openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 00:59 |
---|---|---|
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient master: Updated from global requirements https://review.openstack.org/509449 | 01:10 |
openstackgerrit | Nam Nguyen Hoai proposed openstack/barbican master: [WIP] Migrate to Zuul v3 https://review.openstack.org/509348 | 01:51 |
*** dave-mcc_ has quit IRC | 01:57 | |
*** rarora has quit IRC | 02:00 | |
*** rarora has joined #openstack-barbican | 02:15 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 02:32 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient master: Updated from global requirements https://review.openstack.org/509449 | 02:42 |
openstackgerrit | Davanum Srinivas (dims) proposed openstack/castellan master: Vault based key manager https://review.openstack.org/483080 | 02:45 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 02:48 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient master: Updated from global requirements https://review.openstack.org/509449 | 02:58 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 03:03 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient master: Updated from global requirements https://review.openstack.org/509449 | 03:14 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 03:24 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient master: Updated from global requirements https://review.openstack.org/509449 | 03:34 |
*** tonyb has quit IRC | 03:42 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 03:53 |
*** tonyb has joined #openstack-barbican | 03:56 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient master: Updated from global requirements https://review.openstack.org/509449 | 04:04 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 04:09 |
*** jamielennox has quit IRC | 04:14 | |
*** jamielennox has joined #openstack-barbican | 04:18 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient master: Updated from global requirements https://review.openstack.org/509449 | 04:20 |
*** cerny01_ has quit IRC | 04:23 | |
*** chlong_ has joined #openstack-barbican | 04:51 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 06:01 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient master: Updated from global requirements https://review.openstack.org/509449 | 06:12 |
*** chlong_ has quit IRC | 06:13 | |
*** andreas_s has joined #openstack-barbican | 06:15 | |
*** david_1 has joined #openstack-barbican | 06:56 | |
*** mkoderer_ has joined #openstack-barbican | 06:56 | |
*** dgonzalez_ has joined #openstack-barbican | 06:56 | |
*** seife_ has joined #openstack-barbican | 06:56 | |
*** seife_ has quit IRC | 06:58 | |
*** dgonzalez_ has quit IRC | 06:58 | |
*** mkoderer_ has quit IRC | 06:58 | |
*** david_1 has quit IRC | 06:58 | |
*** pcaruana has joined #openstack-barbican | 07:08 | |
*** nrado has joined #openstack-barbican | 07:46 | |
*** salmankhan has joined #openstack-barbican | 08:00 | |
*** salmankhan has quit IRC | 08:07 | |
nrado | Hi there, I've noticed that the barbican-keystone-listener (in Ocata release) is a consumer of the "notifications.info" queue. I can understand the reason behind this, but in my case the keystone listener is consuming messages which should belong to ceilometer. I have missing telemetry information in my database due to this. It that a known issue? | 08:16 |
nrado | I mean I'm creating and deleting volumes.. why is barbican-keystone-listener picking the messages from the queue? | 08:20 |
*** serlex has joined #openstack-barbican | 08:22 | |
*** hieulq has quit IRC | 08:26 | |
*** hieulq has joined #openstack-barbican | 08:27 | |
*** daidv has quit IRC | 08:29 | |
*** daidv has joined #openstack-barbican | 08:30 | |
jaosorior | nrado: well, you could use allow_requeue in the configuration. | 08:49 |
jaosorior | which is in the keystone_notifications section | 08:50 |
nrado | what is this param used for? Can you please explain it to me jaosorior? | 08:54 |
nrado | but it doesn't help | 08:56 |
nrado | still getting just a part of the event when I enable this | 08:57 |
jaosorior | nrado: allow_requeue=True makes the listener not actually consume the events, it reads it and then sends it back to the queue. | 08:58 |
jaosorior | nrado: that would probably help in the short term, however, what you probably want is to use separate queues, so keystone would be putting the notifications to two queues: one for barbican and one for ceilometer. Then you could have both listeners consume the events, and avoid race-conditions | 08:59 |
