Tuesday, 2018-05-01

*** tosky has quit IRC00:00
*** gildub has joined #openstack-sdks00:20
*** d0ugal__ has joined #openstack-sdks00:27
gtmanfredok, maybe tomorrow00:37
*** d0ugal__ has quit IRC00:58
*** d0ugal__ has joined #openstack-sdks01:06
*** slaweq has joined #openstack-sdks01:11
openstackgerritMark Kirkwood proposed openstack/python-openstackclient master: Add an option to container create for storage policy  https://review.openstack.org/56466501:13
*** slaweq has quit IRC01:16
*** salv-orlando has joined #openstack-sdks01:31
*** thrash is now known as thrash|g0ne01:34
*** salv-orlando has quit IRC01:36
*** d0ugal__ has quit IRC01:39
*** chenyb4 has joined #openstack-sdks02:08
*** dave-mccowan has joined #openstack-sdks02:11
*** chenyb4 has quit IRC02:42
*** slaweq has joined #openstack-sdks02:56
*** slaweq has quit IRC03:01
mnaserit seems like providing 'key_manager_endpoint_override' doesn't seem to properly work in a connection?03:05
*** d0ugal__ has joined #openstack-sdks03:06
*** d0ugal__ has quit IRC03:21
*** d0ugal__ has joined #openstack-sdks03:22
*** d0ugal__ has quit IRC03:29
*** salv-orlando has joined #openstack-sdks03:32
*** d0ugal__ has joined #openstack-sdks03:36
*** salv-orlando has quit IRC03:37
*** d0ugal__ has quit IRC03:47
*** lbragstad has joined #openstack-sdks03:49
*** gildub has quit IRC04:24
*** salv-orlando has joined #openstack-sdks04:33
*** salv-orlando has quit IRC04:37
*** dave-mccowan has quit IRC04:43
*** lbragstad has quit IRC04:51
openstackgerritXiaojueGuan proposed openstack/keystoneauth master: Trivial: Update pypi url to new url  https://review.openstack.org/56541804:59
*** salv-orlando has joined #openstack-sdks05:01
*** salv-orlando has quit IRC05:02
*** slaweq has joined #openstack-sdks06:11
*** slaweq has quit IRC06:15
*** salv-orlando has joined #openstack-sdks07:04
*** jpich has joined #openstack-sdks07:04
*** salv-orlando has quit IRC07:09
*** salv-orlando has joined #openstack-sdks07:10
*** salv-orlando has quit IRC07:14
*** gildub has joined #openstack-sdks07:34
*** gkadam has joined #openstack-sdks07:45
*** slaweq has joined #openstack-sdks08:11
*** slaweq has quit IRC08:15
*** ralonsoh has joined #openstack-sdks08:37
*** d0ugal__ has joined #openstack-sdks08:57
*** d0ugal__ has quit IRC09:02
*** slaweq has joined #openstack-sdks09:07
*** cdent has joined #openstack-sdks09:08
*** slaweq has quit IRC09:22
*** slaweq has joined #openstack-sdks09:23
*** lifeless_ is now known as lifeless09:46
*** cdent has quit IRC09:50
*** gildub has quit IRC09:58
*** d0ugal has joined #openstack-sdks10:00
*** gildub has joined #openstack-sdks10:11
*** gtema has joined #openstack-sdks10:28
mordredmnaser: no?10:36
lifelessmordred: ETIMEZONE10:36
mordredlifeless: so much E10:38
mordredwait - that implies the wrong thing :)10:38
mordredlifeless: also, o hai10:39
*** gtema has quit IRC10:58
openstackgerritMonty Taylor proposed openstack/openstacksdk master: Defer all endpoint discovery to keystoneauth  https://review.openstack.org/56449411:00
*** slaweq has quit IRC11:22
mnasermordred: I’ll paste what I have that’s not working for me. By looking over the code it seems that cloud region get_endpoint_session() uses the override however the normal adapter doesn’t seem to use it11:45
mnasermordred: use case here is using sensu-plugins-python + openstacksdk for small integration monitoring tests11:45
*** thrash|g0ne is now known as thrash11:48
*** slaweq has joined #openstack-sdks11:50
*** dave-mccowan has joined #openstack-sdks12:08
*** tosky has joined #openstack-sdks12:13
mordredmnaser: seems like a good use case - I'm looking through the code to see if I can see where endpoint_override isn't getting passed ...12:19
*** edmondsw has joined #openstack-sdks12:19
mordredmnaser: ah. there it is. it's in the part where it's not getting passed12:20
mnasermordred: currently doing this.. but it also feels a little dirty, i could be doing it the wrong way12:22
mnaseri didnt find a way to just connect to a region with overrides12:22
mordredyah - there'sa bug - patch coming12:22
mnaser(the reason i'm overriding is because i want to test the *local* instance using that monitoring agent rather than going through load balancer, etc)12:23
mnaserand cool thank you!12:23
