*** rcernin has quit IRC | 00:34 | |
*** rcernin has joined #openstack-oslo | 00:35 | |
*** irclogbot_2 has joined #openstack-oslo | 00:55 | |
*** irclogbot_2 has quit IRC | 01:00 | |
*** irclogbot_2 has joined #openstack-oslo | 01:25 | |
*** irclogbot_2 has quit IRC | 01:34 | |
*** irclogbot_1 has joined #openstack-oslo | 02:25 | |
*** irclogbot_1 has quit IRC | 02:30 | |
*** altlogbot_0 has joined #openstack-oslo | 02:47 | |
*** altlogbot_0 has quit IRC | 02:52 | |
*** irclogbot_1 has joined #openstack-oslo | 03:21 | |
*** irclogbot_1 has quit IRC | 03:26 | |
openstackgerrit | pengyuesheng proposed openstack/oslo.config master: Add Python 3 Train unit tests https://review.opendev.org/669233 | 03:36 |
---|---|---|
*** irclogbot_0 has joined #openstack-oslo | 03:51 | |
*** irclogbot_0 has quit IRC | 03:56 | |
*** zbitter has quit IRC | 04:53 | |
*** toabctl has quit IRC | 04:53 | |
*** zbitter has joined #openstack-oslo | 04:53 | |
*** rcernin has quit IRC | 04:54 | |
*** toabctl has joined #openstack-oslo | 04:55 | |
*** irclogbot_0 has joined #openstack-oslo | 04:59 | |
*** pcaruana has joined #openstack-oslo | 05:13 | |
*** irclogbot_0 has quit IRC | 05:16 | |
*** Luzi has joined #openstack-oslo | 05:24 | |
*** zaneb has joined #openstack-oslo | 05:51 | |
*** zbitter has quit IRC | 05:53 | |
*** jbadiapa has quit IRC | 05:55 | |
*** altlogbot_1 has joined #openstack-oslo | 06:09 | |
*** altlogbot_1 has quit IRC | 06:14 | |
*** irclogbot_2 has joined #openstack-oslo | 06:35 | |
*** irclogbot_2 has quit IRC | 06:40 | |
*** tosky has joined #openstack-oslo | 07:19 | |
*** jbadiapa has joined #openstack-oslo | 07:22 | |
*** jbadiapa has quit IRC | 07:27 | |
*** iurygregory has joined #openstack-oslo | 07:30 | |
*** ralonsoh has joined #openstack-oslo | 07:50 | |
*** altlogbot_3 has joined #openstack-oslo | 08:01 | |
*** altlogbot_3 has quit IRC | 08:04 | |
*** altlogbot_2 has joined #openstack-oslo | 08:11 | |
gsantomaggio | @bnemec @hberaud | 08:14 |
gsantomaggio | good morning | 08:14 |
gsantomaggio | I'd like to add an "example" folder inside olso.messaging | 08:14 |
gsantomaggio | does it make sense to you? | 08:14 |
gsantomaggio | I'd start with mandatory stuff | 08:15 |
*** altlogbot_2 has quit IRC | 08:16 | |
*** altlogbot_1 has joined #openstack-oslo | 08:17 | |
*** altlogbot_1 has quit IRC | 08:22 | |
*** altlogbot_0 has joined #openstack-oslo | 08:23 | |
*** altlogbot_0 has quit IRC | 08:29 | |
*** altlogbot_0 has joined #openstack-oslo | 08:53 | |
*** altlogbot_0 has quit IRC | 08:58 | |
*** irclogbot_1 has joined #openstack-oslo | 08:59 | |
*** irclogbot_1 has quit IRC | 09:04 | |
stephenfin | gsantomaggio: Makes sense to me, though I'd personally like to see some documentation around it too | 09:27 |
stephenfin | Perhaps making extensive use of the 'literalinclude' directive | 09:28 |
gsantomaggio | @stephenfin yes, the idea is to add also some documentation inside the examples | 09:29 |
stephenfin | Go for it. The idea seems good to me and we can decide of the specifics of the approach when reviewing | 09:29 |
gsantomaggio | 👍 | 09:34 |
*** irclogbot_1 has joined #openstack-oslo | 10:31 | |
*** irclogbot_1 has quit IRC | 10:38 | |
*** irclogbot_2 has joined #openstack-oslo | 10:42 | |
*** irclogbot_2 has quit IRC | 10:44 | |
*** altlogbot_3 has joined #openstack-oslo | 11:03 | |
*** altlogbot_3 has quit IRC | 11:08 | |
*** stephenfin has quit IRC | 11:08 | |
*** tesseract has joined #openstack-oslo | 11:08 | |
*** stephenfin has joined #openstack-oslo | 11:10 | |
*** altlogbot_2 has joined #openstack-oslo | 11:12 | |
*** altlogbot_2 has quit IRC | 11:16 | |
*** boden has joined #openstack-oslo | 11:50 | |
*** altlogbot_3 has joined #openstack-oslo | 12:08 | |
*** altlogbot_3 has quit IRC | 12:08 | |
*** goldyfruit has quit IRC | 12:22 | |
*** kgiusti has joined #openstack-oslo | 12:30 | |
*** raildo has joined #openstack-oslo | 13:03 | |
*** Luzi has quit IRC | 13:07 | |
*** irclogbot_3 has joined #openstack-oslo | 13:36 | |
*** irclogbot_3 has quit IRC | 13:38 | |
gsantomaggio | @bnemec @kgiusti @hberaud @stephenfin | 13:44 |
gsantomaggio | Currently the "mandatory" implementation is only of RPCClient. | 13:44 |
gsantomaggio | It should be added also server side on the class RPCServer. | 13:44 |
gsantomaggio | `RPCClient` is "easy" to handle the `RPCServer` is a more complex. | 13:44 |
gsantomaggio | So at the moment I'd leave the mandatory only to `RPCClient` then we will see. | 13:44 |
