| *** openstack has joined #openstack-marconi | 13:49 | |
| tongli | Hi, guys, I read the marconi api doc here http://developer.rackspace.com/blog/openstack-marconi-api.html and it says , Marconi support posting 10 msgs at the same time. | 13:51 |
|---|---|---|
| tongli | wonder if the number 10 is an example or it is the up limit? | 13:52 |
| tongli | and the reason why it has such limit. | 13:52 |
| tongli | any one can help with that question will be really appreciated. | 13:52 |
| *** amitgandhi has joined #openstack-marconi | 13:53 | |
| tongli | if a consumer wants to have 10 msgs in one GET request, but there are only 4 on the server, will the request be blocked or marconi will return with 4? | 13:54 |
| malini1 | tongli: marconi will return 4 in tht case | 13:56 |
| sriram | tongli: it should return with 4 | 13:56 |
| sriram | wrt to your other question | 13:56 |
| tongli | @malini1 @sriram, thanks for that answer. makes a lot of sense. | 13:57 |
| tongli | I assume that 10 is the up limit, can not post or get more than 10 in one request, right? | 13:57 |
| *** abettadapur has joined #openstack-marconi | 13:59 | |
| sriram | there is a max_message_size. So I guess the the limit for posting messages, should be such that the entirety of the content length should be less than that. | 14:00 |
| flaper87 | Goooood morning ppl | 14:00 |
| flaper87 | tongli: sriram malini1 FYI, I deployed latest Marconi here: http://marconi-fpercoco.rhcloud.com/ | 14:01 |
| malini1 | tongli: for some reason I have it in my head that max messages tht can be posted is a configurable value grrrr..But cant find that in the code. | 14:01 |
| malini1 | flaper87: goooooood whatever !!! | 14:01 |
| flaper87 | you can play with it using marconi client, curl or whatever client you prefer | 14:01 |
| flaper87 | malini1: LOL | 14:01 |
| malini1 | flaper87: do we just chk or max message size when posting messages? | 14:01 |
| * flaper87 is in Europe now | 14:01 | |
| sriram | flaper87: woooot :D | 14:01 |
| sriram | nice | 14:01 |
| *** amitgandhi has quit IRC | 14:01 | |
| malini1 | flaper87: yayyy! | 14:02 |
| flaper87 | malini1: max message size | 14:02 |
| flaper87 | AFAIR | 14:02 |
| flaper87 | :P | 14:02 |
| *** amitgandhi has joined #openstack-marconi | 14:02 | |
| malini1 | flaper87: tht is what the code has too..my brain has a bug :D | 14:02 |
| sriram | brain bug :O | 14:03 |
| * flaper87 does an emergency surgery on malini1's brain | 14:03 | |
| *** whenry has joined #openstack-marconi | 14:03 | |
| malini1 | in coma now | 14:03 |
| sriram | flaper87: :O you did the surgery on the wrong malini :P | 14:04 |
| sriram | malini2 will be here soon :P | 14:04 |
| malini1 | :D | 14:04 |
| flaper87 | LOL | 14:04 |
| malini1 | tongli: Hope you got your answers | 14:04 |
| tongli | @malini1, if I got your guys right, the 10 is just the default value for the parameter max message size. | 14:05 |
| *** ykaplan has quit IRC | 14:06 | |
| sriram | the default size is 256kb | 14:06 |
| sriram | the sum of content length of your messages shouldnt exceed that | 14:07 |
| sriram | flaper87: ^ does that seem correct? | 14:07 |
| sriram | tongli: ^ | 14:07 |
| malini1 | sriram is right ..also the max message size is a configurable value | 14:07 |
| malini1 | so ur deployment can be configured to have a value other than 256 kn | 14:08 |
| tongli | can not find it in the code. | 14:08 |
| flaper87 | sriram: correct | 14:08 |
| malini1 | tongli: https://github.com/openstack/marconi/blob/master/marconi/queues/transport/validation.py | 14:08 |
