Perl 6 - the future is here, just unevenly distributed

IRC log for #openstack-rally, 2016-06-29

| Channels | #openstack-rally index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:03 diogogmt joined #openstack-rally
00:10 yingjun joined #openstack-rally
00:21 diogogmt joined #openstack-rally
00:52 klindgren joined #openstack-rally
01:23 openstack joined #openstack-rally
01:33 openstackgerrit Merged openstack/rally: [Plugins] Add cleanup context for security groups scenarios  https://review.openstack.org/334277
01:48 ilbot3 joined #openstack-rally
01:48 Topic for #openstack-rally is now ☁ Rally RoadMap: http://goo.gl/JZkmwY ☁ Review Dashboard: http://goo.gl/YuCJfl ☁ IRC logs http://irclog.perlgeek.de/openstack-rally ☁ Key persons to ask:  boris-42, yfried, msdubov, rediskin, andreykurilin, amaretskiy  ☁ Documentation: https://rally.readthedocs.org/en/latest/ ☁ To publish changes to Rally:  https://rally.readthedocs.org/en/latest/contribute.html
01:50 yanyanhu joined #openstack-rally
01:57 liuyulong joined #openstack-rally
02:32 adiantum joined #openstack-rally
02:32 raddaoui joined #openstack-rally
02:42 openstackgerrit lvfm proposed openstack/rally: Fix bug on image_valid_on_flavor  https://review.openstack.org/334741
02:43 arnoldje joined #openstack-rally
03:04 liuyulong_ joined #openstack-rally
03:32 yamamot__ joined #openstack-rally
04:03 MaxPC joined #openstack-rally
04:06 yingjun joined #openstack-rally
04:39 yingjun joined #openstack-rally
04:53 rcernin joined #openstack-rally
05:11 yingjun joined #openstack-rally
05:38 yamamot__ joined #openstack-rally
05:47 psuriset joined #openstack-rally
05:51 yingjun joined #openstack-rally
05:58 kuladeep joined #openstack-rally
06:03 rdas joined #openstack-rally
06:05 rcernin joined #openstack-rally
06:08 M00nr41n joined #openstack-rally
06:10 arcimboldo joined #openstack-rally
06:16 pcaruana joined #openstack-rally
06:17 abalutoiu_ joined #openstack-rally
06:22 M00nr41n joined #openstack-rally
06:31 psuriset joined #openstack-rally
06:34 M00nr41n joined #openstack-rally
07:09 tesseract- joined #openstack-rally
07:18 amaretskiy joined #openstack-rally
07:22 openstackgerrit lvfm proposed openstack/rally: Fix bug on image_valid_on_flavor  https://review.openstack.org/334741
07:22 arxcruz joined #openstack-rally
07:25 abehl joined #openstack-rally
07:25 e0ne joined #openstack-rally
07:43 arcimboldo joined #openstack-rally
08:02 mvk joined #openstack-rally
08:05 alexchadin joined #openstack-rally
08:29 klindgren joined #openstack-rally
08:31 kbaikov joined #openstack-rally
08:33 arcimboldo joined #openstack-rally
08:36 sshnaidm joined #openstack-rally
08:39 kbaikov joined #openstack-rally
08:58 kbaikov joined #openstack-rally
09:06 kbaikov joined #openstack-rally
09:15 alexchadin joined #openstack-rally
09:36 mvk joined #openstack-rally
09:42 alexchadin joined #openstack-rally
09:48 alexchadin amaretskiy: hi :)
09:48 tosky joined #openstack-rally
09:50 amaretskiy alexchadin: hi
09:51 alexchadin About your comment "Is it normal that one specific non-admin user creates audit templates for all users?"
09:51 alexchadin https://review.openstack.org/#/c/333858/1/rally/plugins/openstack/context/watcher/audit_templates.py
09:51 alexchadin It seems that admin should crete audit templates
09:51 alexchadin create*
09:52 alexchadin How should I specify that only admin can create it??