gsantomaggio | WDYT? | 13:44 |
*** goldyfruit has joined #openstack-oslo | 13:49 | |
kgiusti | gsantomaggio: on the server side we don't have a mechanism in the API to alert the application that an rpc reply failed. In that case the server logs an error and recovers. | 13:50 |
kgiusti | gsantomaggio: that being said, we definitely don't want a missing reply q to block the server (DOS in effect). | 13:51 |
kgiusti | gsantomaggio: using the mandatory flag on the server would (I think) help both the error logging and faster recovery from missing reply queues (I would think). | 13:52 |
kgiusti | gsantomaggio: so IMHO we should at least POC that. But I agree maybe we hold off releasing it until RPCClient mandatory stuff is released and we get some feedback on it. | 13:54 |
kgiusti | gsantomaggio: it's really up to you - what do you want to do next :) ? | 13:55 |
gsantomaggio | @kgiusti > using the mandatory flag on the server would (I think) help both the error logging and faster recovery from missing reply queues (I would think). | 13:58 |
gsantomaggio | yes, it will help | 13:58 |
gsantomaggio | @kgiusti > so IMHO we should at least POC that. | 13:58 |
gsantomaggio | I was working on it, so will make a POC | 13:58 |
kgiusti | gsantomaggio: +1 | 13:59 |
gsantomaggio | @kgiusti ok let me see how hard is to introduce it on `RPCServer` will let you know. | 14:00 |
gsantomaggio | Adding a log server side would be a very good point to start, in my opinion, just to have an idea about the number of the fails | 14:00 |
kgiusti | gsantomaggio: yeah, logging seems the Best We Can do. No one IIRC has asked for an API change to hook those failures so I'd rather not add complexity unless there's a clear demand for it (my opinion only!) | 14:02 |
*** irclogbot_3 has joined #openstack-oslo | 14:10 | |
gsantomaggio | @kgiusti if we want to have the same `RPCClient` flexibility we need to do something like: | 14:11 |
gsantomaggio | ``` | 14:11 |
gsantomaggio | options = oslo_messaging.TransportOptions(at_least_once=True) | 14:11 |
gsantomaggio | 14:11 | |
gsantomaggio | server = oslo_messaging.get_rpc_server(transport, target, endpoints, | 14:11 |
gsantomaggio | executor='threading', transport_options=options) | 14:11 |
gsantomaggio | ``` | 14:11 |
gsantomaggio | 14:11 | |
*** FlorianFa has quit IRC | 14:11 | |
kgiusti | gsantomaggio: +1 consistent with client, looks good | 14:12 |
gsantomaggio | @kgiusti 😥 I knew | 14:13 |
gsantomaggio | :D :D let me see how to do that | 14:13 |
kgiusti | :D | 14:13 |
*** altlogbot_1 has joined #openstack-oslo | 14:14 | |
*** lpetrut has joined #openstack-oslo | 14:26 | |
openstackgerrit | Vadym Markov proposed openstack/oslo.policy master: Correctly handle IO errors at policy file load https://review.opendev.org/670571 | 14:27 |
*** bnemec is now known as beekneemech | 14:34 | |
*** boden has quit IRC | 14:38 | |
*** hoonetorg has quit IRC | 14:40 | |
openstackgerrit | Scott Little proposed openstack/oslo.tools master: filter_git_history.sh passes broken parent-filter to git filter-branch https://review.opendev.org/669709 | 14:53 |
*** hoonetorg has joined #openstack-oslo | 14:56 | |
openstackgerrit | Scott Little proposed openstack/oslo.tools master: filter_git_history.sh passes broken parent-filter to git filter-branch https://review.opendev.org/669709 | 14:57 |
*** boden has joined #openstack-oslo | 14:57 | |
*** ansmith has quit IRC | 15:23 | |
*** iurygregory is now known as skolt | 15:27 | |
openstackgerrit | caoyuan proposed openstack/oslo.service master: Move doc related modules to doc/requirements.txt https://review.opendev.org/669408 | 15:53 |
*** lpetrut has quit IRC | 15:53 | |
*** ansmith has joined #openstack-oslo | 15:57 | |
openstackgerrit | Vadym Markov proposed openstack/oslo.policy master: Correctly handle IO errors at policy file load https://review.opendev.org/670571 | 16:04 |
*** ralonsoh has quit IRC | 17:19 | |
*** irclogbot_3 has quit IRC | 18:49 | |
*** edmondsw_ has quit IRC | 18:49 | |
*** irclogbot_3 has joined #openstack-oslo | 18:52 | |
*** tesseract has quit IRC | 19:33 | |
*** skolt has quit IRC | 19:39 | |
*** kgiusti has left #openstack-oslo | 20:32 | |
*** goldyfruit has quit IRC | 20:35 | |
*** boden has quit IRC | 20:59 | |
*** pcaruana has quit IRC | 20:59 | |
*** raildo has quit IRC | 21:01 | |
*** ianychoi has quit IRC | 21:54 | |
*** tosky has quit IRC | 23:20 | |
*** bobh has joined #openstack-oslo | 23:41 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!