*** x00350071 has joined #openstack-smaug | 00:28 | |
*** x00350071_ has joined #openstack-smaug | 00:32 | |
*** c00281451 has quit IRC | 01:31 | |
openstackgerrit | yinwei proposed openstack/smaug: Basic design doc for Protection Service https://review.openstack.org/261455 | 02:08 |
---|---|---|
*** yinweiphone has joined #openstack-smaug | 02:14 | |
*** yinweiphone has quit IRC | 02:23 | |
*** x00350071 has quit IRC | 04:04 | |
*** x00350071 has joined #openstack-smaug | 04:04 | |
*** x00350071 has quit IRC | 05:50 | |
*** x00350071 has joined #openstack-smaug | 05:51 | |
*** gampel has joined #openstack-smaug | 06:38 | |
*** gampel1 has joined #openstack-smaug | 07:07 | |
*** gampel has quit IRC | 07:10 | |
*** gampel has joined #openstack-smaug | 07:11 | |
*** gampel1 has quit IRC | 07:14 | |
*** gampel has quit IRC | 07:47 | |
*** gampel has joined #openstack-smaug | 07:56 | |
openstackgerrit | chenying proposed openstack/smaug: Introduce oslo.versionedobjects to smaug https://review.openstack.org/271189 | 07:58 |
*** gampel has quit IRC | 07:58 | |
*** x00350071_ has quit IRC | 08:07 | |
openstackgerrit | chenying proposed openstack/smaug: Introduce oslo.versionedobjects to smaug https://review.openstack.org/271189 | 09:46 |
*** yinweimac has joined #openstack-smaug | 12:41 | |
yinweimac | saggi? | 12:42 |
yinweimac | can i ask what's the difference between get_prev_task and get_post_task? | 12:43 |
saggi | the former is when you visit the node and post when you leave | 12:44 |
saggi | yinweimac: ^^^^^ | 12:44 |
yinweimac | hmm, per my understanding, on entering the node, we usually make task flow of this node | 12:45 |
yinweimac | on leaving the node, which means we have visited all of its children, then we build dependency between its own task flow to the task flow returned by its chidlren nodes. | 12:46 |
yinweimac | does get_post_task involve making new tasks? | 12:47 |
yinweimac | I mean, is it possible that get_post_task needs children tasks as its input? | 12:51 |
yinweimac | say, for restoration, vm node can build vm_task, and its children vol1 node could build vol1_task | 12:52 |
yinweimac | when the node is leaving, the vm node needs create an attach_vol_task and compose the 3 as vm_task, vol1_task, attach_vol_task->vm_task, attach_vol_task->vm_task | 12:53 |
yinweimac | it seems to me that only one function of get_task_flow(children_task_flow) would be enough. | 12:54 |
yinweimac | i don't see the difference between enter node and leave node | 12:56 |
yinweimac | only the dependency on the tasks matters. whether you create task before or after the children node traverse doesn't matter, right? | 12:57 |
yinweimac | Pls. let me know your idea when you're available. | 12:57 |
*** zhonghua-lee has quit IRC | 13:08 | |
*** zhonghua-lee has joined #openstack-smaug | 13:09 | |
*** yinweimac has quit IRC | 13:22 | |
*** saggi has quit IRC | 13:50 | |
*** yinweimac has joined #openstack-smaug | 14:40 | |
*** yinweimac has quit IRC | 14:42 | |
openstackgerrit | Saggi Mizrahi proposed openstack/smaug: Pluggable protection provider doc https://review.openstack.org/262264 | 16:56 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!