| flaper87 | tongli: https://github.com/openstack/marconi/blob/master/marconi/queues/transport/validation.py#L27 | 14:08 |
| flaper87 | malini1: argghh | 14:08 |
| flaper87 | >.> | 14:08 |
| malini1 | flaper87: but you were more specific ;) | 14:08 |
| flaper87 | w0000t | 14:08 |
| malini1 | the #L27 is what took u more time :D | 14:09 |
| tongli | oh, it is not by number of messages, it is by the size of the request body if I read that code right. | 14:09 |
| flaper87 | tongli: correct | 14:10 |
| tongli | http://developer.rackspace.com/blog/openstack-marconi-api.html | 14:10 |
| flaper87 | it doesn't matter if you've 1 message of 256kb or 10 that added up make 256kb | 14:10 |
| tongli | that docs indicated it is controlled by number of messages, which seems to me not the same thing as message body. | 14:10 |
| flaper87 | tongli: that's old ;) | 14:10 |
| tongli | ah, ok. | 14:11 |
| flaper87 | tongli: we're working on a proper documentation | 14:11 |
| flaper87 | tongli: http://docs.openstack.org/developer/marconi/installing.html | 14:11 |
| tongli | cool, thanks for these excellent answers. | 14:11 |
| flaper87 | WIP | 14:11 |
| flaper87 | tongli: can I ask what your use case for marconi is? | 14:12 |
| flaper87 | are you just looking at it? | 14:12 |
| flaper87 | are you thinking of installing it? | 14:12 |
| tongli | @flaper87, I was told you guys using falcon framework and using ReST API handling messages, | 14:13 |
| tongli | I would like to look at the code and see if we can get first hand experience if falcon is indeed a good framework and it performs under heavy load. | 14:14 |
| tongli | there is a need for monitoring as a service, and few big guys (company) are trying to do a new project, I am thinking if we can simply take marconi and make some changes to it to meet the need. | 14:15 |
| tongli | @flaper87 ,hope that help a bit with your question. | 14:15 |
| *** rwsu has joined #openstack-marconi | 14:16 | |
| tongli | marconi takes message and place it into a mongodb for persistance ? | 14:16 |
| flaper87 | tongli: it does help, thanks a lot :) | 14:17 |
| flaper87 | tongli: yes, as of now mongodb is the only production-recommended store we have | 14:17 |
| tongli | have you guys found falcon is a lot faster comparing to other python frameworks? | 14:18 |
| tongli | in terms ReSTful API support. | 14:18 |
| flaper87 | tongli: based on some benchmarks it is | 14:18 |
| flaper87 | that's the whole point of falcon, being very minimalist and fast | 14:18 |
| *** vkmc has joined #openstack-marconi | 14:18 | |
| *** vkmc has quit IRC | 14:18 | |
| *** vkmc has joined #openstack-marconi | 14:18 | |
| flaper87 | tongli: kgriffs is falcon's leader but he won't be around this week | 14:18 |
| tongli | any drawbacks or gotches in your experience? | 14:19 |
| flaper87 | tongli: alcabrera|afk has contributed quite a bit to falcon too | 14:19 |
| flaper87 | tongli: drawback is that you don't get many of the semi-automatic-builtin-framework-like features | 14:19 |
| tongli | @flaper87, very well aware of that (kgriffs made falcon) | 14:19 |
| flaper87 | but as you can see in marconi's code, the code is still clean, clear and small | 14:19 |
| tongli | @flaper87, thanks so much. | 14:20 |
| flaper87 | tongli: my pleasure | 14:21 |
| tongli | wonder how you inform consumers there are messages? | 14:21 |
| tongli | or consumers simply do pull (GET requests) when the clients feel like to? | 14:21 |
| flaper87 | tongli: as of now, that's the supported message pattern. Users need to poll for messages | 14:22 |
| flaper87 | we're working (or suppose to be working) on push messages for Juno | 14:22 |
| tongli | pointer to the BP if there is one? | 14:23 |
