov_, https://review.openstack.org/#/c/81001/4 JFYI | 16:04 |
| *** Ruetobas has joined #openstack-ceilometer | 16:06 | |
| *** Ruetobas has quit IRC | 16:11 | |
| *** yassine has joined #openstack-ceilometer | 16:15 | |
| *** saju_m has quit IRC | 16:15 | |
| *** Ruetobas has joined #openstack-ceilometer | 16:16 | |
| *** _cjones_ has joined #openstack-ceilometer | 16:23 | |
| *** stzilli__ has joined #openstack-ceilometer | 16:25 | |
| *** stzilli_ has quit IRC | 16:28 | |
| *** ildikov_ has quit IRC | 16:33 | |
| *** tomixxx5 has quit IRC | 16:44 | |
| gordc | luis_: i had a comment here: https://review.openstack.org/#/c/70824/10/ceilometer/service.py | 16:45 |
| gordc | luis_: wanted to make sure it's not relevant. | 16:45 |
| *** yassine has quit IRC | 16:45 | |
| *** admin0 has joined #openstack-ceilometer | 16:51 | |
| luis_ | gordc, thank you for looking into this, responded inlie | 16:54 |
| gordc | luis_: thanks for response. just so i'm clear. if we removed all the 'from ceilometer.openstack.common.gettextutils import _' code, we wouldn't need the enable_laxy() call? | 16:57 |
| *** yassine has joined #openstack-ceilometer | 16:58 | |
| luis_ | gordc, yeah exactly | 16:58 |
| gordc | luis_: cool, i did not know that. i guess we can do that clean up in another patch. | 16:59 |
| luis_ | I tried doing that, but the sphinx build needs them | 16:59 |
| gordc | luis_: thanks for explanation | 16:59 |
| luis_ | no problem, thank you | 16:59 |
| *** ildikov_ has joined #openstack-ceilometer | 17:06 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Make sure use IPv6 sockets for ceilometer in IPv6 environment https://review.openstack.org/67786 | 17:22 |
| *** tomixxx5 has joined #openstack-ceilometer | 17:29 | |
| ildikov_ | dhellmann: hi. are you around? | 17:29 |
| *** yassine has quit IRC | 17:30 | |
| *** sayalilunkad has joined #openstack-ceilometer | 17:31 | |
| *** sayalilunkad has quit IRC | 17:31 | |
| *** annegentle has quit IRC | 17:31 | |
| *** annegentle has joined #openstack-ceilometer | 17:32 | |
| *** nacim has quit IRC | 17:54 | |
| *** admin0 has quit IRC | 17:58 | |
| *** admin0 has joined #openstack-ceilometer | 17:59 | |
| *** admin0 has quit IRC | 18:00 | |
| *** sayalilunkad has joined #openstack-ceilometer | 18:05 | |
| *** tburnes has quit IRC | 18:06 | |
| *** ildikov_ has quit IRC | 18:08 | |
| *** ildikov_ has joined #openstack-ceilometer | 18:09 | |
| *** _nadya_ has joined #openstack-ceilometer | 18:20 | |
| *** anamalagon has joined #openstack-ceilometer | 18:21 | |
| *** sdake_ is now known as sdake | 18:34 | |
| *** changbl has quit IRC | 18:39 | |
| *** tburnes has joined #openstack-ceilometer | 18:51 | |
| *** promulo has quit IRC | 18:58 | |
| *** alexpilotti has joined #openstack-ceilometer | 19:10 | |
| alexpilotti | dhellmann: ping | 19:11 |
| *** tomixxx5 has quit IRC | 19:14 | |
| *** _nadya_ has quit IRC | 19:15 | |
| *** changbl has joined #openstack-ceilometer | 19:29 | |
| *** yassine has joined #openstack-ceilometer | 19:31 | |
| *** _nadya_ has joined #openstack-ceilometer | 19:36 | |
| *** malini is now known as malini_afk | 19:37 | |
| *** admin0 has joined #openstack-ceilometer | 19:38 | |
| *** _nadya_ has quit IRC | 19:49 | |
| dhellmann | ildikov_, alexpilotti : pong | 19:51 |
| alexpilotti | dhellmann: hi | 19:51 |
| alexpilotti | dhellmann: I saw your comment here: https://review.openstack.org/#/c/78802/ | 19:52 |
| ildikov_ | dhellmann: hi, I finally answered my question by testing that fix, thanks | 19:52 |
| ildikov_ | dhellmann: when do you plan to make a release? | 19:52 |
| alexpilotti | dhellmann: we already conbert to MB, see link in the comment | 19:52 |
| dhellmann | ildikov_: sphinxcontrtib-pecanwsme, right? I'll try to do that today. | 19:52 |
| ildikov_ | dhellmann: yes | 19:53 |
| dhellmann | alexpilotti: ah, I missed that, thanks | 19:53 |
| dhellmann | alexpilotti: I was focusing on the key name change, not the math :-/ | 19:53 |
| dhellmann | alexpilotti: +2a | 19:54 |
| ildikov_ | dhellmann: sounds good, thanks | 19:54 |
| *** changbl has quit IRC | 20:00 | |
| *** promulo has joined #openstack-ceilometer | 20:01 | |
| openstackgerrit | gordon chung proposed a change to openstack/pycadf: add OpenStack Foundation as author https://review.openstack.org/81066 | 20:04 |
| alexpilotti | dhellmann: tx! | 20:07 |
| jaypipes | gordc: would you agree that getting the pagination patches for the sqlalchemy driver in asap and enabling a default limit to the get_resources() method would be an easy win for the performance of that method in the gate jobs? | 20:10 |
