| *** jamielennox is now known as jamielennox|away | 00:58 | |
| *** catintheroof has quit IRC | 01:02 | |
| *** jamielennox|away is now known as jamielennox | 01:12 | |
| openstackgerrit | Merged openstack/barbican: Revert "Add ID property to all entities" https://review.openstack.org/426893 | 01:12 |
|---|---|---|
| *** jaosorior has quit IRC | 01:17 | |
| *** catintheroof has joined #openstack-barbican | 01:37 | |
| *** catintheroof has quit IRC | 01:37 | |
| *** catintheroof has joined #openstack-barbican | 01:37 | |
| openstackgerrit | Dave McCowan proposed openstack/castellan: Remove outdated comment https://review.openstack.org/409887 | 02:10 |
| *** chris_hultin is now known as chris_hultin|AWA | 02:13 | |
| *** zz_dimtruck is now known as dimtruck | 02:55 | |
| openstackgerrit | Merged openstack/castellan: Remove outdated comment https://review.openstack.org/409887 | 02:55 |
| *** catintheroof has quit IRC | 03:09 | |
| *** catintheroof has joined #openstack-barbican | 03:11 | |
| *** agrebennikov__ has quit IRC | 03:14 | |
| *** catintheroof has quit IRC | 03:15 | |
| openstackgerrit | Dave McCowan proposed openstack/python-barbicanclient: Add client list filter functionality https://review.openstack.org/400370 | 03:59 |
| *** dave-mccowan has quit IRC | 04:26 | |
| *** dimtruck is now known as zz_dimtruck | 04:48 | |
| *** zz_dimtruck is now known as dimtruck | 05:39 | |
| *** dimtruck is now known as zz_dimtruck | 05:49 | |
| *** zz_dimtruck is now known as dimtruck | 05:50 | |
| *** dimtruck is now known as zz_dimtruck | 06:00 | |
| *** spotz is now known as spotz_zzz | 06:46 | |
| *** zz_dimtruck is now known as dimtruck | 06:51 | |
| *** spotz_zzz is now known as spotz | 06:52 | |
| *** dimtruck is now known as zz_dimtruck | 07:01 | |
| *** zz_dimtruck is now known as dimtruck | 07:12 | |
| *** dimtruck is now known as zz_dimtruck | 07:22 | |
| *** andreas_s has joined #openstack-barbican | 07:36 | |
| *** zz_dimtruck is now known as dimtruck | 08:13 | |
| *** nkinder has joined #openstack-barbican | 08:18 | |
| *** dimtruck is now known as zz_dimtruck | 08:23 | |
| *** zz_dimtruck is now known as dimtruck | 09:14 | |
| *** dimtruck is now known as zz_dimtruck | 09:23 | |
| *** liujiong has joined #openstack-barbican | 09:32 | |
| openstackgerrit | Jeremy Liu proposed openstack/barbican: Make error message clear when no supported secret store found https://review.openstack.org/414473 | 09:52 |
| *** zz_dimtruck is now known as dimtruck | 10:14 | |
| *** liujiong has quit IRC | 10:15 | |
| *** dimtruck is now known as zz_dimtruck | 10:24 | |
| *** shohel has joined #openstack-barbican | 10:37 | |
| *** nkinder has quit IRC | 10:59 | |
| *** dmellado has quit IRC | 11:00 | |
| *** dmellado has joined #openstack-barbican | 11:03 | |
| *** zz_dimtruck is now known as dimtruck | 11:15 | |
| *** dimtruck is now known as zz_dimtruck | 11:25 | |
| *** dave-mccowan has joined #openstack-barbican | 12:06 | |
| *** zz_dimtruck is now known as dimtruck | 12:16 | |
| *** catintheroof has joined #openstack-barbican | 12:22 | |
| *** dimtruck is now known as zz_dimtruck | 12:26 | |
| *** nkinder has joined #openstack-barbican | 12:30 | |
| *** jmckind has joined #openstack-barbican | 13:09 | |
| *** zz_dimtruck is now known as dimtruck | 13:17 | |
| *** dimtruck is now known as zz_dimtruck | 13:26 | |
| *** nkinder has quit IRC | 13:35 | |
| *** catinthe_ has joined #openstack-barbican | 13:36 | |
| *** armstrong has joined #openstack-barbican | 13:39 | |
| *** catintheroof has quit IRC | 13:40 | |
| *** armstrong has quit IRC | 13:41 | |
| *** armstrong has joined #openstack-barbican | 13:41 | |
| *** nkinder has joined #openstack-barbican | 13:58 | |
| *** zz_dimtruck is now known as dimtruck | 14:17 | |
