| *** pmackinn has joined #openstack-trove | 00:10 | |
| *** ashleighfarnham has joined #openstack-trove | 00:14 | |
| *** ashleighfarnham has quit IRC | 00:19 | |
| *** pmackinn has quit IRC | 00:26 | |
| *** pmackinn has joined #openstack-trove | 00:34 | |
| *** pmackinn has quit IRC | 00:37 | |
| *** chlong has quit IRC | 00:52 | |
| *** mingdang1 has joined #openstack-trove | 00:52 | |
| *** _amrith_ is now known as amrith | 00:59 | |
| *** erkules has joined #openstack-trove | 01:09 | |
| *** erkules_ has quit IRC | 01:12 | |
| *** haomaiwang has quit IRC | 01:12 | |
| *** mingdang1 has quit IRC | 01:13 | |
| *** dbennett has joined #openstack-trove | 01:19 | |
| dbennett | @cp16net ping | 01:20 |
|---|---|---|
| *** pmackinn has joined #openstack-trove | 01:25 | |
| *** pmackinn has quit IRC | 01:42 | |
| *** mingdang1 has joined #openstack-trove | 01:43 | |
| *** ashleighfarnham has joined #openstack-trove | 01:59 | |
| *** rjaiswal has quit IRC | 02:04 | |
| *** ashleighfarnham has quit IRC | 02:04 | |
| *** chlong has joined #openstack-trove | 02:11 | |
| *** lexholden has quit IRC | 02:11 | |
| *** mingdang1 has quit IRC | 02:14 | |
| amrith | sushilkm, yt? | 02:22 |
| amrith | sushilkm, when you get this would you please ping me. | 02:22 |
| amrith | would like to get to the bottom of the datastore/version spec quickly | 02:22 |
| amrith | and move it along. | 02:22 |
| amrith | cp16net, yt? | 02:22 |
| amrith | cp16net ^^ | 02:22 |
| *** esmute_ has quit IRC | 02:24 | |
| *** esmute_ has joined #openstack-trove | 02:53 | |
| cp16net | hello | 02:57 |
| cp16net | i just got back to my puter | 02:58 |
| amrith | cp16net, howdy! | 03:19 |
| amrith | gots a couples of minutez to chat | 03:20 |
| amrith | I knew I wanted to chat about something | 03:20 |
| amrith | let me see if I can remember | 03:20 |
| amrith | ah yes. datastore version spec ... would you please take a look at my comments. I'd like to get to some conclusion of it quickly; it seems to be getting old. | 03:21 |
| cp16net | oh yeah | 03:26 |
| cp16net | so you mentioned why have the body? | 03:27 |
| cp16net | for the patch or post on the response | 03:27 |
| cp16net | sushilkm: around? | 03:28 |
| cp16net | amrith: you left? | 03:30 |
| cp16net | maybe you are distracted like i am :-P | 03:30 |
| *** amrith is now known as _amrith_ | 03:34 | |
| *** coolsvap|away is now known as coolsvap | 03:34 | |
| *** ashleighfarnham has joined #openstack-trove | 03:39 | |
| *** rjaiswal has joined #openstack-trove | 03:42 | |
| *** ashleighfarnham has quit IRC | 03:44 | |
| cp16net | _amrith_: disappeared | 03:45 |
| cp16net | not to be confused with amrith | 03:46 |
| openstackgerrit | Peter Stachowski proposed openstack/trove: Add support for Redis replication https://review.openstack.org/203278 | 03:51 |
| *** esmute_ has quit IRC | 04:04 | |
| *** _amrith_ is now known as amrith | 04:07 | |
| amrith | cp16net, I'm back | 04:11 |
| amrith | I have magical powers | 04:12 |
| amrith | I can disappear from time to time ;) | 04:12 |
| cp16net | me too | 04:29 |
| cp16net | i bet you are not here tho | 04:30 |
| *** georgelorch has quit IRC | 04:30 | |
| *** georgelorch has joined #openstack-trove | 04:32 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/trove: Updated from global requirements https://review.openstack.org/203152 | 04:43 |
| *** sgotliv_ has quit IRC | 04:47 | |
| amrith | i'm here | 04:52 |
| amrith | i'm not sure why my hexchat isn't notifying me. | 04:52 |
| amrith | cp16net, yt? | 04:52 |
| georgelorch | dbennett, cp16net IIRC is on east coast US and lefred was helping out. Check back in tomorrow AM. Thanks aging guys, you'll never have to buy a round if I'm present. | 04:52 |
| amrith | maybe that will make your irc client blink | 04:52 |
| cp16net | lol | 04:52 |
| cp16net | yeah something poped out at me | 04:52 |
| amrith | georgelorch, you really are a peach. What do you mean "Thanks aging guys"? | 04:53 |
| amrith | Who is the "aging guys" | 04:53 |
| cp16net | georgelorch: i'm CST in Texas | 04:53 |
| amrith | georgelorch is AZ | 04:53 |
| georgelorch | LOL I meant again, it seems I'm the one aging the most | 04:53 |
| amrith | Yeah right | 04:53 |
| cp16net | :-D | 04:53 |
| amrith | as they would say in TX, "turd blossom" which apparently is a term of high endearment. | 04:54 |
| amrith | why, I know not. | 04:54 |
| amrith | I'm not a texan | 04:54 |
| georgelorch | ahh cp16net, cool, dbennett is actually really close to you, lefred is in Belgium | 04:54 |
| cp16net | there much stranger terms i've heard | 04:54 |
| cp16net | but perfer not to use them in public | 04:54 |
| georgelorch | ans amrith is well...he's amrith, what else is there to say | 04:55 |
| amrith | oh puhleeze! | 04:55 |