| flaper87 | tongli: it'll happen as part of this work: https://blueprints.launchpad.net/marconi/+spec/notifications | 14:24 |
| tongli | @flaper87, thanks so much, hope see you guys in Paris. | 14:24 |
| flaper87 | tongli: absolutely, looking forward to it. | 14:24 |
| *** cpallares has joined #openstack-marconi | 14:27 | |
| *** ChanServ sets mode: +o kgriffs | 14:27 | |
| *** kgriffs is now known as kgriffs|afk | 14:28 | |
| *** ykaplan has joined #openstack-marconi | 14:29 | |
| prashanthr_ | Good morning all :) | 14:31 |
| vkmc | morning! :) | 14:32 |
| vkmc | hi prashanthr_! | 14:32 |
| prashanthr_ | hi vkmc , how u doing ? | 14:32 |
| vkmc | I woke up with an intense headache -.- now I'm feeling better though... hoping to recover a bit from the morning | 14:34 |
| vkmc | prashanthr_, you? | 14:34 |
| flaper87 | vkmc: heyyyyy | 14:34 |
| prashanthr_ | vkmc: Co-incidence, i had a headache today as well | 14:36 |
| prashanthr_ | and had to shift homes with that. | 14:36 |
| vkmc | flaper87, hiiiiii :) o/ | 14:40 |
| malini1 | prashanthr_, vkmc: tylenol pop-tarts for you! | 14:41 |
| malini1 | hope you feel better | 14:41 |
| vkmc | prashanthr_, boo | 14:41 |
| vkmc | malini1, hiiii :) | 14:41 |
| prashanthr_ | malini1: thank you :) | 14:41 |
| malini1 | hello vkmc, prashanthr_@ | 14:41 |
| vkmc | malini1, thanks | 14:42 |
| malini1 | On a side note, we got another tempest patch merged over th weekend.. | 14:42 |
| *** stannie has left #openstack-marconi | 14:42 | |
| vkmc | malini1, I saw it... congrats! | 14:42 |
| malini1 | With tht, we have positive tests for all v1 APIs in Tempest | 14:42 |
| malini1 | thanks vkmc! | 14:43 |
| *** Obulpathi has quit IRC | 14:43 | |
| flaper87 | vkmc: we gotta talk | 14:43 |
| flaper87 | vkmc: you're doing a very good job and we don't like that here | 14:43 |
| vkmc | malini1, that one was 1 month ++ on the gates... it was time! | 14:43 |
| flaper87 | Marconi is full of crappy coders | 14:43 |
| flaper87 | you ain't be good | 14:43 |
| *** Obulpathi has joined #openstack-marconi | 14:43 | |
| malini1 | vkmc: yes ! | 14:44 |
| vkmc | flaper87, lol | 14:44 |
| *** AAzza is now known as AAzza_afk | 14:44 | |
| malini1 | I am sure flaper87 typed in the wrong channel :D | 14:44 |
| * flaper87 hasn't had enough gummy bears lately | 14:44 | |
| flaper87 | malini1: LOL, no no. I sent those to the right channel | 14:45 |
| vkmc | flaper87, that's not good, your sugar must be low | 14:45 |
| flaper87 | vkmc: IT ISSSSS :( | 14:45 |
| flaper87 | I just had like 3 packs today, that's not enough | 14:45 |
| vkmc | flaper87, which is the daily minimum amount of gummy bears for a flaper? | 14:48 |
| flaper87 | vkmc: That x10^100000 | 14:48 |
| vkmc | lol | 14:49 |
| *** AAzza_afk is now known as AAzza | 14:49 | |
| AAzza | prashanthr_: ping | 14:53 |
| *** ametts has joined #openstack-marconi | 14:54 | |
| prashanthr_ | AAzza: pong :) | 14:54 |
| prashanthr_ | good morning | 14:54 |
| AAzza | prashanthr_: good evening) | 14:55 |
| AAzza | prashanthr_: i want to talk with your about ways we can support py3k compatibility of your redis driver | 14:56 |
| prashanthr_ | Sure. | 14:56 |
| AAzza | prashanthr_: it should be difficult, but it is important) | 14:57 |
| prashanthr_ | I understand the importance of it. difficult is good :) | 14:57 |
| AAzza | prashanthr_: should not be very difficult) | 14:57 |
| AAzza | prashanthr_: missing words( | 14:57 |
| AAzza | prashanthr_: basically first step is to try to run tests under py33) | 14:58 |
| prashanthr_ | AAzza : okay. just py33 is good enough ? | 14:58 |