| *** agrebennikov__ has joined #openstack-barbican | 14:27 | |
| *** dimtruck is now known as zz_dimtruck | 14:27 | |
| *** jmckind_ has joined #openstack-barbican | 14:33 | |
| *** jmckind has quit IRC | 14:34 | |
| *** armstrong has quit IRC | 14:40 | |
| *** armstrong has joined #openstack-barbican | 14:41 | |
| *** armstrong has quit IRC | 14:46 | |
| *** armstrong has joined #openstack-barbican | 14:48 | |
| *** jmckind has joined #openstack-barbican | 14:48 | |
| *** zz_dimtruck is now known as dimtruck | 14:51 | |
| *** armstrong has quit IRC | 14:52 | |
| *** jmckind_ has quit IRC | 14:52 | |
| *** armstrong has joined #openstack-barbican | 14:52 | |
| *** armstrong has quit IRC | 14:57 | |
| *** tdink has joined #openstack-barbican | 15:00 | |
| *** dimtruck is now known as zz_dimtruck | 15:04 | |
| *** armstrong has joined #openstack-barbican | 15:13 | |
| *** nkinder has quit IRC | 15:13 | |
| *** jmckind has quit IRC | 15:17 | |
| *** jaosorior has joined #openstack-barbican | 15:20 | |
| *** nkinder has joined #openstack-barbican | 15:21 | |
| *** edtubill has joined #openstack-barbican | 15:26 | |
| *** jmckind has joined #openstack-barbican | 15:26 | |
| openstackgerrit | Dave McCowan proposed openstack/barbican: Split serial and parallel functional test runs https://review.openstack.org/427251 | 15:28 |
| *** chris_hultin|AWA is now known as chris_hultin | 15:29 | |
| *** chris_hultin is now known as chris_hultin|AWA | 15:30 | |
| *** chris_hultin|AWA is now known as chris_hultin | 15:33 | |
| *** armstrong has quit IRC | 15:34 | |
| *** catinthe_ has quit IRC | 15:45 | |
| openstackgerrit | Merged openstack/barbican: Clean imports in code https://review.openstack.org/424413 | 15:59 |
| *** tdink has quit IRC | 16:14 | |
| *** tdink has joined #openstack-barbican | 16:14 | |
| *** shohel has quit IRC | 16:16 | |
| *** nkinder has quit IRC | 16:19 | |
| *** jmckind has quit IRC | 16:20 | |
| *** zz_dimtruck is now known as dimtruck | 16:20 | |
| *** tdink_ has joined #openstack-barbican | 16:24 | |
| *** tdink has quit IRC | 16:24 | |
| *** nkinder has joined #openstack-barbican | 16:25 | |
| dave-mccowan | https://review.openstack.org/427251 | 16:29 |
| dave-mccowan | alee jaosorior ^^ | 16:29 |
| dave-mccowan | this cuts two of the gate jobs from 60 minutes to 30 and stops the time outs. | 16:30 |
| jaosorior | dave-mccowan: the dogtag tests actually suffer a lot from that | 16:31 |
| jaosorior | dave-mccowan: maybe it would be appropriate to hvae another gate job, that only runs the parallel tests | 16:31 |
| dave-mccowan | jaosorior +1 i'll add that later. for now, i just want to stop the timeouts so we can clear the backlog and cut RC1. | 16:33 |
| jaosorior | dave-mccowan: that makes sense. Got my +2. Thanks for this patch! | 16:33 |
| dave-mccowan | jaosorior to help the dogtag gate, do we need another change? | 16:33 |
| jaosorior | verifying | 16:35 |
| jaosorior | dave-mccowan: As far as I can see that patch should help | 16:36 |
| alee | jaosorior, looks good to me | 16:37 |
| alee | dave-mccowan, ^^ | 16:37 |
| alee | jaosorior, is the dogtag gate still running parallel tests? | 16:37 |
| jaosorior | alee: it shouldn't with that patch | 16:37 |
| alee | hmm ., except that its still failing .. maybe timeout .. | 16:38 |
| jaosorior | alee: it did time out | 16:38 |
| dave-mccowan | jaosorior alee we need to figure out how to get the yum installs to go faster. i think that takes a big chunk of time. | 16:39 |
| jaosorior | dave-mccowan: I think that was the purpose of bindep.txt | 16:41 |
| jaosorior | but I'm not entirely sure | 16:41 |
| alee | jaosorior, dave-mccowan yeah -- just looking at the timing .. | 16:43 |
| alee | the job starts at 15:28 | 16:44 |
| alee | at 15:37 the node is up and the job timeout is set to 47 minutes | 16:44 |