| amrith | georgelorch, amen to that. | 04:55 |
| cp16net | ... yeah i dunno what else to say | 04:55 |
| amrith | so, cp16net, about the datastore registration spec | 04:55 |
| cp16net | word | 04:56 |
| amrith | I wonder if sushilkm is around | 04:56 |
| amrith | is he up in seattle now or back home? | 04:56 |
| cp16net | i'm not sure | 04:56 |
| cp16net | i think home | 04:56 |
| amrith | me too | 04:56 |
| cp16net | cool so you can knock on his door | 04:57 |
| cp16net | :-P | 04:57 |
| amrith | let me do that. one second, I think I have his phone number at least. | 04:57 |
| cp16net | lolz | 04:57 |
| cp16net | doesnt seem to be online at least | 04:57 |
| amrith | calling him | 04:59 |
| amrith | he'll be getting online in a couple of minutes | 05:00 |
| amrith | just spooked with him. | 05:00 |
| cp16net | ok cool | 05:00 |
| cp16net | so the registration... | 05:01 |
| amrith | yes ... | 05:01 |
| amrith | that thing | 05:01 |
| amrith | ok to just go through from the top? | 05:01 |
| cp16net | ok | 05:01 |
| amrith | so at a high level is this intended to be a synchronous or async api? | 05:02 |
| amrith | the spec talks about a return code of 200 and accepted. | 05:02 |
| sushilkm | hello amrith cp16net | 05:02 |
| amrith | not sure which he intended. | 05:02 |
| cp16net | since it just updates the db i'd say sync | 05:03 |
| amrith | hi sushilkm ... | 05:03 |
| cp16net | yeah thats something i brought up as well | 05:03 |
| amrith | then the code must be updated. | 05:03 |
| amrith | either 202 accepted or 200 success | 05:03 |
| cp16net | yeah 200 OK | 05:03 |
| * SlickNik joins the party | 05:03 | |
| * amrith wonders whether SlickNik brought anything nice to drink | 05:03 | |
| cp16net | 202 for async stuffs | 05:03 |
| cp16net | not allowed in the party unless drinks in hand | 05:04 |
| cp16net | :-P | 05:04 |
| SlickNik | 200 - success would be my call, fwiw | 05:04 |
| * SlickNik returns to the kitchen to grab some beers. | 05:04 | |
| amrith | hey! I had a shitload more comments | 05:04 |
| amrith | where did they go? | 05:04 |
| * amrith needs more beer | 05:04 | |
| cp16net | patch before? | 05:04 |
| *** esmute_ has joined #openstack-trove | 05:05 | |
| amrith | looking | 05:05 |
| amrith | don't see them | 05:05 |
| sushilkm | are you saing "200 OK" is not good | 05:05 |
| cp16net | no | 05:06 |
| cp16net | we are saying the spec says... 200 accepted | 05:06 |
| cp16net | which is contradicting | 05:06 |
| amrith | you are correct | 05:06 |
| cp16net | should be 200 OK or SUCCESS | 05:06 |
| amrith | patch set 18 | 05:06 |
| amrith | just pushed them. | 05:06 |
| sushilkm | yeah i see that i updated the code but missed to update the message | 05:06 |
| amrith | as cp16net says ... 200 OK | 05:07 |
| amrith | not 200 Acepted | 05:07 |
| amrith | accepted | 05:07 |
| amrith | comments pushed on PS 18 | 05:07 |
| amrith | let me walk through those quickly | 05:07 |
| * cp16net feels accomplished (just solved my rubics cube again) | 05:08 | |
| amrith | re: patch (around line 220) | 05:08 |
| amrith | so the patch is allowing me to update some parameters. | 05:08 |
| amrith | my comment is longish so let me not retype it here; y'all can look there. | 05:08 |
| * cp16net reading | 05:09 | |
| *** esmute_ has quit IRC | 05:09 | |
| * amrith wonders how long it will take SlickNik to return with beers. | 05:09 | |
| amrith | I noticed a typo in my comment | 05:10 |
| amrith | sadly I can't fix it. | 05:10 |
| amrith | "I may be wrong here but the RESTful way to do things is to send all the data and the server ignores the things that are changed OR send only the things that are changed." | 05:10 |
| amrith | should read | 05:10 |
| amrith | I may be wrong here but the RESTful way to do things is to send all the data and the server ignores the things that are UNchanged OR send only the things that are changed. | 05:10 |
| amrith | changed word "changed" to "UNchanged" in one place. | 05:10 |
| * amrith runs for power | 05:11 | |
| cp16net | yeah that was a little off but i think i got what you mean | 05:11 |
| cp16net | so for a PATCH verb i like the idea of only sending what needs to change | 05:11 |
| sushilkm | how i had thought this was you may send the only attribute you want to update | 05:11 |
| cp16net | especially when there are 50 fields | 05:11 |
| cp16net | there is no need to send them all | 05:11 |
| sushilkm | and the attributes i have listed in spec are those which are updateable | 05:11 |
| amrith | so please don't call "image": "new-image-id" ;) | 05:12 |
| cp16net | i think showing or not a uuid in the field is a fruitless point | 05:12 |
| amrith | or you give me the feeling that I can change it. | 05:12 |
| sushilkm | thus not allowing to update datastore_version_id, datastore_version_name and datastore to which this version is attached. | 05:12 |
