*** ChanServ changes topic to "Wiki: https://wiki.openstack.org/wiki/Murano | Launchpad: https://launchpad.net/Murano" | 00:01 | |
*** drupalmonkey has quit IRC | 00:13 | |
*** rakhmerov has joined #murano | 00:18 | |
*** rakhmerov has quit IRC | 00:22 | |
*** asalkeld has quit IRC | 00:24 | |
*** asalkeld has joined #murano | 00:37 | |
*** rakhmerov has joined #murano | 01:18 | |
*** rakhmerov has quit IRC | 01:23 | |
*** julienvey has joined #murano | 01:46 | |
*** julienvey has quit IRC | 01:51 | |
*** julienvey has joined #murano | 02:45 | |
*** julienvey has quit IRC | 02:50 | |
*** sergmelikyan has quit IRC | 03:10 | |
*** sergmelikyan has joined #murano | 03:13 | |
*** rakhmerov has joined #murano | 03:20 | |
*** rakhmerov has quit IRC | 03:24 | |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-dashboard: Fix DynamicSelect choice field https://review.openstack.org/87836 | 03:26 |
---|---|---|
*** julienvey has joined #murano | 03:46 | |
*** julienvey has quit IRC | 03:51 | |
*** IlyaE has quit IRC | 04:03 | |
*** rakhmerov has joined #murano | 04:05 | |
*** asalkeld has quit IRC | 04:14 | |
*** IlyaE has joined #murano | 04:17 | |
*** sergmelikyan has quit IRC | 04:21 | |
*** sergmelikyan has joined #murano | 04:23 | |
*** saju_m has joined #murano | 04:27 | |
*** saju_m has quit IRC | 04:27 | |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-dashboard: Add generateHostname function https://review.openstack.org/87851 | 04:46 |
*** julienvey has joined #murano | 04:47 | |
*** julienvey has quit IRC | 04:52 | |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-api: Fix issue with hostname https://review.openstack.org/87852 | 04:59 |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-api: Handle unicode strings in merge-dict https://review.openstack.org/87853 | 05:00 |
*** IlyaE has quit IRC | 05:07 | |
*** sab has joined #murano | 05:10 | |
*** chandan_kumar has joined #murano | 05:31 | |
sergmelikyan | http://andrewbrookins.com/tech/one-year-later-an-epic-review-of-pycharm-2-7-from-a-vim-users-perspective/ | 05:37 |
sergmelikyan | Interesting post :) | 05:37 |
*** julienvey has joined #murano | 05:51 | |
*** IlyaE has joined #murano | 05:54 | |
*** julienvey has quit IRC | 05:55 | |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-api: Fix engine results processing https://review.openstack.org/87857 | 06:21 |
sergmelikyan | tsufiev, I need your help to squash bug #1305142 | 06:30 |
sergmelikyan | It seems like dashboard store app_id as additional param for instance of the application. Why? | 06:31 |
sab | there might be conflicts and dependencies for a services running on different nodes. will murano take care of such cases? i know the question is blunt. but just wanted to know if murano is handling conflicts or dependencies anywhere | 06:31 |
sergmelikyan | sab, about which conflicts you are talking about? And what does it mean 'services running on different nodes' | 06:38 |
* sergmelikyan didn't get question and need some more information | 06:39 | |
sab | sergmelikyan, Suppose service A is dependent on service B which may be in same or different nodes. while deploying, is there a way to speicfy in manifest that service A is dependent on service B, so service B should be deployed first and service A next? | 06:43 |
sab | sergmelikyan, i am still trying to get into murano, apologies if question is too stupid :) | 06:44 |
sergmelikyan | There is no such option for now, every application gets deploy command in the same time. (so all applications are deployed in parallel) | 06:49 |
sab | sergmelikyan, thank you :) | 06:52 |
*** IlyaE has quit IRC | 06:55 | |
sergmelikyan | sab, I am not sure that I have answered on your question: let's start from beginning and consider actual two application (use-case that I am working on right now) | 06:59 |
sergmelikyan | PostgreSQL and Tomcat | 06:59 |