| alee | at 16:20 devstack run is done (that includes all the dogtag setup/yum installs) | 16:45 |
| alee | so at that poitn test runs start | 16:45 |
| alee | and it times out at 16:24 | 16:45 |
| alee | so yeah the devstack setup is taking about 40 minutes , leaving us 7 minutes to run tests | 16:46 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/barbican: Add Dogtag plugin dependencies to bindep.txt https://review.openstack.org/427287 | 16:46 |
| jaosorior | alee, dave-mccowan: Trying it out ^^ | 16:46 |
| alee | dave-mccowan, jaosorior -- its not completely yum install issues .. http://logs.openstack.org/51/427251/1/check/gate-barbican-dogtag-devstack-fedora-25-nv/7622f53/logs/devstacklog.summary.txt.gz | 16:49 |
| dave-mccowan | alee jaosorior i see lots of ceilometer processes in enabled_services. could they be using a chunk of time unnecessarily? | 16:50 |
| jaosorior | dave-mccowan: any reason we're enabling ceilometer there? | 16:51 |
| dave-mccowan | jaosorior i can't think of one | 16:51 |
| jaosorior | dave-mccowan: any idea where we're enabling it? | 16:52 |
| jaosorior | dave-mccowan: oh, right, for the tempest devstack gate we just enable all default services | 16:54 |
| jaosorior | didn't know dogtag was using that too | 16:54 |
| dave-mccowan | jaosorior the extra time is probably due to yum install of dogtag. maybe we can get a mirror of it inside to zuul? | 16:57 |
| jaosorior | dave-mccowan: not sure how to do that. but it sounds like a good idea | 17:00 |
| alee | dave-mccowan, jaosorior so I took a look at the dogtag install times .. | 17:02 |
| alee | based on the log .. | 17:02 |
| alee | let me paste them -- | 17:02 |
| alee | we might be able to reduce time by consolidating some steps but not much .. | 17:03 |
| alee | dave-mccowan, jaosorior https://paste.fedoraproject.org/541685/14858825/ | 17:09 |
| alee | dave-mccowan, jaosorior not a whole lot f time is doing yum installs - though we would likely do better by consolidating the calls | 17:10 |
| jaosorior | crap... so what's taking so long then? | 17:10 |
| *** andreas_s has quit IRC | 17:10 | |
| alee | right now we call yum 3 times... | 17:10 |
| alee | first time takes 12 seconds, second time 16 seconds, 3rd time 2 seconds | 17:11 |
| jaosorior | is there a way we can just disable the telemetry services from our devstack runs? | 17:11 |
| alee | so if we do all the installs in a single call, we can save some time | 17:12 |
| jaosorior | alee: that should be an easy refactor | 17:13 |
| alee | pkispawn of the ca and kra takes some time .. | 17:13 |
| alee | not much we can do with those .. | 17:13 |
| jaosorior | alee: should we officially deprecate the CA bits and remove that from the tests? | 17:13 |
| alee | jaosorior, yeah we can stop doing those tests - that will save some time | 17:13 |
| alee | jaosorior, dave-mccowan - so we have to pkispawn a ca and then a kra , and we have to wait in between for the ca to restart .. | 17:14 |
| alee | ca takes 42 seconds, wait takes 11 seconds, kratakes 35 seconds | 17:15 |
| alee | if we do the ca, then do other stuff (which will likely take more than 11 seconds) and then do kra in a later script | 17:16 |
| alee | we can save the 11 wait in between) | 17:16 |
| alee | maybe do the kra in the script that configures barbican to use kra | 17:16 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/barbican: Move dogtag plugin dependencies to one yum call https://review.openstack.org/427313 | 17:17 |
| jaosorior | alee: ^^ | 17:17 |
| alee | all in all - these refactorings might give us 20-25 seconds .. | 17:17 |
| alee | jaosorior, works for me -- we might get 10 seconds out of that | 17:18 |
| alee | dave-mccowan, jaosorior I suspect greater time savings will be gained by killing telemetry etc. | 17:23 |