| cp16net | you should be able to change the id of the image that the version has | 05:13 |
| amrith | and please identify which fields are changeable and which are not in some way. | 05:13 |
| amrith | so cp16net I don't understand what "image" means. Is it a UUID | 05:13 |
| sushilkm | i did not get the point of "new-image-id" | 05:13 |
| amrith | or is it some other id. | 05:13 |
| amrith | sushilkm, I don't either | 05:13 |
| amrith | what is "new image id" | 05:13 |
| amrith | is it a glance id? | 05:14 |
| amrith | by some chance? | 05:14 |
| cp16net | image is the glance id yeah | 05:14 |
| cp16net | its what we store in the db | 05:14 |
| cp16net | to boot the instance | 05:14 |
| amrith | if it is a glance id, I understand it; I'm just not thrilled trying to guess which are UUID's and which aren't but you think that's moot. | 05:14 |
| cp16net | this is sort of a copy from the trove-manage cmd | 05:14 |
| cp16net | but putting it in the mgmt api now | 05:15 |
| amrith | cp16net, that I think is nikhil's and my issue; the trove manage command is not a good thing to point to as a good "api". | 05:15 |
| amrith | and making an API based on that is not generating a very nice API IMHO | 05:15 |
| openstackgerrit | Sergey Vilgelm proposed openstack/trove: Switch to the oslo_log library https://review.openstack.org/203828 | 05:15 |
| cp16net | no i agree | 05:15 |
| amrith | on to line 232 then. | 05:15 |
| amrith | why send me a response? | 05:15 |
| amrith | I want to patch one field. | 05:16 |
| amrith | so if you tell me you succeeded, I know what the final thing is. | 05:16 |
| amrith | why do I need to get all this crap back. | 05:16 |
| amrith | the bigger issue is whether we want this to be the API or we want to make a good API and make the trove CLI use that; deprecate the trove-manage command. | 05:17 |
| cp16net | hmm | 05:17 |
| cp16net | i invisioned thats the direction this would go | 05:17 |
| cp16net | theres no need to use the manage cmd after we can update this with the api | 05:18 |
| cp16net | and the api should be up and running any way once we get to the point of building images | 05:18 |
| amrith | so let's make an API that is not just an API version of the current trove-manage command. instead make an API without the shackles of how trove-manage did things and then write the command to use it. | 05:18 |
| cp16net | and creating datastores/versions | 05:18 |
| amrith | so the operations are these. | 05:19 |
| amrith | create a datastore | 05:19 |
| amrith | create a datastore version | 05:19 |
| amrith | set a default datastore | 05:19 |
| amrith | set a default datastore version | 05:19 |
| amrith | for each of these | 05:19 |
| amrith | list, update, delete | 05:19 |
| amrith | the workflow would be natural with that. | 05:20 |
| cp16net | well there was a thought that why should be have operation of datastore | 05:20 |
| amrith | why is that? | 05:20 |
| cp16net | that doesnt really make sense and makes the process more complicated | 05:20 |
| cp16net | when all we REALLY care about is a version | 05:20 |
| amrith | won't an operator want to add datastores? without editing files that is. | 05:20 |
| cp16net | so this hides the idea that datastore is a separate object | 05:20 |
| cp16net | if the datastore doesnt exist it would create it | 05:21 |
| cp16net | or use it | 05:21 |
| amrith | so the current trove-manage workflow is ... | 05:22 |
| cp16net | SlickNik: mentioned this and i think it makes alot of sense when you usually want to just create a version with an image and manager and other goodies | 05:22 |
| amrith | first make a datastore with no version; you'll add that later. | 05:22 |
| amrith | that's step 1. | 05:22 |
| amrith | that's also my step 1, make a datastore. | 05:22 |
| sushilkm | it was SlickNik's suggestion to simplify the process rather than immitating how trove-manage does it | 05:22 |
| amrith | we are talking about an API, not a command | 05:22 |
| amrith | your command may be to create a datastore/version in one shot; that command may make three API calls. | 05:22 |
| amrith | my point is this; command invocations do not need to have a 1:1 matching with API's. | 05:23 |
| cp16net | ok i follow | 05:23 |
| amrith | your command may be; just as it is today; trove-new-manage register-datastore-version mysql version-5.5 manager-class available-to-all this-validation-rule.json my-name min-flavor ... | 05:24 |
| amrith | that could translate into a half dozen API calls. | 05:24 |
| amrith | trove list datastores | 05:24 |
| amrith | is there a mysql? | 05:24 |
| amrith | no? | 05:24 |
| amrith | trove make-datastore mysql | 05:24 |
| amrith | now make version | 05:24 |
| amrith | then register class | 05:24 |
