*** mat128 is now known as mat128|mtg | 16:35 | |
ntpttr | lin_yang: have a moment? | 20:14 |
---|---|---|
ntpttr | I think I just found one possible reason we might want to at least be aware of the redfish assigned UUID for nodes | 20:15 |
ntpttr | working on the manage command, it's possible to run it on a node you already have saved in the database, which results in having the same node in the DB twice with two different valence-assigned UUIDs | 20:16 |
ntpttr | lin_yang: *or* we could use the index value, with the caveat that we will need to be checking only against nodes managed by the same pod manager | 20:32 |
ntpttr | I think it would be good to get that info stored in the DB for resources anyways - index, uuid, and which pod manager are they managed by | 20:32 |
ntpttr | in the form of a UUID so that all pod manager IDs are unique | 20:32 |
lin_yang | just finished the lunch | 21:32 |
lin_yang | yep, the podm uuid + index should be unique | 21:32 |
lin_yang | and the corresponding podm uuid should be store in db also as you mentioned, I remember chester update this on api spec | 21:35 |
lin_yang | https://review.openstack.org/#/c/410570/5/api-ref/source/valence-api-v1-nodes.inc | 21:37 |
lin_yang | How about we add podm uuid in another commit, combine with support compose node on multiply podm? | 21:40 |
openstackgerrit | Lin Yang proposed openstack/valence master: Support post action to composed node https://review.openstack.org/428403 | 21:46 |
*** ramineni has quit IRC | 22:20 | |
*** ramineni has joined #openstack-valence | 22:21 | |
*** ramineni has quit IRC | 22:50 | |
*** ramineni has joined #openstack-valence | 22:51 | |
ntpttr | lin_yang: Sounds good, a different commit would be best | 23:44 |
ntpttr | running into a bit of an issue with my patch at the moment with node listing not showing index value | 23:45 |
lin_yang | if need to any update for current node related function, please let me know | 23:48 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!