| jaosorior | alee: I agree, but I'm unsure how to do that | 17:23 |
| alee | jaosorior, dont we configure which services to install in devstack? | 17:23 |
| jaosorior | alee: we use the defaults. I don't know if there's a way to just epxlicitly remove telemtry | 17:26 |
| alee | jaosorior, well the other way is to specify what we want | 17:26 |
| dave-mccowan | jaosorior alee in the xenial/simple-crypto run it takes devstack 5 minutes to get to "starting rabbit". in the fedora/dogtag run it takes 17 minutes. | 17:28 |
| dave-mccowan | both runs have the same enabled openstack services | 17:29 |
| alee | dave-mccowan, what about the other runs? | 17:30 |
| jaosorior | it makes no sense that it takes 17 minutes to start rabbit there :/ | 17:31 |
| alee | dave-mccowan, it might be that rpms installs in general are slower .. | 17:31 |
| alee | (in this environment) | 17:31 |
| alee | so yeah - speeding up rpm installs would affect the whole install - not just the dogtag bits which take up a relatively small part of the install .. | 17:32 |
| alee | dave-mccowan, jaosorior so basically the dogtag bits alone add 2 minutes and 12 seconds to the process. | 17:35 |
| alee | dave-mccowan, jaosorior that leaves 17 - 5 - 2.12 minutes unaccounted for | 17:36 |
| alee | almost 10 minutes of generally taking longer | 17:37 |
| jaosorior | dave-mccowan: anything in the rabbitmq logs? | 17:38 |
| * jaosorior has been having issues with rabbitmq all week... | 17:38 | |
| jaosorior | this just adds to it | 17:38 |
| *** jmckind has joined #openstack-barbican | 17:43 | |
| dave-mccowan | alee jaosorior the kolla project adds these lines to use mirrors local to zuul: https://github.com/openstack/kolla/blob/master/tools/setup_gate.sh#L11 | 17:45 |
| jaosorior | dave-mccowan: nice, yet we would need to ask the infra folks to serve the needed packages in their mirrors, right? | 17:46 |
| *** nkinder has quit IRC | 17:46 | |
| dave-mccowan | jaosorior yep | 17:46 |
| *** catintheroof has joined #openstack-barbican | 17:55 | |
| *** jamielennox is now known as jamielennox|away | 17:58 | |
| *** dimtruck is now known as zz_dimtruck | 18:04 | |
| redrobot | good mornin' barbican! | 18:05 |
| redrobot | sorry I was out yesterday | 18:05 |
| *** jperry has joined #openstack-barbican | 18:10 | |
| dave-mccowan | hi redrobot! | 18:23 |
| jaosorior | rm_work: around? | 18:27 |
| jaosorior | redrobot: hey dude, how's it going? | 18:28 |
| redrobot | jaosorior pretty good! finally closed on a house | 18:32 |
| redrobot | jaosorior that's where I was yesterday... signing my money away, lol | 18:33 |
| *** zz_dimtruck is now known as dimtruck | 18:51 | |
| rm_work | jaosorior: yes now | 18:57 |
| rm_work | redrobot: ah nice grats! saw you were at PAX but didn't managed to catch you in the crowd | 18:58 |
| jaosorior | rm_work: are you acquainted with the Octavia HAProxy Amphora bits? | 18:58 |
| rm_work | jaosorior: yes | 18:58 |
| rm_work | ... pretty intimately | 18:58 |
| rm_work | we've spent a lot of nights together | 18:58 |
| jaosorior | rm_work: so, any hints on how the TLS certs are handled there? haven't found much documentation around that | 18:59 |
| rm_work | on the Amphorae themselves? | 18:59 |
| rm_work | they're stored in a ramfs | 18:59 |
| rm_work | encrypted ramfs | 18:59 |
| rm_work | though, i think there is something broken with that right now | 19:00 |
| rm_work | there might be a bug in LP even... | 19:00 |
| * rm_work looks | 19:00 | |
| jaosorior | rm_work: and who provisions those certs, and how are they passed on-to the nova nodes? | 19:01 |
| rm_work | err | 19:01 |
| rm_work | sec | 19:02 |
| jaosorior | dave-mccowan: can you check this out https://review.openstack.org/#/c/427313/ ? | 19:13 |
| jaosorior | redrobot: congrats dude! | 19:14 |