| amrith | then set availability | 05:24 |
| amrith | or maybe part of prev call | 05:24 |
| amrith | then set validation rules etc., | 05:24 |
| amrith | sorry; you said "ok i follow". should have shut up ;) | 05:24 |
| cp16net | i see your work flow | 05:25 |
| * cp16net processing | 05:25 | |
| * amrith calls security for a wellbeing check on SlickNik (actually a wellbeing check on the beer) | 05:26 | |
| cp16net | i think there is merit to having a separation | 05:26 |
| cp16net | before SlickNik mentioned to simplify the calls i was in the same thought you had amrith | 05:27 |
| *** ashleighfarnham has joined #openstack-trove | 05:28 | |
| amrith | let's wait and see what SlickNik has to say. in general my thought re: API's is make them small atomic things and build on that. | 05:28 |
| cp16net | yeah i think of rest apis modeled after the data | 05:29 |
| cp16net | which was what this bp showed before | 05:29 |
| cp16net | if we didnt have datastores and only versions .... | 05:30 |
| cp16net | then i think this would simplify things | 05:30 |
| cp16net | if only.... | 05:30 |
| amrith | but the datastore/version split does have some benefits | 05:32 |
| *** ashleighfarnham has quit IRC | 05:32 | |
| amrith | you get to associate a bunch of attributes with the datastore | 05:32 |
| sushilkm | it was SlickNik's suggestion on patchset 14 "the trove-manage command provides a really poor user experience ......." | 05:32 |
| amrith | and it automatically gets inherited to the version. | 05:32 |
| cp16net | umm what attrs you talking about? | 05:33 |
| cp16net | a name? | 05:33 |
| amrith | sushilkm, I agree with that comment | 05:33 |
| amrith | manager class | 05:33 |
| cp16net | the datastore table has an id, name, default_version_id | 05:34 |
| cp16net | that on the version | 05:34 |
| cp16net | i can specify what ever manager i want | 05:35 |
| amrith | cp16net, I'm saying the split as it is now is not much use; but a split (done slightly differently) could be very useful. | 05:35 |
| amrith | right now, the name merely helps to determine which section of the conf file to look in and a couple of other things. | 05:35 |
| cp16net | ah thats possible | 05:35 |
| amrith | like sushilkm's other chkin on datastore specific timeouts. | 05:36 |
| amrith | now, either make that datastore version specific or more meaningfully implement the datastore/version split | 05:36 |
| amrith | like for example, allow the user to provide parameters at the global level; at the datastore level or the datastore version level. use the most specific one preferentially. | 05:37 |
| amrith | that's how I'd like this to have been ... | 05:37 |
| cp16net | yeah i think the default for all leads to a bit more complication | 05:38 |
| amrith | so, I think SlickNik isnt around and we should wait for his input. You know what I'm talking about. How about we wait for him and then continue? | 05:38 |
| cp16net | i think thats fair | 05:38 |
| sushilkm | i think i am not able to grasp some ideas, about what is the new thing being looked into | 05:39 |
| amrith | sushilkm, how about we wait for SlickNik and then one of us can summarize | 05:40 |
| sushilkm | cool | 05:40 |
| amrith | i'd hate to have you spin another patch set | 05:40 |
| amrith | before there's some consensus. | 05:40 |
| cp16net | amrith: after this convo i'm split on the topic | 05:40 |
| amrith | ok, let's unsplit you. | 05:40 |
| *** exploreshaifali has joined #openstack-trove | 05:40 | |
| cp16net | maybe after some sleep i'll wake up on one side of the other of the bed | 05:40 |
| amrith | ok, go take a nap, let's talk tomorrow | 05:41 |
| cp16net | yea | 05:41 |
| amrith | sushilkm, you in your local office or wfh? | 05:41 |
| sushilkm | wfh | 05:41 |
| amrith | k | 05:42 |
| sushilkm | r u in India, amrith | 05:42 |
| amrith | ok all, time for a break. | 05:42 |
| *** bhunter71 has quit IRC | 05:42 | |
| amrith | will talk to y'all later | 05:42 |
| amrith | nope | 05:42 |
| amrith | g'night all | 05:43 |
| cp16net | nite | 05:43 |
| cp16net | amrith: good talk tho | 05:43 |
| sushilkm | gd nyt | 05:43 |
| amrith | cp16net, i think SlickNik is sleep walking | 05:43 |
| cp16net | to the store | 05:43 |
| cp16net | or bar | 05:43 |
| cp16net | -P | 05:43 |
| amrith | ;) | 05:44 |
| amrith | whatever it is, no beer arrived. | 05:44 |
| amrith | just saying. | 05:44 |
| SlickNik | back | 05:44 |
| *** rjaiswal has quit IRC | 05:44 | |
| * amrith looks for the beer | 05:44 | |
| cp16net | oh just in time | 05:44 |
| SlickNik | I have beer here | 05:44 |
| * SlickNik reads scrollback | 05:44 | |
| amrith | cp16net, should we let SlickNik read the scrollback and reconvene tomorrow after you've had a chance to sleep on it. | 05:44 |
