*** lxkong has quit IRC | 01:48 | |
congnt95 | lxkong: I think we should allow user create replica with flavor, because replica maybe just only backup, it don't need a large vcpu and ram. What are you think? | 02:23 |
---|---|---|
congnt95 | https://github.com/openstack/trove/blob/master/trove/instance/service.py#L370 | 02:23 |
*** spatel has joined #openstack-trove | 02:38 | |
*** spatel has quit IRC | 02:42 | |
-openstackstatus- NOTICE: The Gerrit service on review.opendev.org is being restarted quickly to enable support for Git protocol v2, downtime should be less than 5 minutes | 02:55 | |
*** lifeless has quit IRC | 03:31 | |
*** lifeless has joined #openstack-trove | 03:33 | |
*** rcernin has quit IRC | 03:39 | |
*** rcernin has joined #openstack-trove | 03:40 | |
*** lxkong has joined #openstack-trove | 06:16 | |
lxkong | congnt95: we intentionally set the same flavor for replicas as the primary to avoid the situation that in particular around configuration the user may set (eg sort area) that may not work/fit a smaller replica | 06:23 |
lxkong | congnt95: and same sizes gives known predictable outcomes. | 06:23 |
lxkong | congnt95: as you know, databases have configuration settings that impact how much memory they allocate and how the memory is utilised to speed up database operations (such as caching data in RAM or having more memory space to sort data) | 06:24 |
lxkong | when using a smaller flavor for replicas, there is a chance that the replica won't come up | 06:26 |
lxkong | being that said, the user is still able to resize the replica at their own discretion | 06:30 |
lxkong | congnt95: does that make sense? | 06:30 |
congnt95 | lxkong: i still can't understand why we allow user resize replica, but not allow user create replica with different flavor. Because, if user resize replica to smaller and with configuration, it may still not work anymore. | 07:38 |
*** rcernin has quit IRC | 08:07 | |
*** rcernin has joined #openstack-trove | 08:14 | |
*** tosky has joined #openstack-trove | 08:47 | |
*** openstackgerrit has joined #openstack-trove | 08:49 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/trove-dashboard master: Imported Translations from Zanata https://review.opendev.org/c/openstack/trove-dashboard/+/766687 | 08:49 |
lxkong | congnt95: as I said above, the cloud provider should be using the recommended way to set up the replication cluster by default, but still provide API for the users to make changes by themselves at their own risk. | 08:52 |
lxkong | congnt95: and FYI, Azure is doing that in a similar way, see https://docs.microsoft.com/en-us/azure/mysql/concepts-read-replicas#replica-configuration | 08:53 |
*** rcernin has quit IRC | 09:27 | |
*** e0ne has joined #openstack-trove | 10:04 | |
openstackgerrit | Merged openstack/trove-dashboard master: Imported Translations from Zanata https://review.opendev.org/c/openstack/trove-dashboard/+/766687 | 10:30 |
*** e0ne has quit IRC | 10:34 | |
*** e0ne has joined #openstack-trove | 10:48 | |
openstackgerrit | Merged openstack/trove master: Disable some tempest tests that always failed in CI https://review.opendev.org/c/openstack/trove/+/765930 | 11:57 |
*** lxkong has quit IRC | 12:56 | |
*** congnt96 has joined #openstack-trove | 13:36 | |
*** rcernin has joined #openstack-trove | 15:24 | |
*** rcernin has quit IRC | 15:28 | |
*** e0ne has quit IRC | 16:53 | |
*** tosky has quit IRC | 17:46 | |
*** __ministry1 has joined #openstack-trove | 17:49 | |
*** __ministry1 has quit IRC | 18:08 | |
*** congnt96 has quit IRC | 18:13 | |
*** e0ne has joined #openstack-trove | 18:54 | |
*** e0ne has quit IRC | 19:08 | |
*** e0ne has joined #openstack-trove | 19:29 | |
*** e0ne has quit IRC | 19:30 | |
*** e0ne has joined #openstack-trove | 19:45 | |
*** e0ne has quit IRC | 19:47 | |
*** e0ne has joined #openstack-trove | 19:48 | |
*** lxkong has joined #openstack-trove | 20:24 | |
*** e0ne has quit IRC | 21:02 | |
*** tosky has joined #openstack-trove | 21:05 | |
*** lxkong has quit IRC | 22:32 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!