| *** yamamoto has joined #openstack-dragonflow | 00:08 | |
| *** yamamoto has quit IRC | 00:10 | |
| *** yamamoto has joined #openstack-dragonflow | 00:10 | |
| *** yamamoto has quit IRC | 00:10 | |
| *** yamamoto has joined #openstack-dragonflow | 01:11 | |
| *** yamamoto has quit IRC | 01:16 | |
| *** yamamoto has joined #openstack-dragonflow | 02:13 | |
| *** yamamoto has quit IRC | 02:18 | |
| *** yamamoto has joined #openstack-dragonflow | 03:15 | |
| *** yamamoto has quit IRC | 03:20 | |
| *** yamamoto has joined #openstack-dragonflow | 04:16 | |
| *** yamamoto has quit IRC | 04:22 | |
| *** yamamoto has joined #openstack-dragonflow | 05:18 | |
| *** yamamoto has quit IRC | 05:24 | |
| leyal | Good morning :) | 06:14 |
|---|---|---|
| *** yamamoto has joined #openstack-dragonflow | 06:20 | |
| *** yamamoto has quit IRC | 06:25 | |
| *** irenab has joined #openstack-dragonflow | 06:54 | |
| *** snapiri has joined #openstack-dragonflow | 06:59 | |
| dimak | Good morning | 07:06 |
| snapiri | Morning :) | 07:12 |
| *** yamamoto has joined #openstack-dragonflow | 07:22 | |
| *** yamamoto has quit IRC | 07:27 | |
| irenab | morning | 07:41 |
| *** yamamoto has joined #openstack-dragonflow | 08:23 | |
| *** yamamoto has quit IRC | 08:28 | |
| oanson | Morning | 08:37 |
| oanson | leyal, ping | 08:47 |
| leyal | oanson, pong | 08:58 |
| oanson | Yo. I wanted to ask you about a patch | 08:58 |
| oanson | (one sec, retrieving (: ) | 08:59 |
| oanson | Ah, I see you answered. | 08:59 |
| oanson | It's the redis driver patch https://review.openstack.org/#/c/496299/ | 08:59 |
| oanson | leyal, I see essentially, all keys are generated via _key_name. So we can just change that in the future and do benchmarks to see what works best | 09:00 |
| oanson | Basically, the key structure isn't a real change from the previous version. | 09:01 |
| oanson | leyal, you thoughts? | 09:02 |
| oanson | your* | 09:02 |
| *** openstackgerrit has joined #openstack-dragonflow | 09:03 | |
| openstackgerrit | Omer Anson proposed openstack/dragonflow master: Support Grenade upgrade scenario https://review.openstack.org/496837 | 09:03 |
| leyal | oanson, i think the change is more significant than the the _key_name function - as it's require rewrite the set and the get_all function each in the optimize way according to key structure .. | 09:03 |
| oanson | Looking through the code, it still goes through _key_name . In any case, it's still contained. | 09:05 |
| leyal | I also think it's more nice for user to have same view when they use the df-db util - now it's depends on the driver. but don't think that critical - and i don't have real understanding about redis QL . so i removed my -1 .. | 09:06 |
| oanson | leyal, yes to the df-db util. But it goes through the db-api (and now the nb_api? I don't remember if that patch was merged). Which is a unified API that everyone has to implement anyway. | 09:07 |
| leyal | oanson, i think that you can remove the __key_name_infer_topic in if you change the key - and than you can reduce the scan in such cases .. | 09:08 |
| oanson | Hmm, I see what you mean | 09:09 |
| oanson | But this just means we change two methods (_key_name and _key_name_infer_topic) instead of one (just _key_name) | 09:10 |
| oanson | Still a contained change, and an optimization issue. | 09:10 |
| leyal | about the DF-DB in ls we print the list of keys - see: https://github.com/openstack/dragonflow/blob/58f3a46c127f5d9b94f34e0ebcf2e11fcb47a30a/dragonflow/cli/df_db.py#L70-L70 | 09:10 |
| oanson | If you like, I can add a future patch with a configuration option whether to use one method or the other | 09:11 |
| oanson | leyal, yes. But the DB driver returns the keys correctly (if it doesn't, that's a bug). But note that it calls _strip_table_topic on each key | 09:11 |
| leyal | oanson , agree that not critical just an optimization (and i don't sure which one is better)- that way i removed the -1 :) | 09:13 |
| oanson | Sure. Just trying to reach a consensus :) | 09:13 |
| leyal | oanson, thanks , didn't noticed that | 09:14 |
| oanson | We were having an 'open exchange of ideas' :) | 09:14 |
| leyal | sure | 09:16 |
| *** yamamoto has joined #openstack-dragonflow | 09:24 | |
| *** yamamoto has quit IRC | 09:30 | |
| *** yamamoto has joined #openstack-dragonflow | 09:57 | |
| *** yamamoto_ has joined #openstack-dragonflow | 09:58 | |
| *** yamamoto has quit IRC | 10:02 | |
| openstackgerrit | Merged openstack/dragonflow master: Updated from global requirements https://review.openstack.org/545535 | 10:46 |
| openstackgerrit | Omer Anson proposed openstack/dragonflow master: Detect source port for DNAT ingress https://review.openstack.org/539546 | 11:40 |
| openstackgerrit | Omer Anson proposed openstack/dragonflow master: [WIP][DNR]Refactor sg to use reg6/reg7 and not IPs https://review.openstack.org/543863 | 11:43 |
| *** yamamoto_ has quit IRC | 12:33 | |
| *** yamamoto has joined #openstack-dragonflow | 12:35 | |
| *** yamamoto has quit IRC | 12:37 | |
| *** yamamoto has joined #openstack-dragonflow | 12:37 | |
| -openstackstatus- NOTICE: Zuul has been restarted and queues were saved. However, patches uploaded after 14:40UTC may have been missed. Please recheck your patchsets where needed. | 15:54 | |
| *** leyal has quit IRC | 16:49 | |
| *** leyal has joined #openstack-dragonflow | 16:50 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!