| amrith | and he's had a chance to drink the beer that he isn't sharing. | 05:45 |
| amrith | ok, it's late for me I'm calling it quits for now. | 05:45 |
| cp16net | so here is what i am thinking right now around this... | 05:50 |
| cp16net | 1. the api should be modeled around the objects so that it makes sense in a restful way | 05:50 |
| cp16net | 2. if we need a way to make things even easier like this helper that does multiple calls from a single command this might call for a new api call | 05:52 |
| amrith | cp16net, I'll look for this in scrollback later and maybe post you a response. we've typed a bunch so let's also give SlickNik a chance to drink his beer and process what we've been talking about. | 05:52 |
| amrith | bye for now. | 05:52 |
| amrith | will look to read what you've written in a couple of hours. cp16net ^^ cheerios | 05:53 |
| cp16net | 3. calling multiple api call from a single cmd gets the job done but seems wrong | 05:53 |
| cp16net | peace out i'm just putting what my current thoughts are here. | 05:53 |
| cp16net | i'm out | 05:58 |
| *** amrith is now known as _amrith_ | 06:02 | |
| lefred | cp16net: did you find the issue with mysql ? | 06:02 |
| cp16net | not yet | 06:06 |
| cp16net | i'm trying out pxc56 now | 06:07 |
| *** esmute_ has joined #openstack-trove | 06:07 | |
| cp16net | its time for sleep i'm going to pick it back up in the morning | 06:09 |
| lefred | cp16net: ok but I'm 100% sure this is not related to the mysql/pxc version ;) | 06:09 |
| cp16net | thats my thought as well but it shouldnt be hard to update that | 06:10 |
| cp16net | i tryed installing pxc55 on a vagrant box | 06:10 |
| cp16net | and tried to replicate whats happening and i had no luck replicating the weird issue | 06:11 |
| cp16net | i'll poke some more tomorrow though. | 06:11 |
| *** esmute_ has quit IRC | 06:11 | |
| lefred | no prob, have a good nigh | 06:11 |
| lefred | t | 06:11 |
| *** exploreshaifali has quit IRC | 06:15 | |
| *** exploreshaifali has joined #openstack-trove | 06:19 | |
| *** chlong has quit IRC | 06:49 | |
| *** pmackinn has joined #openstack-trove | 06:50 | |
| *** pmackinn has quit IRC | 07:07 | |
| *** ashleighfarnham has joined #openstack-trove | 07:16 | |
| *** ashleighfarnham has quit IRC | 07:21 | |
| *** exploreshaifali has quit IRC | 07:42 | |
| *** exploreshaifali has joined #openstack-trove | 07:56 | |
| *** tosky has joined #openstack-trove | 07:58 | |
| *** mingdang1 has joined #openstack-trove | 08:24 | |
| *** sgotliv has joined #openstack-trove | 08:33 | |
| *** haomaiwang has joined #openstack-trove | 08:34 | |
| *** sgotliv has quit IRC | 08:34 | |
| *** sgotliv has joined #openstack-trove | 08:34 | |
| *** exploreshaifali has quit IRC | 08:38 | |
| *** mingdang1 has quit IRC | 08:52 | |
| *** ashleighfarnham has joined #openstack-trove | 09:05 | |
| *** ashleighfarnham has quit IRC | 09:10 | |
| *** saban has joined #openstack-trove | 09:29 | |
| saban | I have a database image | 09:29 |
| saban | what to do to have trove-guest agent installed on it? | 09:29 |
| saban | do service.py file has to run on that image? | 09:30 |
| *** k4n0 has joined #openstack-trove | 09:31 | |
| *** saban has quit IRC | 09:40 | |
| *** saban has joined #openstack-trove | 09:40 | |
| sushilkm | @saban, you can simply install trove on ur image | 09:42 |
| saban | that's what i am asking | 09:44 |
| saban | how to install trove on image | 09:44 |
| saban | ? | 09:44 |
| sushilkm | just clone it and run setup.py | 09:45 |
| sushilkm | how r u building ur image | 09:45 |
| saban | using DIB | 09:48 |
| sushilkm | ok | 09:48 |
| sushilkm | also you can have a look at trove-integration to find the elements | 09:49 |
| *** exploreshaifali has joined #openstack-trove | 09:50 | |
| sushilkm | also a gud document for building images is in trove | 09:50 |
| sushilkm | https://github.com/openstack/trove/blob/master/doc/source/dev/building_guest_images.rst | 09:50 |
| *** erkules has quit IRC | 09:53 | |
| *** erkules has joined #openstack-trove | 09:53 | |
| *** saban has quit IRC | 09:54 | |
| openstackgerrit | Merged openstack/trove: Switch to the oslo.context library https://review.openstack.org/203829 | 10:05 |
| openstackgerrit | Merged openstack/trove: Switch to the oslo.serialization library https://review.openstack.org/203827 | 10:06 |
| openstackgerrit | Merged openstack/trove: MongoDB prepare needs to wait for Mongo to start https://review.openstack.org/203650 | 10:06 |
| openstackgerrit | Merged openstack/python-troveclient: Updated from global requirements https://review.openstack.org/203145 | 10:07 |
| *** coolsvap is now known as coolsvap|away | 10:27 | |
| *** ashleighfarnham has joined #openstack-trove | 10:54 | |
| *** ashleighfarnham has quit IRC | 10:58 | |
| *** chlong has joined #openstack-trove | 11:08 | |