| * jaypipes been thinking of the easiest ways to reduce the pressure in the gate job. | 20:11 | |
| jaypipes | without massive changes | 20:11 |
| jaypipes | gordc: also, IIRC, somebody already removed the meter dependent sub-gets in the get_resources() method... | 20:11 |
| gordc | jaypipes: i'm not sure pagination is easiest solution... one of the reasons it's not merged is there was so much controversy on how to correctly implement it. | 20:12 |
| gordc | jaypipes: there is the ability to filter by time but that's not really for gates. | 20:13 |
| jaypipes | gordc: right, there is, as I pointed out, but there isn't any *default* filter :) | 20:13 |
| gordc | i believe the patch i'm working on should solve the gate issues... beyond that pagination or limits would probably be needed. | 20:13 |
| jaypipes | cool. | 20:13 |
| gordc | jaypipes: what kind of defaults were thinking of? | 20:14 |
| jaypipes | gordc: oh, anything really... :) one minute? two minutes? | 20:14 |
| jaypipes | gordc: at a minimum, we should be able to change the tempest gate test to pass something like 10 seconds. | 20:14 |
| dhellmann | ildikov_: new version of sphinxcontrib-pecanwsme released as 0.7.1 | 20:15 |
| ildikov_ | dhellmann: thanks! | 20:16 |
| gordc | jaypipes: hmm... do you mean to add a filter to tempest test or literally change what http://url/v2/resources returns by default? the latter probably won't be a good idea... | 20:16 |
| jaypipes | gordc: I mean to propose both. | 20:16 |
| jaypipes | gordc: first is easy, latter is more involved of course. | 20:17 |
| jaypipes | gordc: and latter is backwards incompatible, so would need to be done on a next major version of API. | 20:17 |
| gordc | jaypipes: agreed. i'd put game for the former... | 20:17 |
| jaypipes | absolutely, they aren't mutually exclusive :) | 20:17 |
| gordc | jaypipes: the latter would be highly dependent on what meters/loads a user has... i don't know any good way to gauge what a good 'average' limit is. | 20:18 |
| jaypipes | gordc: a config optiohn? :) | 20:18 |
| gordc | jaypipes: i guess. but you might as well define a timestamp filter then. :) although i guess you'd have to consistently caculate when the start timestamp is. | 20:19 |
| gordc | jaypipes: i guess having an option might work... we'll see how these incremental patches help for now. | 20:20 |
| jaypipes | gordc: well that is easy enough.. if the config option is something like default_since_filter_seconds and the value is 60, then you would set the filter for start_timestamp to utcnow() - 60 seconds, yeah? | 20:20 |
| gordc | jaypipes: agreed. i just don't want more options if we can avoid them. | 20:21 |
| jaypipes | :) understood completely! | 20:21 |
| openstackgerrit | A change was merged to openstack/pycadf: add OpenStack Foundation as author https://review.openstack.org/81066 | 20:29 |
| *** admin0 has left #openstack-ceilometer | 20:35 | |
| *** julim_ has quit IRC | 20:37 | |
| *** yassine has quit IRC | 20:38 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Fix bug in get_capabilities behavior in DB drivers https://review.openstack.org/80852 | 20:43 |
| openstackgerrit | A change was merged to openstack/ceilometer: 028 migration script incorrectly skips over section https://review.openstack.org/80455 | 20:44 |
| *** ravikumar_hp has joined #openstack-ceilometer | 20:44 | |
| ravikumar_hp | One question - what is the Ceilometer version for Icehouse release . is it V2? | 20:47 |
| *** rwsu has quit IRC | 21:16 | |
| *** admin0_ has joined #openstack-ceilometer | 21:31 | |
| *** admin0_ has quit IRC | 21:31 | |
| *** rwsu has joined #openstack-ceilometer | 21:31 | |
| *** changbl has joined #openstack-ceilometer | 21:38 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Fixes Hyper-V Inspector network metrics values https://review.openstack.org/78799 | 21:50 |
| *** ryanpetrello has quit IRC | 22:17 | |
| *** openstackgerrit has quit IRC | 22:39 | |
| *** openstackgerrit has joined #openstack-ceilometer | 22:39 | |
| *** _nadya_ has joined #openstack-ceilometer | 22:49 | |
| *** _nadya_ has quit IRC | 22:53 | |
| *** malini_afk is now known as malini | 23:02 | |
| *** sayalilunkad has quit IRC | 23:05 | |
| *** ryanpetrello has joined #openstack-ceilometer | 23:13 | |
| *** dhellmann has quit IRC | 23:18 | |
| *** dhellmann has joined #openstack-ceilometer | 23:23 | |
| *** ryanpetrello has quit IRC | 23:30 | |
| *** ryanpetrello has joined #openstack-ceilometer | 23:42 | |
| *** ryanpetrello has quit IRC | 23:47 | |
| *** ravikumar_hp has quit IRC | 23:47 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!