09:58 e0ne joined #openstack-rally
10:08 adiantum joined #openstack-rally
10:12 abehl joined #openstack-rally
10:25 amaretskiy alexchadin: sorry for my being out
10:27 amaretskiy alexchadin: it is simple to use admin user in contexts, for example: https://github.com/openstack/rally/blob/master/rally/plugins/openstack/context/nova/flavors.py#L78
10:29 amaretskiy alexchadin: I've added a comment with this note
11:04 alexchadin joined #openstack-rally
11:10 alexchadin joined #openstack-rally
11:11 liuyulong joined #openstack-rally
11:32 liuyulong joined #openstack-rally
11:37 openstackgerrit Alex Krzos proposed openstack/rally: Add keystone validate token using ceilometer scenario  https://review.openstack.org/335245
11:41 kuladeep joined #openstack-rally
11:43 stpierre joined #openstack-rally
11:45 alexchadin joined #openstack-rally
11:57 yamamoto joined #openstack-rally
11:59 alexchadin joined #openstack-rally
12:09 e0ne joined #openstack-rally
12:21 openstackgerrit Merged openstack/rally: Fix bug on image_valid_on_flavor  https://review.openstack.org/334741
12:31 yingjun joined #openstack-rally
12:35 liuyulong_ joined #openstack-rally
12:36 yingjun_ joined #openstack-rally
12:38 yingjun joined #openstack-rally
12:40 aernhart joined #openstack-rally
12:51 alexchadin joined #openstack-rally
12:52 diogogmt joined #openstack-rally
13:02 yingjun joined #openstack-rally
13:02 rook_ joined #openstack-rally
13:11 yingjun joined #openstack-rally
13:11 abehl joined #openstack-rally
13:12 rook_ joined #openstack-rally
13:13 stpierre_ joined #openstack-rally
13:18 kuladeep_ joined #openstack-rally
13:18 stpierre joined #openstack-rally
13:19 kuladeep__ joined #openstack-rally
13:22 openstackgerrit Merged openstack/rally: Add NoverServers.boot_server_from_volume_snapshot  https://review.openstack.org/330291
13:27 yingjun joined #openstack-rally
13:38 yingjun joined #openstack-rally
13:44 alexchadin joined #openstack-rally
13:46 openstackgerrit Chris St. Pierre proposed openstack/rally: Convert ResourceType.transform to ResourceType.preprocess  https://review.openstack.org/317674
13:46 openstackgerrit Chris St. Pierre proposed openstack/rally: Test: Fail jobs if resources are added or removed  https://review.openstack.org/300159
13:46 openstackgerrit Chris St. Pierre proposed openstack/rally: Make glance v2 the default  https://review.openstack.org/307415
13:46 openstackgerrit Chris St. Pierre proposed openstack/rally: Add ability to specify superclass for cleanup  https://review.openstack.org/275340
13:48 abalutoiu hello, does anyone have any idea why I get the actions in random order in the results? For example I get "nova.delete_floating_ip" before "nova.creat_floating_ip" and so on
13:49 stpierre because dicts are inherintly unordered
13:50 stpierre or did you mean the HTML report?
13:50 stpierre in the html report they should be in alphabetical order, but they are run in a random-looking order and will be in a random-looking order in the JSON results
13:51 abalutoiu they are the same in the html report as well
13:54 raddaoui joined #openstack-rally
13:58 arcimboldo joined #openstack-rally
14:00 ajo_ joined #openstack-rally
14:00 rook- joined #openstack-rally
14:04 ajo_ joined #openstack-rally
14:04 yingjun joined #openstack-rally
14:05 openstackgerrit Andrey Kurilin proposed openstack/rally: Tie endpoint_type to interface for keystone  https://review.openstack.org/320056
14:07 ajo_ joined #openstack-rally
14:19 sai joined #openstack-rally
14:29 yingjun joined #openstack-rally
14:33 arcimboldo joined #openstack-rally
14:40 arnoldje joined #openstack-rally
14:43 yamamoto joined #openstack-rally
14:50 yingjun joined #openstack-rally
14:55 KevinE joined #openstack-rally
14:57 tosky joined #openstack-rally
15:00 amaretskiy rook ping
15:00 amaretskiy rook- ping
15:01 rook- heya
15:02 amaretskiy rook-: andreykurilin__ said you are interested about RaaS
15:02 amaretskiy rook-: I'm currently working on this feature, and some demo code is expected to be proposed next week
15:04 rook- sweet
15:08 alan__ joined #openstack-rally
15:16 catintheroof joined #openstack-rally
15:17 openstackgerrit Roman Vasilets proposed openstack/rally: [WIP] Use keystoneauth instead of keystoneclient directly  https://review.openstack.org/335555
15:30 rcernin joined #openstack-rally
15:32 pcaruana joined #openstack-rally
15:41 kuladeep joined #openstack-rally
15:41 kuladeep__ joined #openstack-rally
15:41 kuladeep_ joined #openstack-rally
15:44 openstackgerrit Roman Vasilets proposed openstack/rally: [WIP] Use keystoneauth instead of keystoneclient directly  https://review.openstack.org/335555
15:48 klindgren joined #openstack-rally
15:52 klindgren_ joined #openstack-rally
15:53 alan__ joined #openstack-rally
15:54 alan__ joined #openstack-rally
15:55 diogogmt joined #openstack-rally
15:59 alan__ joined #openstack-rally
16:12 KevinE stpierre: Hey, is your superclass fix tied to https://blueprints.launchpad.net/rally/+spec/add-cleanup-option too?