| *** esmute_ has joined #openstack-trove | 11:12 | |
| *** esmute_ has quit IRC | 11:16 | |
| *** mingdang1 has joined #openstack-trove | 11:28 | |
| *** mingdang1 has quit IRC | 11:28 | |
| *** mingdang1 has joined #openstack-trove | 11:28 | |
| *** daemontool_ has joined #openstack-trove | 11:36 | |
| *** daemontool_ has quit IRC | 11:36 | |
| *** daemontool_ has joined #openstack-trove | 11:37 | |
| *** exploreshaifali has quit IRC | 11:46 | |
| *** tomblank has joined #openstack-trove | 11:59 | |
| *** _amrith_ is now known as amrith | 12:15 | |
| *** pboros has joined #openstack-trove | 12:15 | |
| *** lexholden has joined #openstack-trove | 12:24 | |
| *** ashleighfarnham has joined #openstack-trove | 12:43 | |
| openstackgerrit | Merged openstack/trove: Configuration Groups for Redis https://review.openstack.org/191860 | 12:45 |
| *** ashleighfarnham has quit IRC | 12:47 | |
| *** amrith is now known as _amrith_ | 12:48 | |
| *** _crobertsrh is now known as crobertsrh | 12:48 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/trove: Updated from global requirements https://review.openstack.org/203152 | 12:54 |
| *** k4n0_ has joined #openstack-trove | 13:06 | |
| *** k4n0 has quit IRC | 13:06 | |
| *** _amrith_ is now known as amrith | 13:07 | |
| *** esmute_ has joined #openstack-trove | 13:14 | |
| *** esmute_ has quit IRC | 13:18 | |
| *** k4n0_ has quit IRC | 13:31 | |
| dbennett | cp16net ping me if you need any help with PXC | 13:40 |
| *** exploreshaifali has joined #openstack-trove | 13:48 | |
| *** mingdang1 has quit IRC | 14:14 | |
| *** thedodd has joined #openstack-trove | 14:18 | |
| *** rbrooker has joined #openstack-trove | 14:20 | |
| *** ashleighfarnham has joined #openstack-trove | 14:31 | |
| *** ashleighfarnham has quit IRC | 14:36 | |
| *** sabeen1 has joined #openstack-trove | 14:39 | |
| *** amrith is now known as _amrith_ | 14:45 | |
| *** exploreshaifali has quit IRC | 14:46 | |
| *** georgelorch has quit IRC | 14:51 | |
| *** georgelorch has joined #openstack-trove | 14:59 | |
| *** sabeen3 has joined #openstack-trove | 15:01 | |
| *** sabeen1 has quit IRC | 15:02 | |
| *** rjaiswal has joined #openstack-trove | 15:06 | |
| *** bhunter71 has joined #openstack-trove | 15:07 | |
| *** sriram_tesora has joined #openstack-trove | 15:10 | |
| *** daemontool_ is now known as marzif_ | 15:10 | |
| *** sgotliv has quit IRC | 15:10 | |
| *** david-lyle has joined #openstack-trove | 15:17 | |
| *** marzif_ has quit IRC | 15:28 | |
| *** rbrooker has quit IRC | 15:30 | |
| *** ashleighfarnham has joined #openstack-trove | 15:32 | |
| *** ashleighfarnham has quit IRC | 15:37 | |
| *** rbrooker has joined #openstack-trove | 15:38 | |
| openstackgerrit | Morgan Jones proposed openstack/python-troveclient: Redis Clustering Initial Implementation https://review.openstack.org/203701 | 15:42 |
| *** zhiyan has quit IRC | 15:58 | |
| *** zhiyan has joined #openstack-trove | 15:58 | |
| *** exploreshaifali has joined #openstack-trove | 16:08 | |
| *** rbrooker has quit IRC | 16:12 | |
| *** esmute_ has joined #openstack-trove | 16:13 | |
| *** serverascode has quit IRC | 16:14 | |
| *** esmute_ has quit IRC | 16:14 | |
| *** serverascode has joined #openstack-trove | 16:16 | |
| *** sgotliv has joined #openstack-trove | 16:24 | |
| *** ashleighfarnham has joined #openstack-trove | 16:25 | |
| openstackgerrit | Sergey Vilgelm proposed openstack/trove: Switch to the oslo_log library https://review.openstack.org/203828 | 16:32 |
| *** lexholden has quit IRC | 16:34 | |
| cp16net | dbennett: i tried out pxc56 | 16:37 |
| cp16net | but the test instance size we have of 512MB wont run pxc apparently | 16:38 |
| *** georgelorch has quit IRC | 16:38 | |
| cp16net | it keeps dieing from not enough memory | 16:38 |
| cp16net | lefred: if you are arround ^^ | 16:38 |
| lefred | cp16net: dying at what point ? | 16:40 |
| *** georgelorch has joined #openstack-trove | 16:41 | |
| cp16net | on start | 16:42 |
| lefred | cp16net: in 5.6 (mysql) the buffer pool default is larger (128M) so you should add in your config something like innodb_buffer_pool_size = 10M | 16:42 |
| lefred | and check if it starts | 16:42 |
| cp16net | lefred: https://gist.github.com/cp16net/9dba437e7b357f62a4ba | 16:47 |
| cp16net | i get this sort of errors on start | 16:47 |
| cp16net | mysql/error.log and syslog | 16:48 |
| lefred | with a buffer pool of 10M ? | 16:48 |
| cp16net | yeah with or without | 16:48 |
| lefred | what's the value of key_buffer_size ? | 16:49 |
| lefred | gcache.size = 300M; | 16:49 |
| lefred | you need to change that too | 16:49 |
| dbennett | cp16net, no chance of increasing VM ram? | 16:49 |
| lefred | wsrep_provider_options="gcache.size=16M" | 16:50 |
| cp16net | i could but its our default vm size for testing | 16:50 |