sergmelikyan | Tomcat - is a application that deploys Apache Tomcat and installs any Servlet application on top of Tomcat from git | 07:00 |
sergmelikyan | After servlet app is deployed it's need to be pointed to the PostgreSQL, so during Tomcat deployment we need to ask some connection string from Postgres (that may deployed right now, along side with Tomcat, or no) | 07:02 |
*** chandan_kumar has quit IRC | 07:02 | |
sergmelikyan | We can ask connection string from PostgreSQL only if it is finished own deployment, so we in Deploy action of Tomcat Application call something like: $database.deploy() and right after that call $cs = $database.getConnectionString() | 07:03 |
sergmelikyan | At this moment we don't actually know is $database (link to PostgreSQL application instance) deployed or not | 07:04 |
sergmelikyan | If it is not deployed yet: deployment is triggered | 07:04 |
sergmelikyan | If it deploying right now, this call to deploy one more time will be superseded by previous one | 07:05 |
sergmelikyan | If it is deployed - deploy will be called once more time, and will start from scratch | 07:05 |
sergmelikyan | So one major assumption is that all actions are idempotent or at least should be | 07:07 |
*** slagun has quit IRC | 07:07 | |
sergmelikyan | We have Attributes mechanism to support implementation of idempotent actions | 07:07 |
*** chandan_kumar has joined #murano | 07:08 | |
sergmelikyan | Since you called deploy from Tomcat you may be sure that PostgreSQL is deployed, after that you call some $cs = $database.getConnectionString to obtain CS and pass connection string obtained from PostgreSQL to configureApplicationWithConnectionString($cs) in TomCat | 07:12 |
sab | sergmelikyan, so basically the dependent applications are installed while the main application is being deployed. am i right? | 07:14 |
sergmelikyan | sab, there is no clean definition of dependency in Murano. | 07:15 |
sergmelikyan | sab, you just have some object hierarchy and each object can call any action on another one or get some attribute from other one. | 07:16 |
sergmelikyan | One convention: engine call action deploy() on all objects that extends Application class | 07:17 |
sergmelikyan | If you need to call some action and be sure that application is deployed - you call deploy once more time. | 07:18 |
sab | meaning tomcat needs postgresql, while deploying tomcat we install postgresql. can we make it explicit in the manifest file that tomcat needs postgresql, so that murano knows what to deploy first? it can be a proposal :) | 07:18 |
sergmelikyan | sab, there is no reason to do this explicitly in manifest, since you can't be sure that there is need to have another application deployed. | 07:20 |
sergmelikyan | For example you can get in Tomcat Application username param from PostgreSQL without even deploying PostgreSQL | 07:20 |
sergmelikyan | $postgres_uname = $database.username | 07:21 |
sergmelikyan | since username is supplied by the used in UI, this param is available from begging for all other objects that reference PostgreSQL | 07:21 |
sergmelikyan | *by the user | 07:22 |
sergmelikyan | *begging = beginning | 07:22 |
* sergmelikyan is typo master | 07:22 | |
sab | hmmm, yep but here it's only tomcat and postgres.. I thought it might be cleaner if there are many applications like a heirachical definition. agreed that is not important, may be desired :) | 07:26 |
*** rakhmero_ has joined #murano | 07:27 | |
sergmelikyan | sab, in many application is even more important to NOT have such dependencies stated in manifest - you will end up with some huge list of dependencies, and be sure with circular ones. | 07:29 |
sergmelikyan | And not for all them there is a reason to call deploy. Do not forget that deploy could take much time. | 07:29 |
sergmelikyan | And one other thing - not all dependencies are dependencies to Applications. | 07:30 |
sergmelikyan | Object reference any other objects, and only some of them are Applications | 07:30 |