| AAzza | prashanthr_: if you need help with it, i' m here. I could try your driver and say what is the problem zones. | 14:58 |
| prashanthr_ | Sure, I will start up with it. I have some basic driver tests written up | 14:59 |
| AAzza | prashanthr_: basically yes, py33 enough, | 14:59 |
| prashanthr_ | but have'nt worked with them using py33 | 14:59 |
| AAzza | prashanthr_: most problems should be with strings\bytes | 14:59 |
| AAzza | prashanthr_: such as results from redis will be in bytes (just my predictions from experience) | 15:00 |
| AAzza | prashanthr_: just encode/decode staff in proper places should solve most problems | 15:00 |
| AAzza | prashanthr_: many thanks | 15:00 |
| prashanthr_ | okay. That's a common issue. But AFAIK redis library is py3k compatible. | 15:00 |
| AAzza | prashanthr_: if you need any-any help - i'm here) | 15:00 |
| *** tedross has quit IRC | 15:01 | |
| prashanthr_ | Thanks a lot. Will startup soon and come back with some bugs. :) Great effort from your side must say related to py3k. | 15:01 |
| AAzza | prashanthr_: it was pretty exiting to see py33 failing tests count decreasing) thanks) | 15:02 |
| AAzza | prashanthr_: honesty, I wanted to do your project with redis for internship initially) | 15:02 |
| prashanthr_ | ha ha yeah. I remember you once said "5 versions of py is not what someone expects from life". | 15:03 |
| prashanthr_ | Is it ? Yeah redis is pretty awesome. Fits the bill well for Marconi. | 15:03 |
| AAzza | prashanthr_: aha, it is awesome) and your are doing good job with it | 15:04 |
| prashanthr_ | thank you :) | 15:05 |
| *** ykaplan has quit IRC | 15:06 | |
| *** tedross has joined #openstack-marconi | 15:16 | |
| AAzza | malini1: hi, i have a question about your POP review. You added new method to storage/base.py, shouldn't it be added also to storage/pooling controller's wrapper? | 15:17 |
| *** chandan_kumar has quit IRC | 15:18 | |
| *** ykaplan has joined #openstack-marconi | 15:18 | |
| *** prashanthr_ has quit IRC | 15:20 | |
| *** oz_akan_ has quit IRC | 15:22 | |
| *** oz_akan_ has joined #openstack-marconi | 15:23 | |
| malini1 | AAzza: good question..I didnt think I have to. flaper87: thoughts? | 15:23 |
| flaper87 | malini1: yup, you've to add it there too | 15:24 |
| flaper87 | AAzza: good catch | 15:24 |
| malini1 | thanks AAzza!! you are our new jenkins ;) | 15:25 |
| AAzza | flaper87: i just noticed earlier, that our /pooling is not precisely reflecting /base | 15:25 |
| flaper87 | AAzza: :( | 15:25 |
| flaper87 | it should, we need to add a test for that | 15:25 |
| flaper87 | or find a better way to do it | 15:25 |
| AAzza | flaper87: yeah, i was tried to add the tests for storage/pooling and see these problems. | 15:26 |
| *** oz_akan_ has quit IRC | 15:27 | |
| AAzza | flaper87: btw if you read my comment in the review i give a link to you in the morning, you know it a little bit earlier:P | 15:27 |
| flaper87 | AAzza: ahahahahahaha, I'm sorryyyyyyyyyyyyyyyyyyyyyy | 15:28 |
| * flaper87 has been under a pile of things to finish | 15:28 | |
| flaper87 | I'll read it before heading to bed :P | 15:28 |
| flaper87 | and that means I may be reading it at 3am :P | 15:28 |
| AAzza | flaper87: oh, i appreciate this:) | 15:30 |
| flaper87 | :D | 15:30 |
| flaper87 | :D | 15:30 |
| * flaper87 is messing with AAzza | 15:30 | |
| * flaper87 will read the patch | 15:30 | |
| *** oz_akan_ has joined #openstack-marconi | 15:53 | |
| *** ykaplan has quit IRC | 15:53 | |
| *** oz_akan__ has joined #openstack-marconi | 15:55 | |
| *** oz_akan_ has quit IRC | 15:57 | |
| *** AAzza_afk has joined #openstack-marconi | 15:58 | |