openstackgerritMonty Taylor proposed openstack/openstacksdk master: Honor endpoint_override for get_session_client  https://review.openstack.org/56548912:24
mordredmnaser: ^^12:24
mnaserlet me try and test that locally12:24
mnasermordred: your patch is missing a , after 'endpoint_override=self.get_endpoint(service_key)'12:29
mnaser(before the **kwargs)12:29
mnaserand it also fixes the issue-- CheckBarbicanIntegration CRITICAL: Unable to establish connection to http://foobar/secrets: HTTPConnectionPool(host='foobar', port=80): Max retries exceeded with url: /secrets (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x30922d0>: Failed to establish a new connection: [Errno -2] Name or service not known',)) :D12:30
mordredmnaser: whoops12:32
openstackgerritMonty Taylor proposed openstack/openstacksdk master: Honor endpoint_override for get_session_client  https://review.openstack.org/56548912:32
mordredShrews: ^^ if you get a sec12:35
*** slaweq has quit IRC12:47
*** crunchengine has joined #openstack-sdks13:05
*** slaweq has joined #openstack-sdks13:11
*** slaweq has quit IRC13:15
*** lbragstad has joined #openstack-sdks13:32
*** openstackgerrit has quit IRC13:34
*** tosky has quit IRC13:51
*** slaweq has joined #openstack-sdks13:52
*** slaweq has quit IRC13:53
*** slaweq has joined #openstack-sdks13:53
*** gtema has joined #openstack-sdks13:55
*** gtema has quit IRC14:15
mordredShrews: you may also enjoy https://review.openstack.org/54930714:15
mordredslaweq: ^^ you may also enjoy both of those patches (or you might not :) )14:15
slaweqmordred: looking14:15
*** bobh has joined #openstack-sdks14:19
*** gildub has quit IRC14:19
*** bobh has quit IRC14:23
*** bobh has joined #openstack-sdks14:24
*** slaweq has quit IRC14:34
*** gkadam has quit IRC14:43
gtmanfredmordred: ping14:57
mordredgtmanfred: hi! I'm about to be on a phone call for an hour - but then I'll be here14:59
gtmanfredawesome, thanks14:59
mordredgtmanfred: however, that race condition you mention should not exist on a reasonably modern openstack - we use the create-floaing-ip-with-server-port approach rather than the 'create-floating-ip-then-attach-to-server' approach which suffers from the race condition mentioned15:00
mordredif the cloud in quetsion does not expose the neutron endpoints needed to safely create floating ips in a multi-process or multi-threaded manner, we will fall back to create-and-attach15:01
mordred(this is all just backgound info)15:02
gtmanfredlemme check what they are running, it is older15:04
mordredyah- the pure nova api version of this is ... not great .. for the reasons listed in the story15:06
gtmanfredmordred: he is using mitaka15:10
*** slaweq has joined #openstack-sdks15:11
mordredhrm. that should be new enough15:12
*** slaweq has quit IRC15:15
gtmanfredI also think I tested with pike, and was running into the same problem.15:17
gtmanfredwhat it seems like the problem is, is that we are trying to use floating ips that have already been allocated15:18
gtmanfredand since multiple processes list all the available ips, and then pick one, multiple processes try to assign the same ip15:18
gtmanfredthis is why we went with `create and assign by default.  if there are no free ips, log a warning that we are going to be not multiprocess safe, because we need to list available ones and pick one from a list`15:19
*** openstackgerrit has joined #openstack-sdks15:23
openstackgerritHongbin Lu proposed openstack/openstacksdk master: Add 'port_details' to Floating IP  https://review.openstack.org/53381115:23
mordredyah. multi-process re-use of pre-existing floating ips is basically unpossible - without some sort of shared locking/brokering system15:23
mordredin nodepool (which does this process in many many parallel threads) - we make sure we delete floating ips when we delete servers - and also have a cleanup thread that deletes any unattached floating ips it finds15:24
mordredsince if the create-with-server-port method is being used, there is no such thing as a validly unattached floating ip15:25
gtmanfredok, i think what I am just going to do is change the default `reuse_ips` to False, and allow them to set it if they really really really really really really really really  are dumb and want to use it15:25