*** rakhmerov has quit IRC | 07:30 | |
sergmelikyan | sab, take a look on Tomcat example that I had talked about: http://paste.openstack.org/show/75877/ | 07:31 |
sab | sergmelikyan, hmm.. yeah i am having a look | 07:33 |
openstackgerrit | Dmitry Teselkin proposed a change to stackforge/murano-api: Add murano-dashboard to devstack https://review.openstack.org/87751 | 08:06 |
*** chandan_kumar has quit IRC | 09:06 | |
*** chandan_kumar has joined #murano | 09:20 | |
*** chandan_kumar has quit IRC | 09:26 | |
*** chandan_kumar has joined #murano | 09:26 | |
*** rakhmerov has joined #murano | 10:28 | |
*** rakhmero_ has quit IRC | 10:30 | |
openstackgerrit | Dmitry Teselkin proposed a change to stackforge/murano-api: Get rid of murano-common https://review.openstack.org/87921 | 10:36 |
openstackgerrit | Dmitry Teselkin proposed a change to stackforge/murano-api: Fix some options set by devstack https://review.openstack.org/87929 | 10:47 |
openstackgerrit | A change was merged to stackforge/murano-api: Add tag assignment during manifest parsing https://review.openstack.org/87581 | 11:46 |
*** sab has quit IRC | 12:08 | |
openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-api: Setup doc build infrastructure https://review.openstack.org/87950 | 12:23 |
openstackgerrit | A change was merged to stackforge/murano-api: Devstack scripts update https://review.openstack.org/87690 | 12:58 |
openstackgerrit | A change was merged to stackforge/murano-api: Add murano-dashboard to devstack https://review.openstack.org/87751 | 12:59 |
openstackgerrit | A change was merged to stackforge/murano-api: Get rid of murano-common https://review.openstack.org/87921 | 12:59 |
openstackgerrit | A change was merged to stackforge/murano-api: Fix some options set by devstack https://review.openstack.org/87929 | 12:59 |
*** rakhmero_ has joined #murano | 13:53 | |
*** rakhmerov has quit IRC | 13:55 | |
*** julienvey has joined #murano | 13:59 | |
*** drupalmonkey has joined #murano | 14:02 | |
*** sjmc7 has joined #murano | 14:05 | |
sjmc7 | good afternoon, akuznetsova. were you able to reproduce https://bugs.launchpad.net/murano/+bug/1308247 ? | 14:05 |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-dashboard: Add generateHostname function https://review.openstack.org/87851 | 14:07 |
akuznetsova | sjmc7, hello, actually I didn't work with this bug yet, but in my environment I don't see this problem(Chrome, Firefox), later I will test it in IE | 14:08 |
sjmc7 | it may be specific to the application description or image. the data i was using was faked; i'll spend a few minutes trying to narrow down the cause | 14:09 |
ruhe | sjmc7: hi! do you have the latest version of dashboard code? | 14:13 |
sjmc7 | yep, ruhe | 14:13 |
sjmc7 | i updated again after georgy had trouble reproducing it | 14:13 |
ruhe | tsufiev: any idea why this bug might happen? | 14:13 |
*** rakhmerov has joined #murano | 14:14 | |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-api: Fix issue with hostname https://review.openstack.org/87852 | 14:14 |
*** rakhmerov has quit IRC | 14:16 | |
*** rakhmero_ has quit IRC | 14:17 | |
openstackgerrit | A change was merged to stackforge/murano-dashboard: Fix DynamicSelect choice field https://review.openstack.org/87836 | 14:23 |
*** rakhmerov has joined #murano | 14:24 | |
openstackgerrit | A change was merged to stackforge/murano-api: Fix engine results processing https://review.openstack.org/87857 | 14:26 |
*** julienvey has quit IRC | 14:31 | |
*** IlyaE has joined #murano | 15:03 | |
*** julienvey has joined #murano | 15:05 | |
*** rakhmerov has quit IRC | 15:12 | |
openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-conductor: Depracate murano-conductor https://review.openstack.org/87995 | 15:17 |
*** rakhmerov has joined #murano | 15:20 | |
openstackgerrit | Stan Lagun proposed a change to stackforge/murano-api: Default MuranoPL function argument value was evaluated incorrectly https://review.openstack.org/87997 | 15:24 |