16:13 stpierre i think that blueprint is probably obsoleted by the cleanup-refactoring spec
16:14 stpierre or if not obsoleted, then certainly basically abandoned
16:16 KevinE stpierre: Okay I was just asked to look into this so I'll just redirect interest to your blueprint instead
16:16 stpierre my blueprint doesn't really cover the same thing, so this is kind of a grey area
16:17 stpierre i think it's unlikely that we'd want to go around mucking with cleanup options before cleanup-refactoring is complete, though
16:17 stpierre i don't think wataru is active any more -- i took over the cleanup-refactoring spec from him
16:17 stpierre so if this is something you're interested in, then you could propose add-cleanup-option as a spec, taking the comments into account, and note that it depends on the cleanup-refactoring spec
16:19 KevinE stpierre: yeah I think his whole group went away from rally. And yeah today is the first day I'm looking at this so I need to investigate what this cleanup-refactoring spec even is
16:22 andreykurilin joined #openstack-rally
16:24 openstackgerrit Chris St. Pierre proposed openstack/rally: [WIP] Test: Fail jobs if resources are added or removed  https://review.openstack.org/300159
16:26 KevinE andreykurilin__: Is it over? :P is this patch actually going to get merged?
16:37 andreykurilin KevinE: yeah. it will be merge soon
16:39 KevinE stpierre: wouldn't the cleaner option be to add that to the current cleanup-refactoring spec? Or do you just not want it in there.
16:39 KevinE stpierre: also I can't really tell from the review, how's it going?
16:39 KevinE andreykurilin: :D
16:39 stpierre i'd just rather not add scope creep to that spec, as far along as it is
16:39 stpierre slowly
16:51 andreykurilin stpierre: I'm happy to see you activity at rally again:)
16:51 stpierre i was just on vacation for a couple of weeks
17:09 abehl joined #openstack-rally
17:21 pvaneck joined #openstack-rally
17:27 abalutoiu joined #openstack-rally
17:44 rcernin joined #openstack-rally
17:50 yingjun joined #openstack-rally
18:00 mvk joined #openstack-rally
18:20 abehl joined #openstack-rally
19:06 cu5 joined #openstack-rally
19:08 yamamoto joined #openstack-rally
19:58 mgagne joined #openstack-rally
20:23 KevinE stpierre: Hey do you know why Nate Johnston commented on this patch saying it didn't work, when he didn't even include the keystone_endpoint_type in his config?
20:23 stpierre nope, i'd ask him :)
20:24 stpierre i didn't include it in mine either, and it worked
20:24 njohnston Oh, I thought it was in there.  I guess trying to set it when creating the deployment did not work.  Apologies for my sloppy reading there.
20:25 njohnston KevinE: How do you set the keystone_endpoint_type when creating a deployment?  I am running against an icehouse keystone that is centralized, not local to my region.
20:28 arcimboldo joined #openstack-rally
20:28 KevinE njohnston: sorry I looked for a matching username but didn't see you lol. and my env has been having issues so I couldn't test it myself, I suppose I was just wondering if that was your "rally deployment config" or your literally deployment config
20:29 njohnston yes, that is what I get when I run `rally deployment config` :-)
20:32 njohnston How do you set keystone_endpoint_type - edit the DB, or do you set it on deployment creation somehow?
20:32 KevinE njohnston: I assumed you set it on deployment creation, like: https://review.openstack.org/#/c/320056/11/samples/deployments/existing-with-keystone-endpoint-type.json
20:33 njohnston I can create the deployment that way, I was using --fromenv
20:34 KevinE njohnston: I'm not entirely sure how it works when using --fromenv. Some OS_INTERFACE magic
20:34 njohnston KevinE: That... sounds magical.  I'll create it from a file as you did.