openstackgerritXiaojueGuan proposed openstack/keystoneauth master: Trivial: Update pypi url to new url  https://review.openstack.org/56541815:26
*** gkadam has joined #openstack-sdks15:26
mordredgtmanfred: yah15:26
mordredgtmanfred: so - that said - there is an option we can do15:27
mordred(it's on my list already = it's just deep down there and there's a few big ticket items we need to likely get done first)15:27
mordredtask number 1 is to finish making the caching tier use dogpile.cache everywhere (servers, ports and floating ips are currently done manually if caching is enabled - and there are some tricky interactions as it relates to batched operations and thundering herd that we hvae to be careful about when fixing that)15:28
mordredbut once we've got that, then we should be able to add in a locking/multi-process aware TaskManager (similar to the multithreaded TaskManager that's in nodepool right now)15:29
mordredand leverge dogpile for the shared locking15:29
gtmanfredLemme know what I can help with, i need to figure out what to contribute to so I can go to berlin.15:30
mordredso if someone then wants to do multi-process with reuse_ips there will be a story for it15:30
mordredcool. I'll write up the above into a more consumable story that doesn't involve memories in my head and ping you with it15:30
gtmanfredthat would be great15:30
mordred\o/ winning15:30
*** jpich has quit IRC15:32
gtmanfredother than that, the new shade cloud driver has gotten rave reviews, and everyone says it is much easier to configure and more reliable than what we had before15:34
openstackgerritMerged openstack/openstacksdk master: Drop bogus attributes from network port resource  https://review.openstack.org/56521715:54
*** tosky has joined #openstack-sdks16:02
*** ralonsoh has quit IRC16:20
*** pabelanger has left #openstack-sdks16:20
*** ralonsoh has joined #openstack-sdks16:36
*** tosky has quit IRC16:37
mordredgtmanfred: yay!16:56
*** e0ne has joined #openstack-sdks17:01
*** gkadam has quit IRC17:17
*** thrash is now known as thrash|biab17:26
*** ralonsoh has quit IRC17:47
*** slaweq has joined #openstack-sdks18:11
*** slaweq has quit IRC18:15
*** slaweq has joined #openstack-sdks19:11
*** slaweq has quit IRC19:15
*** e0ne has quit IRC19:19
gtmanfredmordred: also, with the cache stuff for servers, it is about 10 times faster19:21
mordredyah. the cache stuff is really important - and the other thing is also about being able to enable at least in-memory caching by default19:23
*** slaweq has joined #openstack-sdks19:50
*** bobh has quit IRC19:51
crunchenginemordred: hello19:56
mordredcrunchengine: hiya19:57
crunchengineI am listing active instances, but compute give me migrating instances too19:57
crunchengineself.conn.compute.servers(host=self.hostname, all_tenants=True, status="ACTIVE")19:57
crunchenginethose instances are still active, from kvm point of view but the filter should be state consistent no ?19:58
mordredI dunno about the kvm POV - but the filter from a REST API perspective only knows about the value of teh status field. doesn't migration status show up in vm_state?19:59
mordredah - sorry - progress19:59
mordrednope. thinking out loud - sorry :)19:59
mordredyah - I'd expect for the status of the vm to be 'MIGRATING' from looking at the api docs20:01
crunchengineprint("uuid: {} status: {} task_state: {} vm_state: {}".format(i.id, i.status, i.task_state, i.vm_state))20:01
crunchengineuuid: 5a562937-b0e2-4e67-8b2e-490cd44bbc3d status: MIGRATING task_state: migrating vm_state: active20:01
crunchenginetoo much state !20:01
crunchengineokay thanks !20:01
crunchengineso naming is inconsistent too: return self.conn.compute.servers(host=self.hostname, all_tenants=True, status="ACTIVE")20:03
crunchenginestatus => vm_state in lowercase...20:03
*** crunchengine has quit IRC20:55
*** edmondsw has quit IRC21:48
openstackgerritMerged openstack/keystoneauth master: Allow tuples and sets in interface list  https://review.openstack.org/56449521:58
*** gildub has joined #openstack-sdks22:08
*** lbragstad has quit IRC22:22
*** thrash|biab is now known as thrash|g0ne22:30
*** lbragstad has joined #openstack-sdks22:46
*** slaweq has quit IRC22:46
*** slaweq has joined #openstack-sdks22:47
*** gildub has quit IRC23:17
*** slaweq has quit IRC23:27

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!