*** rakhmerov has quit IRC | 15:25 | |
openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-deployment: Deprecate Devstack scripts https://review.openstack.org/88001 | 15:28 |
*** rakhmerov has joined #murano | 15:31 | |
openstackgerrit | Julien Vey proposed a change to stackforge/murano-api: Add 'cp' of murano-dashboard lib file in devstack README https://review.openstack.org/88003 | 15:35 |
*** rakhmerov has quit IRC | 15:36 | |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-dashboard: Add generateHostname function https://review.openstack.org/87851 | 15:37 |
openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-common: Deprecate murano-common https://review.openstack.org/88006 | 15:38 |
openstackgerrit | A change was merged to stackforge/murano-api: Setup doc build infrastructure https://review.openstack.org/87950 | 15:40 |
sjmc7 | ruhe - you're on a deprecating roll today :) | 15:41 |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-dashboard: Add generateHostname function https://review.openstack.org/87851 | 15:43 |
*** chandan_kumar has quit IRC | 15:46 | |
openstackgerrit | Timur Sufiev proposed a change to stackforge/murano-api: Replace 'applications' in deployments data with 'services' https://review.openstack.org/88008 | 15:46 |
openstackgerrit | Stan Lagun proposed a change to stackforge/murano-api: Basic Security Groups implementation https://review.openstack.org/88013 | 15:59 |
*** rakhmerov has joined #murano | 16:00 | |
*** IlyaE has quit IRC | 16:20 | |
ruhe | sjmc7: more to go :) there was a thread in ML about devstack scripts for murano. newcomers get confused by all this repos | 16:21 |
sjmc7 | my team all got confused by it :) | 16:21 |
*** rakhmerov has quit IRC | 16:27 | |
ruhe | i hope we will just move all those deprecated repos to openstack-attic right after release | 16:28 |
ruhe | and have enough developer documentation before the release https://blueprints.launchpad.net/murano/+spec/murano-dev-doc-05 | 16:29 |
*** IlyaE has joined #murano | 16:30 | |
julienvey | Hi guys! I'm starting to play with murano. Is there a sample catalog I could build and import to test ? any link on the doc that could help me ? | 16:37 |
sergmelikyan | julienvey, there are two packages for now to play with: https://github.com/stackforge/murano-api/tree/master/meta | 16:37 |
sjmc7 | sergmelikyan - did the import code get merged yet? | 16:38 |
sergmelikyan | sjmc7, julienvey not yet: [8:37:44 PM] Kate Fedorova: у меня опять та же ошибка | 16:39 |
sergmelikyan | sorry wrong quote | 16:39 |
sjmc7 | i'll take your word for it :) | 16:39 |
sergmelikyan | https://review.openstack.org/86039 | 16:39 |
sergmelikyan | ^^ link to the fix for package import | 16:39 |
julienvey | sergmelikyan: ok, thanks for the link | 16:39 |
sergmelikyan | tsufiev, could you explore more on comment https://review.openstack.org/#/c/87851/4/muranodashboard/dynamic_ui/yaql_functions.py | 16:44 |
openstackgerrit | A change was merged to stackforge/murano-api: Fix issue with hostname https://review.openstack.org/87852 | 16:45 |
sergmelikyan | Got it... you writing that logic of the method is not clear without comments (not docstrings) | 16:45 |
*** gokrokve has joined #murano | 16:47 | |
openstackgerrit | Serg Melikyan proposed a change to stackforge/murano-dashboard: Add generateHostname function https://review.openstack.org/87851 | 16:52 |
sergmelikyan | tsufiev, updated ^^ | 16:52 |
*** julienvey has quit IRC | 17:02 | |
openstackgerrit | A change was merged to stackforge/murano-api: Replace 'applications' in deployments data with 'services' https://review.openstack.org/88008 | 17:08 |
openstackgerrit | A change was merged to stackforge/murano-api: Handle unicode strings in merge-dict https://review.openstack.org/87853 | 17:10 |