20:37 KevinE njohnston: let me know what happens!
20:37 njohnston same problem
20:38 njohnston "admin endpoint for keystone service in mimi region not found"
20:38 KevinE when you check your deployment config now, does it pass that keystone_endpoint_type?
20:39 njohnston yes, I have it set to 'admin' like the example you showed me.  What other values can there be?
20:39 KevinE njohnston: can you set it to public?
20:40 njohnston same result
20:40 KevinE okay then I think andreykurilin__ may have to read this chat when he gets a chance, I'm not too sure without being able to check myself
20:41 njohnston I previously sent him a copy of my entire keystone catalog I believe, so he can reference that as well, see if there is anything wonky in it
20:43 KevinE njohnston: The error you're getting is different from the error this fix was working to resolve which is interesting to me. My error was always this: WARNING keystoneclient.auth.identity.base [-] Failed to contact the endpoint at http://admin.com:35357 for discovery. Fallback to using that endpoint as the base url. WARNING rally.common.broker [-] Failed to consume a task from the queue: Unable to establish conne
20:43 KevinE ction tohttp://admin.com:35357/domains/default
20:45 njohnston Interesting; Andrey emailed me to see if this patch fixed my issue, which is the source of my interest in it.
20:46 KevinE njohnston: and the error you commented is the one you've always gotten?
20:46 KevinE njohnston: it's the same with me, this patch was my big interest, but it appears we have similar but not exact issues, where it could maybe 'solve mine' but not yet solve yours?
20:47 njohnston KevinE: Yes, the error I commented is the same result as always, sadly.
20:48 njohnston Based on your description I think our issues are definitely different but are close cousins, most likely.
20:48 njohnston I just emailed Andrey with a pointer to the eavesdrop logs of this, so he can take a look
20:49 KevinE njohnston: so I guess if it works at my env, then the patch is good and yours is still not addressed
20:50 njohnston I defer to Andrey's expertise and yours, but I would tenatively say you are correct.
20:50 KevinE njohnston: my issue is/was "Me: Ok rally, hit the public endpoint. Rally: error hitting admin endpoint Me: what"
20:51 KevinE and you can take that back, I still have 0 expertise, I actually laughed out lout when I read that
20:52 njohnston :-)
20:53 KevinE stpierre: same review, you requested a unit test, I'm looking and it seems that Andrey actually removed a test from that exact spot a couple revisions back. Do you have any idea why?
20:53 KevinE stpierre: https://review.openstack.org/#/c/320056/8..11/tests/unit/test_osclients.py
20:54 stpierre probably because keystone_endpoint_type is None, and when it's None interface doesn't get set at all
20:54 stpierre we need something to cover the case where keystone_endpoint_type isn't None
20:55 KevinE stpierre: so... wouldn't it be a copy paste of that if statement from osclients.py... to test_osclients.py?
20:56 KevinE stpierre: this one https://review.openstack.org/#/c/320056/11/rally/osclients.py
20:56 stpierre no, it would be creating a new test case where you explicitly set keystone_endpoint_type and then make sure that interface is set appropriately
20:56 stpierre copying that if statement won't do a thing if we never set keystone_endpoint_type to anything in the unit tests
20:56 stpierre we need to actually test that keystone_endpoint_type is respected by the code
20:56 stpierre and right now we don't
20:57 njohnston and since this is user input it'd be good to have a test to ensure the correct thing happens if the user puts in garbage input, like 'keystone_endpoint_type: arglebargle'.
20:59 * KevinE stares blankly
22:23 andreykurilin__ njohnston: we have a validation for config, but it is no strict yet
22:23 andreykurilin__ *for deployment config
22:24 andreykurilin__ njohnston: Could you share logs after applying latest patch-set? Also, please sahre your deployment config
22:26 KevinE andreykurilin__: I'm pretty sure his logs are stated as a comment on the review, right?
22:27 andreykurilin__ KevinE: logs from gerrit were made before usage of keystone_endpoint_type variable
22:33 aernhart joined #openstack-rally
22:35 KevinE andreykurilin__: ok. I tried to look into adding that unit test but I still don't know enough to have done it
22:56 diogogmt joined #openstack-rally
23:39 sai joined #openstack-rally
23:58 sai joined #openstack-rally

| Channels | #openstack-rally index | Today | | Search | Google Search | Plain-Text | summary