| *** chandan_kumar has joined #openstack-marconi | 15:59 | |
| *** oz_akan__ has quit IRC | 15:59 | |
| *** AAzzal has joined #openstack-marconi | 16:00 | |
| *** AAzza has quit IRC | 16:01 | |
| *** AAzzal is now known as AAzza | 16:01 | |
| *** AAzza_afk has quit IRC | 16:04 | |
| *** reed_ has joined #openstack-marconi | 16:05 | |
| *** tedross has quit IRC | 16:45 | |
| *** oz_akan_ has joined #openstack-marconi | 16:49 | |
| *** oz_akan_ has quit IRC | 16:54 | |
| *** tedross has joined #openstack-marconi | 17:08 | |
| *** cpallares has quit IRC | 17:35 | |
| *** oz_akan_ has joined #openstack-marconi | 17:43 | |
| *** oz_akan_ has quit IRC | 17:48 | |
| *** oz_akan_ has joined #openstack-marconi | 18:02 | |
| *** oz_akan_ has quit IRC | 18:03 | |
| *** oz_akan_ has joined #openstack-marconi | 18:03 | |
| *** cpallares has joined #openstack-marconi | 18:29 | |
| *** oz_akan_ has quit IRC | 18:30 | |
| *** oz_akan_ has joined #openstack-marconi | 18:31 | |
| *** oz_akan__ has joined #openstack-marconi | 18:34 | |
| *** oz_akan_ has quit IRC | 18:38 | |
| *** Obulpathi has quit IRC | 18:51 | |
| *** Obulpathi has joined #openstack-marconi | 18:52 | |
| *** oz_akan__ has quit IRC | 18:52 | |
| *** oz_akan_ has joined #openstack-marconi | 18:53 | |
| *** Obulpathi has quit IRC | 18:56 | |
| *** Obulpathi has joined #openstack-marconi | 18:56 | |
| *** oz_akan_ has quit IRC | 18:57 | |
| *** oz_akan_ has joined #openstack-marconi | 19:10 | |
| *** oz_akan_ has quit IRC | 19:24 | |
| *** keith_newstadt has quit IRC | 19:25 | |
| *** malini1 has quit IRC | 19:30 | |
| *** reed_ is now known as reed | 19:36 | |
| *** reed has quit IRC | 19:36 | |
| *** reed has joined #openstack-marconi | 19:36 | |
| *** amitgandhi has quit IRC | 19:38 | |
| *** amitgandhi has joined #openstack-marconi | 19:38 | |
| *** oz_akan_ has joined #openstack-marconi | 19:39 | |
| *** chandan_kumar has quit IRC | 19:52 | |
| *** oz_akan_ has quit IRC | 20:04 | |
| *** oz_akan_ has joined #openstack-marconi | 20:04 | |
| vkmc | flaper87, what was the meaning of '~' in amqp://~0.0.0.0 again? | 20:11 |
| vkmc | I'm recall you mentioned that it meant to use the default port, but not I'm not sure since it's failing on amqp://127.0.0.1:8888 | 20:12 |
| *** fifieldt has quit IRC | 20:25 | |
| *** fifieldt has joined #openstack-marconi | 20:26 | |
| *** sriram has quit IRC | 20:33 | |
| *** abettadapur has quit IRC | 20:34 | |
| *** Obulpathi has quit IRC | 21:03 | |
| flaper87 | vkmc: it means, it will listen on that address instead of trying to connect to it | 21:07 |
| vkmc | flaper87, makes sense :) thanks mentor | 21:13 |
| flaper87 | vkmc: anytime :) | 21:13 |
| *** cpallares has quit IRC | 21:15 | |
| *** oz_akan__ has joined #openstack-marconi | 22:05 | |
| *** oz_akan_ has quit IRC | 22:08 | |
| *** oz_akan__ has quit IRC | 22:10 | |
| *** tedross has quit IRC | 22:16 | |
| *** amitgandhi has quit IRC | 22:30 | |
| *** flaper87 is now known as flaper87|afk | 22:34 | |
| openstackgerrit | A change was merged to openstack/marconi: Add None checking for the Mongodb uri https://review.openstack.org/103307 | 22:37 |
| *** flwang has quit IRC | 22:38 | |
| *** flwang has joined #openstack-marconi | 22:56 | |
| *** openstackgerrit has quit IRC | 23:19 | |
| *** openstackgerrit has joined #openstack-marconi | 23:21 | |
| *** amitgandhi has joined #openstack-marconi | 23:31 | |
| *** kgriffs|afk has quit IRC | 23:41 | |
| *** kgriffs|afk has joined #openstack-marconi | 23:41 | |
| *** kgriffs|afk is now known as kgriffs | 23:42 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!