openstackgerrit | A change was merged to stackforge/murano-tests: Refactor WebUI tests https://review.openstack.org/86297 | 17:11 |
*** IlyaE has quit IRC | 17:26 | |
*** ankurrr has joined #murano | 17:30 | |
openstackgerrit | A change was merged to stackforge/murano-api: Add 'cp' of murano-dashboard lib file in devstack README https://review.openstack.org/88003 | 17:35 |
*** rakhmerov has joined #murano | 17:50 | |
*** gokrokve has quit IRC | 17:57 | |
*** rakhmerov has quit IRC | 17:57 | |
openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-metadataclient: Deprecate murano-metadataclient https://review.openstack.org/88041 | 18:02 |
openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-repository: Depreacte murano-repository https://review.openstack.org/88045 | 18:11 |
openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-repository: Deprecate murano-repository https://review.openstack.org/88045 | 18:12 |
ruhe | for those who might be curious - here is a patch to infra which disables all the jenkins jobs repositories i'm deprecating - https://review.openstack.org/#/c/88032/ | 18:13 |
ruhe | and i'm going to send first group of patches for murano developer docs | 18:14 |
*** gokrokve has joined #murano | 18:18 | |
*** tsufiev_ has joined #murano | 18:26 | |
tsufiev_ | guys, i have some thoughts about fully qualified names (FQNs) of package and main package class | 18:27 |
tsufiev_ | if we are to keep only service FQN in object model of Environment being deployed by engine - to later obtain Application name using that FQN, we should make Package FQN equal to the FQN of main package's class | 18:29 |
tsufiev_ | consider this line: https://github.com/stackforge/murano-api/blob/master/meta/io.murano.windows.ActiveDirectory/UI/ui.yaml#L28 | 18:31 |
tsufiev_ | regardless of what should be the proper FQN for Active Directory Service, that FQN will be later used to retrieve human readable type of the Service (i.e. name of its Application) | 18:32 |
tsufiev_ | so, that 2 names should be the same | 18:32 |
tsufiev_ | here 2nd question arises: should that Active Directory's FQN be equal to the FQN of main Active Directory class here https://github.com/stackforge/murano-api/blob/master/meta/io.murano.windows.ActiveDirectory/manifest.yaml#L20 ? | 18:34 |
tsufiev_ | also i'd like to point out that the relation between Application (class of some service) and Service (instance of that class) has coined on its own | 18:36 |
tsufiev_ | are we ok with it or should replace one of these terms with something more suitable? | 18:37 |
sjmc7 | tsufiev_: reading | 18:45 |
sjmc7 | is the 'type' then redundant? | 18:46 |
*** chandan_kumar has joined #murano | 18:49 | |
sjmc7 | tsufiev_: it seems a package's identifier should be independent from what it refers to | 18:50 |
sjmc7 | i think it's confusing having package and class identifiers (fqns) look similar since they're logically quite different | 18:50 |
*** rakhmerov has joined #murano | 18:58 | |
*** rakhmerov has quit IRC | 19:12 | |
gokrokve | Hi. I want to share my idea about conditional resource creation and I want to understand how it is possible to implement with current DSL. | 19:14 |
gokrokve | The idea is the following: | 19:15 |
gokrokve | I have an application which asks in UI if user wants to use standard instance or Heat HARestarter instance | 19:15 |
gokrokve | If user does not select HA option, application definition creates a standard instance type like murano.instance | 19:16 |
gokrokve | if user selects HA option, application definition uses murano.HAinstance which uses Heat HArestarter with alarms. | 19:17 |
gokrokve | The question is how to do this in current DSL? | 19:17 |
gokrokve | currently application definition sets an instance in application properties, but is it possible to create a new instance inside deploy() method? | 19:18 |
*** chandan_kumar has quit IRC | 19:30 | |
tsufiev_ | sjmc7: yes, it is confusing because we have 'Library' packages which don't have 'main' class | 19:34 |