| lefred | because this is mmaped | 16:50 |
| dbennett | ack | 16:50 |
| cp16net | key_buffer is 16M | 16:51 |
| cp16net | its the default install 56 | 16:51 |
| cp16net | so i havnt modified any configs | 16:51 |
| dbennett | you might follow some of the advice here: http://www.tocker.ca/2014/03/10/configuring-mysql-to-use-minimal-memory.html | 16:51 |
| dbennett | don't set max_connections to 1 obviously | 16:52 |
| dbennett | actually, before you do that... | 16:53 |
| *** esmute_ has joined #openstack-trove | 16:54 | |
| *** thedodd has quit IRC | 16:54 | |
| dbennett | cp16net, can you dump your SHOW GLOBAL VARIABLES somewhere? | 16:55 |
| cp16net | mysql reads the last value you set for those right? | 16:55 |
| cp16net | so i can copy and paste that at the bottom and it should be ok as long as its in the [mysqld] section? | 16:56 |
| cp16net | dbennett: how do you do that when its not starting? | 16:56 |
| lefred | cp16net: you should have some changes somewhere, for example the gcache default size is 128M not 300 | 17:00 |
| dbennett | cp16net, good point, how about /usr/sbin/mysqld --verbose --help | 17:03 |
| dbennett | it may take a minute with PXC but it will eventually spit out all of the settings | 17:03 |
| edmondk | 2015-07-23 16:57:59.328 8777 DEBUG trove.common.base_wsgi [-] Empty body provided in request deserialize_body /opt/stack/venvs/trove/local/lib/python2.7/site-packages/trove/common/base_wsgi.py:661 | 17:04 |
| edmondk | 2015-07-23 16:57:59.328 8777 DEBUG trove.common.wsgi [-] Getting schema for type:index get_schema /opt/stack/venvs/trove/local/lib/python2.7/site-packages/trove/common/wsgi.py:373 | 17:04 |
| edmondk | 2015-07-23 16:57:59.329 8777 INFO eventlet.wsgi [-] 127.0.0.1 - - [23/Jul/2015 16:57:59] "GET / HTTP/1.0" 200 281 0.001627 | 17:04 |
| edmondk | 2015-07-23 16:58:00.045 8775 INFO eventlet.wsgi [-] (8775) accepted ('127.0.0.1', 57449) | 17:04 |
| edmondk | 2015-07-23 16:58:00.048 8775 DEBUG trove.common.base_wsgi [-] Empty body provided in request deserialize_body /opt/stack/venvs/trove/local/lib/python2.7/site-packages/trove/common/base_wsgi.py:661 | 17:04 |
| edmondk | 2015-07-23 16:58:00.048 8775 DEBUG trove.common.wsgi [-] Getting schema for type:index get_schema /opt/stack/venvs/trove/local/lib/python2.7/site-packages/trove/common/wsgi.py:373 | 17:04 |
| edmondk | 2015-07-23 16:58:00.050 8775 INFO eventlet.wsgi [-] 127.0.0.1 - - [23/Jul/2015 16:58:00] "GET / HTTP/1.0" 200 281 0.002957 | 17:04 |
| edmondk | 2015-07-23 16:58:00.060 8778 INFO eventlet.wsgi [-] (8778) accepted ('127.0.0.1', 57450) | 17:04 |
| edmondk | 2015-07-23 16:58:00.062 8778 DEBUG trove.common.base_wsgi [-] Empty body provided in request deserialize_body /opt/stack/venvs/trove/local/lib/python2.7/site-packages/trove/common/base_wsgi.py:661 | 17:04 |
| edmondk | 2015-07-23 16:58:00.063 8778 DEBUG trove.common.wsgi [-] Getting schema for type:index get_schema /opt/stack/venvs/trove/local/lib/python2.7/site-packages/trove/common/wsgi.py:373 | 17:04 |
| edmondk | 2015-07-23 16:58:00.065 8778 INFO eventlet.wsgi [-] 127.0.0.1 - - [23/Jul/2015 16:58:00] "GET / HTTP/1.0" 200 281 0.003132 | 17:04 |
| edmondk | whoops meant to paste that in a paste | 17:05 |
| edmondk | Gist doesn't appear to be working | 17:05 |
| cp16net | lol | 17:05 |
| edmondk | I guess my question was has anyone ever seen this message constantly repeated in trove-api | 17:05 |
| edmondk | trove-api is just constantly outputting this and seems not be handling any requests | 17:06 |
| openstackgerrit | Sergey Vilgelm proposed openstack/trove: Remove openstack.common package https://review.openstack.org/203826 | 17:06 |
| cp16net | edmondk: looks like that might be getting the versions | 17:06 |
| *** mingdang1 has joined #openstack-trove | 17:15 | |
| *** mingdang1 has quit IRC | 17:19 | |
| *** coolsvap|away is now known as coolsvap | 17:22 | |
| *** lexholden has joined #openstack-trove | 17:39 | |
| *** lexholden has quit IRC | 17:42 | |
| *** pmackinn has joined #openstack-trove | 17:43 | |
| openstackgerrit | Petr Malik proposed openstack/trove: Configuration Groups for MongoDB https://review.openstack.org/194739 | 18:21 |
| openstackgerrit | Petr Malik proposed openstack/trove: Improve the guestagent configuration manager https://review.openstack.org/198595 | 18:27 |
| *** exploreshaifali has quit IRC | 18:35 | |
| *** sriram_ has joined #openstack-trove | 18:53 | |
| *** pmackinn has quit IRC | 18:53 | |
| *** bitblt has joined #openstack-trove | 18:55 | |
| *** sriram_tesora has quit IRC | 18:55 | |
| *** bitblt has quit IRC | 18:58 | |
| *** annashen has joined #openstack-trove | 18:59 | |
| cp16net | lefred: dunno if you are still around... but i lookedinto the time the mysql table was updated | 19:07 |