jaosorior | And the race-condition I'm talking about is because ceilometer doesn't requeue. So if ceilometer consumes the event before the keystone-listener does, then the keystone-listener will never get it. | 09:02 |
*** namnh has joined #openstack-barbican | 09:04 | |
nrado | Ok, I got it, the problem is that I need both events (create.start and create.end), but I'm getting only one of these. In this case, it's more important for me that ceilometer gets both events | 09:04 |
nrado | I'll try it now with the 2nd queue | 09:04 |
jaosorior | nrado: well, if you do allow_requeue, you should get the events | 09:06 |
jaosorior | nrado: unless you got something else in your system that's consuming them. | 09:06 |
jaosorior | so yeah, under [keystone_notifications] set allow_requeue=true in barbican.conf | 09:06 |
nrado | just in barbican, not in ceilometer? | 09:07 |
jaosorior | nrado: so that would be for a quick test that it works. But yeah, if you don't set it in ceilometer, then nothing will consume those events.... though you could mitigate that by setting a TTL for your messages in rabbitmq (if that's what you're using as a message broker) https://www.rabbitmq.com/blog/2014/01/23/preventing-unbounded-buffers-with-rabbitmq/ | 09:10 |
nrado | Unfortunately, it doesn't work when I set allow_requeue=true in barbican.conf. Still getting just one part of my event | 09:12 |
jaosorior | nrado: did you set it in the keystone_notifications section? | 09:12 |
nrado | yes: | 09:13 |
nrado | [keystone_notifications] | 09:13 |
nrado | enable = True | 09:13 |
nrado | allow_requeue = True | 09:13 |
jaosorior | nrado: well, funky. Anything else in your system that might be consuming events? | 09:14 |
jaosorior | nrado: that option was available in ocata for both barbican and oslo.messaging (which ultimately consumes that option) | 09:14 |
jaosorior | brb | 09:15 |
nrado | well, on this notification.info queue I have two consumers: barbican-keystone-listener and ceilomter-notfication | 09:15 |
nrado | those two options (barbican and oslo.messaging) should be available in barbican.conf? What do you mean? Where should I configure oslo.messaging to ulimately consume that option? | 09:19 |
jaosorior | oslo messaging gets that option from barbican.conf, as it's a library that barbican uses (same as a lot of openstack services) | 09:20 |
jaosorior | I'll be back in a bit, gotta go run an errand | 09:21 |
nrado | https://github.com/openstack/barbican/blob/stable/ocata/etc/barbican/barbican.conf#L243 set this one, so it should be enough, right? | 09:21 |
jaosorior | yeah | 09:21 |
nrado | okay | 09:21 |
*** pbourke has quit IRC | 10:00 | |
*** pbourke has joined #openstack-barbican | 10:01 | |
nrado | @jaosorior: I tried to do this another queue. In barbican.conf I set "barbican_notifications" as the topic under [keystone_notifications]. In Keystone.conf I set multiple topics: | 10:15 |
nrado | [oslo_messaging_notifications] | 10:15 |
nrado | topics = notifications,barbican_notifications | 10:15 |
nrado | I can see in rabbitmq that the queue was created and the the barbican_keystone_listener is consuming the queue, but I don't get any messages there | 10:16 |
nrado | Did I forget something? When I compare the two queues (notifications.info and barbican_notifications.info) I can see that there are some more bindings in notifications.info | 10:17 |
nrado | So I could configure another binding in rabbitmq, but how is it solved in OpenStack? | 10:18 |
*** andreas_s_ has joined #openstack-barbican | 10:56 | |
*** andreas_s has quit IRC | 10:58 | |
*** dave-mccowan has joined #openstack-barbican | 11:04 | |
jaosorior | nrado: weird. We had exactly the same issue with novajoin (another service) and solved it that way. | 11:18 |
jaosorior | nrado: not sure what you're missing there :/ | 11:18 |
nrado | hmm | 12:07 |
nrado | this is indeed weird | 12:08 |
nrado | You did this also in that way in Ocata release? | 12:08 |
*** namnh has quit IRC | 12:10 | |
*** raildo has joined #openstack-barbican | 12:11 | |
*** nrado1 has joined #openstack-barbican | 12:18 | |
jaosorior | nrado: we did it for pike | 12:20 |
*** nrado has quit IRC | 12:21 | |
*** alee has quit IRC | 12:22 | |
nrado1 | Ok, I found the reason | 12:22 |