tsufiev_ | sjmc7: i think we still need a way to somehow display human-readable name of class of deployed Service (i.e. Application name) after deploy is finished | 19:35 |
tsufiev_ | it seems to me that problem exists at junction of dashboard, api and engine, so it's not obvious how it is better to solve it | 19:38 |
tsufiev_ | sjmc7: in my opinion type is not redundant (because we want to show it in UI) | 19:40 |
tsufiev_ | what confuses me more is the lack of good candidate for 'an instance of an Application being configured/deployed'. i used the term 'Service', but it comes mostly from Windows' part of Murano history, so i doubt we should use it | 19:45 |
tsufiev_ | using term 'Application' for both cases (entity in AppCatalog and entity in the Environment) surely will confuse the users | 19:46 |
sjmc7 | 'Solution', tsufiev_ ? | 19:55 |
sjmc7 | "Library" packages are classes or collections of classes, right? | 19:56 |
sjmc7 | Or can they also be installable things? | 19:56 |
tsufiev_ | sjmc7: thank you, an interesting idea! should ponder over it... | 19:56 |
tsufiev_ | 'Library' are collections of classes | 19:57 |
openstackgerrit | Ankur Rishi proposed a change to stackforge/murano-api: The previous commit, 3362ee1337d47ab8930407c58311a811d471a7a4, a sync with Openstack Common, removed all definitions of the post_process() function from the Distro class and all of its child classes in install_venv_common.py. https://review.openstack.org/88074 | 19:57 |
*** chandan_kumar has joined #murano | 19:57 | |
tsufiev_ | https://github.com/stackforge/murano-api/tree/master/meta/io.murano an example of Library package (unpacked) | 19:57 |
sjmc7 | Yep, i've seen that one | 19:58 |
sjmc7 | That makes sense, that's a good word i think | 19:58 |
sjmc7 | A 'service' is a deployed collection of 'Applications' | 19:58 |
*** IlyaE has joined #murano | 19:58 | |
openstackgerrit | Ankur Rishi proposed a change to stackforge/murano-api: Fix missing function error in install_venv https://review.openstack.org/88074 | 20:01 |
*** rakhmerov has joined #murano | 20:08 | |
* tsufiev_ went away | 20:09 | |
*** tsufiev_ has quit IRC | 20:09 | |
*** rakhmerov has quit IRC | 20:12 | |
openstackgerrit | A change was merged to stackforge/python-muranoclient: Update requirements.txt due to changes in global-requirements.txt https://review.openstack.org/87548 | 20:28 |
openstackgerrit | A change was merged to stackforge/murano-api: Typo in murano-api.conf https://review.openstack.org/87774 | 20:32 |
openstackgerrit | A change was merged to stackforge/murano-api: Update requirements.txt due to changes in global-requirements.txt https://review.openstack.org/87549 | 20:32 |
*** gokrokve has quit IRC | 20:42 | |
*** asalkeld has joined #murano | 20:57 | |
*** gokrokve has joined #murano | 21:02 | |
*** rakhmerov has joined #murano | 21:09 | |
*** rakhmerov has quit IRC | 21:13 | |
*** rakhmerov has joined #murano | 21:29 | |
*** rakhmerov has quit IRC | 21:33 | |
*** drupalmonkey has quit IRC | 22:27 | |
*** rakhmerov has joined #murano | 22:29 | |
*** sjmc7 has quit IRC | 22:34 | |
*** rakhmerov has quit IRC | 22:34 | |
*** IlyaE has quit IRC | 22:44 | |
*** IlyaE has joined #murano | 23:06 | |
*** killer_prince has quit IRC | 23:14 | |
*** ankurrr has quit IRC | 23:20 | |
*** lazy_prince has joined #murano | 23:20 | |
*** lazy_prince is now known as killer_prince | 23:20 | |
*** gokrokve has quit IRC | 23:23 | |
*** asalkeld_ has joined #murano | 23:30 | |
*** rakhmerov has joined #murano | 23:30 | |
*** asalkeld has quit IRC | 23:30 | |
*** rakhmerov has quit IRC | 23:35 | |
*** chandan_kumar has quit IRC | 23:37 | |
*** asalkeld_ is now known as asalkeld | 23:43 | |
openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-api: Added developer installation guide https://review.openstack.org/88107 | 23:51 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!