| *** jaosorior has quit IRC | 19:18 | |
| *** jaosorior has joined #openstack-barbican | 19:18 | |
| rm_work | jaosorior: so uhh | 19:20 |
| rm_work | jaosorior: assuming you mean the user certs that are used for the TLS termination... the user provisions those however, and stores them in Barbican themselves | 19:21 |
| rm_work | jaosorior: from there, the user passes the container-id to Octavia, which pulls them from barbican and passes them as data over HTTPS to our agent on the amphora, which writes them into the encrypted ramfs | 19:21 |
| rm_work | so, our controller-worker on the control plane does the fetch | 19:22 |
| rm_work | the amphorae themselves are only passed some data over HTTPS, they can't pull anything themselves | 19:22 |
| openstackgerrit | Peter Hamilton proposed openstack/barbican-tempest-plugin: Add an OrderClient and order API tests https://review.openstack.org/423346 | 19:29 |
| rm_work | I really wish we could figure out the consumer/ACL issues once and for all | 19:30 |
| rm_work | it MIGHT be in a better state now actually, I should re-test, just haven't had a lot of time | 19:30 |
| *** pcaruana has quit IRC | 19:31 | |
| openstackgerrit | Merged openstack/barbican: Split serial and parallel functional test runs https://review.openstack.org/427251 | 19:32 |
| rm_work | yeah we're updating to this I think: https://review.openstack.org/#/c/403817/ | 19:33 |
| rm_work | we are currently using ecryptfs | 19:33 |
| *** tdink has joined #openstack-barbican | 19:34 | |
| *** tdink_ has quit IRC | 19:36 | |
| rm_work | jaosorior: ^^ does that answer your questions? | 19:44 |
| openstackgerrit | Merged openstack/barbican: Move dogtag plugin dependencies to one yum call https://review.openstack.org/427313 | 19:48 |
| jaosorior | rm_work: more or less, still got a bucnh to read haha | 20:06 |
| *** dave-mccowan has quit IRC | 20:46 | |
| *** tdink has quit IRC | 20:48 | |
| *** tdink has joined #openstack-barbican | 20:49 | |
| *** diazjf has joined #openstack-barbican | 20:50 | |
| *** jaosorior has quit IRC | 21:23 | |
| *** diazjf has quit IRC | 21:24 | |
| *** kfarr has joined #openstack-barbican | 21:28 | |
| *** tdink has quit IRC | 21:30 | |
| *** tdink has joined #openstack-barbican | 21:30 | |
| *** diazjf has joined #openstack-barbican | 21:38 | |
| *** catintheroof has quit IRC | 21:38 | |
| *** catintheroof has joined #openstack-barbican | 21:38 | |
| *** catintheroof has quit IRC | 21:39 | |
| *** jaosorior has joined #openstack-barbican | 21:55 | |
| *** tdink has quit IRC | 22:02 | |
| *** chris_hultin is now known as chris_hultin|AWA | 22:05 | |
| *** jmckind has quit IRC | 22:05 | |
| *** catintheroof has joined #openstack-barbican | 22:05 | |
| *** catintheroof has quit IRC | 22:05 | |
| *** catintheroof has joined #openstack-barbican | 22:06 | |
| *** catintheroof has quit IRC | 22:11 | |
| *** jamielennox|away is now known as jamielennox | 22:15 | |
| *** diazjf has quit IRC | 22:16 | |
| *** nkinder has joined #openstack-barbican | 22:23 | |
| openstackgerrit | Peter Hamilton proposed openstack/barbican-tempest-plugin: Add a QuotaClient and quota API tests https://review.openstack.org/427488 | 22:31 |
| *** kfarr has quit IRC | 22:33 | |
| *** tdink has joined #openstack-barbican | 22:34 | |
| *** tdink has quit IRC | 22:36 | |
| *** tdink has joined #openstack-barbican | 22:36 | |
| *** dimtruck is now known as zz_dimtruck | 22:49 | |
| *** edtubill has quit IRC | 23:02 | |
| *** jperry has quit IRC | 23:10 | |
| *** jperry has joined #openstack-barbican | 23:21 | |
| *** diazjf has joined #openstack-barbican | 23:31 | |
| *** dave-mccowan has joined #openstack-barbican | 23:43 | |
| *** diazjf has quit IRC | 23:52 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!