| cp16net | the mysq/user.* files were last modified in the same minute that the grant happened | 19:08 |
| cp16net | but looking at the db using root i dont see the grant there | 19:09 |
| *** tomblank has quit IRC | 19:10 | |
| *** pboros has quit IRC | 19:12 | |
| *** pmackinn has joined #openstack-trove | 19:19 | |
| openstackgerrit | Petr Malik proposed openstack/trove: Add generic int-test classes https://review.openstack.org/203274 | 19:23 |
| *** sriram_tesora has joined #openstack-trove | 19:28 | |
| *** mattgriffin has joined #openstack-trove | 19:28 | |
| *** sriram_ has quit IRC | 19:29 | |
| *** annashen has quit IRC | 19:31 | |
| *** annashen has joined #openstack-trove | 19:32 | |
| *** pmackinn has quit IRC | 19:34 | |
| openstackgerrit | Merged openstack/trove: Updated from global requirements https://review.openstack.org/203152 | 19:34 |
| *** esmute_ has quit IRC | 19:37 | |
| *** esmute_ has joined #openstack-trove | 19:39 | |
| *** mattgriffin has quit IRC | 19:43 | |
| *** mattgriffin has joined #openstack-trove | 19:45 | |
| openstackgerrit | Morgan Jones proposed openstack/python-troveclient: Redis Clustering Initial Implementation https://review.openstack.org/203701 | 20:04 |
| openstackgerrit | Morgan Jones proposed openstack/trove: Redis Cluster Initial Implementation https://review.openstack.org/203702 | 20:07 |
| *** john_f_ has joined #openstack-trove | 20:11 | |
| *** bhunter71 has quit IRC | 20:20 | |
| *** lexholden has joined #openstack-trove | 20:25 | |
| *** crobertsrh is now known as _crobertsrh | 20:33 | |
| *** pmackinn has joined #openstack-trove | 20:46 | |
| *** mattgriffin has quit IRC | 20:53 | |
| *** mattgriffin has joined #openstack-trove | 20:57 | |
| *** pmackinn has quit IRC | 21:16 | |
| sabeen3 | Hi All, i'm trying to create a trove instance in devstack and i'm getting the following error - TroveError: Error creating security group for instance: b88b7037-9730-47a4-86b8-17bd2e234767. Any advice on how to fix this will be appreciated! :) | 21:20 |
| *** john_f_ has quit IRC | 21:23 | |
| *** pmackinn has joined #openstack-trove | 21:26 | |
| *** bhunter71 has joined #openstack-trove | 21:50 | |
| *** rjaiswal has quit IRC | 21:54 | |
| *** thedodd has joined #openstack-trove | 21:55 | |
| *** mattgriffin has quit IRC | 21:56 | |
| *** mattgriffin has joined #openstack-trove | 22:03 | |
| *** mattgriffin has quit IRC | 22:03 | |
| sabeen3 | pmackinn: i'm getting an error while trying to create a trove db in devstack - TroveError: Error creating security group for instance: xxxxxxxx | 22:09 |
| sabeen3 | i'm using heat to create the trove db.. i've added use_heat=True in /etc/trove/trove/conf | 22:10 |
| sabeen3 | i see that it shouldn't go into the if statement if use_heat is set to true: https://github.com/openstack/trove/blob/master/trove/taskmanager/models.py#L363 | 22:11 |
| sabeen3 | do you know what could be wrong? | 22:11 |
| *** pmackinn_ has joined #openstack-trove | 22:13 | |
| *** pmackinn has quit IRC | 22:13 | |
| cp16net | sabeen3: sorry i've not run with heat before :-/ | 22:14 |
| cp16net | i'd look in the trove-taskmanager logs and see if there are hints | 22:15 |
| cp16net | it sounds like ther permissions to create a security group might be missing for the user creating the heat stack? | 22:15 |
| cp16net | from what you shared i'm not sure if there is a issue with heat creating the security group or trove | 22:17 |
| cp16net | i dont recall how that works but i *think* its heat that should be creatig the security group from the heat template we have | 22:17 |
| cp16net | my 2 cents hopefully that helps you debugging | 22:18 |
| sabeen3 | cp16net: thank you! | 22:20 |
| sabeen3 | cp16net: i'll check in the trove-taskmanager | 22:20 |
| *** pmackinn_ has quit IRC | 22:24 | |
| *** annashen has quit IRC | 22:25 | |
| *** esmute_ has quit IRC | 22:27 | |
| *** sabeen3 has quit IRC | 22:32 | |
| *** annashen has joined #openstack-trove | 22:32 | |
| *** esmute_ has joined #openstack-trove | 22:33 | |
| *** tosky has quit IRC | 22:43 | |
| *** chlong has quit IRC | 22:52 | |
| *** coolsvap is now known as coolsvap|away | 23:00 | |
| *** david-lyle has quit IRC | 23:02 | |
| *** lexholden has quit IRC | 23:08 | |
| *** ashuk has joined #openstack-trove | 23:10 | |
| *** esmute_ has quit IRC | 23:20 | |
| *** esmute_ has joined #openstack-trove | 23:21 | |
| *** rjaiswal has joined #openstack-trove | 23:23 | |
| *** sriram_tesora has quit IRC | 23:28 | |
| *** annashen has quit IRC | 23:31 | |
| *** david-lyle has joined #openstack-trove | 23:36 | |
| *** mingdang1 has joined #openstack-trove | 23:45 | |
| *** thedodd has quit IRC | 23:50 | |
| *** ashleighfarnham has quit IRC | 23:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!