nrado1 | I configured the additional topic just in keystone | 12:22 |
nrado1 | but as I'm creating a volume I need to add this barbican topic in cinder.conf | 12:23 |
nrado1 | then it will receive messages in in this queue | 12:23 |
nrado1 | but one question: why does Barbican need all this messages | 12:24 |
nrado1 | is it not enough to just get a message when a keystone was deleted? | 12:24 |
nrado1 | I thought this was the intention behind the barbican keystone listener | 12:24 |
nrado1 | I'm really confused | 12:34 |
*** namnh has joined #openstack-barbican | 12:42 | |
*** namnh has quit IRC | 12:43 | |
jaosorior | nrado1: you're right, we don't need all those notifications | 12:43 |
*** namnh has joined #openstack-barbican | 12:43 | |
jaosorior | nrado1: that could easily be fixed :D not sure if for ocata though | 12:43 |
jaosorior | nrado1: basically, we parse in code the notifications, instead of using a notificationfilter from oslo.messaging (which would make life easier) | 12:44 |
jaosorior | nrado1: basically, we're missing something like this https://github.com/openstack/novajoin/blob/master/novajoin/notifications.py#L56 | 12:47 |
jaosorior | nrado1: if you would like to write a commit it would be most welcome :D | 12:47 |
*** catintheroof has joined #openstack-barbican | 12:50 | |
*** catintheroof has quit IRC | 12:51 | |
*** catintheroof has joined #openstack-barbican | 12:51 | |
*** noslzzp has joined #openstack-barbican | 13:05 | |
*** namnh has quit IRC | 13:15 | |
*** namnh has joined #openstack-barbican | 13:16 | |
*** namnh has quit IRC | 13:20 | |
*** salmankhan has joined #openstack-barbican | 13:25 | |
*** jaosorior has quit IRC | 13:35 | |
*** salmankhan has quit IRC | 13:36 | |
*** alee has joined #openstack-barbican | 13:36 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 13:39 |
*** salmankhan has joined #openstack-barbican | 13:43 | |
*** chlong_ has joined #openstack-barbican | 13:47 | |
*** salmankhan has quit IRC | 13:47 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient master: Updated from global requirements https://review.openstack.org/509449 | 13:50 |
*** jaosorior has joined #openstack-barbican | 14:10 | |
*** namnh has joined #openstack-barbican | 14:16 | |
*** serlex has quit IRC | 14:17 | |
nrado1 | jaosorior: First I have to find out how to get keystone notifications :D I'm struggling with this problem | 14:31 |
nrado1 | There are no keystone notifications in the notification.info queue | 14:32 |
*** jaosorior has quit IRC | 15:07 | |
*** andreas_s_ has quit IRC | 15:21 | |
nrado1 | I'm getting this error when I delete a project in keystone (Ocata release): https://gist.github.com/neki06/1bdd5247daef4c790be7fd1127dd71d2 | 15:31 |
nrado1 | Expected that the secrets created within this project will be also deleted, but it's not the case | 15:32 |
nrado1 | Can someone help? | 15:32 |
*** pcaruana has quit IRC | 15:39 | |
*** nrado1 has quit IRC | 15:44 | |
*** namnh has quit IRC | 15:52 | |
*** namnh has joined #openstack-barbican | 15:53 | |
*** namnh has quit IRC | 15:57 | |
*** noslzzp has quit IRC | 16:03 | |
*** noslzzp has joined #openstack-barbican | 16:07 | |
*** namnh has joined #openstack-barbican | 16:22 | |
*** nrado has joined #openstack-barbican | 16:43 | |
openstackgerrit | Merged openstack/barbican master: Updated from global requirements https://review.openstack.org/506611 | 16:48 |
*** alee has quit IRC | 16:56 | |
*** alee has joined #openstack-barbican | 16:59 | |
*** noslzzp has quit IRC | 17:15 | |
*** raildo has quit IRC | 17:31 | |
*** namnh has quit IRC | 17:38 | |
*** raildo has joined #openstack-barbican | 17:47 | |
*** chlong_ has quit IRC | 18:44 | |
*** chlong_ has joined #openstack-barbican | 19:00 | |
*** alee has quit IRC | 19:07 | |
*** alee has joined #openstack-barbican | 19:20 | |
*** melwitt has joined #openstack-barbican | 19:29 | |
*** melwitt has left #openstack-barbican | 19:29 | |
*** chlong_ has quit IRC | 19:44 | |
*** catintheroof has quit IRC | 20:56 | |
*** nrado has quit IRC | 21:05 | |
*** alee has quit IRC | 21:21 | |
*** raildo has quit IRC | 21:22 | |
*** alee has joined #openstack-barbican | 22:30 | |
*** nkinder has quit IRC | 22:31 | |
*** namnh has joined #openstack-barbican | 23:28 | |
*** ssathaye has quit IRC | 23:39 | |
*** ssathaye has joined #openstack-barbican | 23:39 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!