*** brinzhang has joined #openstack-kolla | 00:13 | |
openstackgerrit | Michal Arbet proposed openstack/kolla master: Add proxysql image https://review.opendev.org/c/openstack/kolla/+/769385 | 00:41 |
---|---|---|
*** LinPeiWen has quit IRC | 01:51 | |
*** LinPeiWen has joined #openstack-kolla | 02:03 | |
*** kevko has quit IRC | 02:07 | |
*** wuchunyang has joined #openstack-kolla | 02:45 | |
*** skramaja has joined #openstack-kolla | 03:13 | |
*** ricolin has quit IRC | 03:19 | |
*** wuchunyang has quit IRC | 04:00 | |
*** zzzeek has quit IRC | 04:06 | |
*** zzzeek has joined #openstack-kolla | 04:07 | |
*** ricolin_ has joined #openstack-kolla | 04:17 | |
*** brinzhang has quit IRC | 05:29 | |
*** brinzhang has joined #openstack-kolla | 05:29 | |
*** vishalmanchanda has joined #openstack-kolla | 05:35 | |
*** wuchunyang has joined #openstack-kolla | 05:35 | |
*** cah_link has joined #openstack-kolla | 06:01 | |
*** zzzeek has quit IRC | 07:02 | |
*** ivan_lin has joined #openstack-kolla | 07:03 | |
*** devfaz has joined #openstack-kolla | 07:04 | |
*** zzzeek has joined #openstack-kolla | 07:07 | |
*** bsanjeewa has joined #openstack-kolla | 07:20 | |
*** nikparasyr has joined #openstack-kolla | 07:27 | |
*** sorin-mihai__ has joined #openstack-kolla | 07:30 | |
*** dave-mccowan has quit IRC | 07:30 | |
*** hjensas has quit IRC | 07:59 | |
*** rpittau|afk is now known as rpittau | 08:08 | |
*** cah_link has quit IRC | 08:09 | |
*** kevko has joined #openstack-kolla | 08:09 | |
*** gfidente|afk is now known as gfidente | 08:13 | |
*** andrewbonney has joined #openstack-kolla | 08:17 | |
*** bsanjeewa has quit IRC | 08:20 | |
*** bengates has joined #openstack-kolla | 08:20 | |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Edit services roles to support database sharding https://review.opendev.org/c/openstack/kolla-ansible/+/770216 | 08:26 |
*** hjensas has joined #openstack-kolla | 08:35 | |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Add proxysql support for database https://review.opendev.org/c/openstack/kolla-ansible/+/770215 | 08:45 |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Edit services roles to support database sharding https://review.opendev.org/c/openstack/kolla-ansible/+/770216 | 08:45 |
*** bengates has quit IRC | 08:47 | |
*** bengates has joined #openstack-kolla | 08:47 | |
*** dougsz has joined #openstack-kolla | 08:49 | |
*** cah_link has joined #openstack-kolla | 08:50 | |
*** e0ne has joined #openstack-kolla | 08:53 | |
*** cah_link has quit IRC | 08:57 | |
*** cah_link has joined #openstack-kolla | 09:01 | |
*** cah_link has quit IRC | 09:05 | |
*** maharg101 has joined #openstack-kolla | 09:05 | |
*** cah_link has joined #openstack-kolla | 09:07 | |
kevko | morning ! :) | 09:20 |
kevko | i'm probably done with proxysql/mariadb services/sharding thing :) -> https://review.opendev.org/q/topic:%22proxysql%22+(status:open%20OR%20status:merged) | 09:21 |
hrw | https://review.opendev.org/c/openstack/kolla/+/755222 anyone? yoctozepto? | 09:26 |
*** mgoddard has joined #openstack-kolla | 09:32 | |
openstackgerrit | Mark Goddard proposed openstack/kolla stable/victoria: CI: revert to public package mirrors after build https://review.opendev.org/c/openstack/kolla/+/770088 | 09:33 |
openstackgerrit | Mark Goddard proposed openstack/kolla stable/victoria: DNM: testing mirror revert https://review.opendev.org/c/openstack/kolla/+/770091 | 09:33 |
hrw | kevko: https://review.opendev.org/c/openstack/kolla/+/770217 https://review.opendev.org/c/openstack/kolla/+/770219 https://review.opendev.org/c/openstack/kolla/+/769385 https://review.opendev.org/c/openstack/kolla/+/769385 - decide on one please | 09:34 |
kevko | that was pushed by mistake ... but i abandoned it | 09:36 |
hrw | kevko: I am going to -1 it with info why | 09:38 |
*** ricolin_ has quit IRC | 09:39 | |
kevko | to which review ? | 09:39 |
hrw | active one | 09:39 |
kevko | only one proxysql image is active and +1 from zuul | 09:39 |
*** cah_link1 has joined #openstack-kolla | 09:39 | |
*** cah_link has quit IRC | 09:40 | |
*** cah_link1 is now known as cah_link | 09:40 | |
hrw | kevko: aarch64 ci jobs have retry_limit again ;( | 09:40 |
kevko | hrw: I usually draw from others docker images in tree :) , so, for me as a template was keepalived ..proxysql has aarch64 packages | 09:44 |
hrw | kevko: not added in repos.yaml | 09:45 |
kevko | ah, I see | 09:45 |
kevko | w8, adding | 09:45 |
hrw | ;) | 09:47 |
kevko | btw, should I reorder entries in repos.yaml if I see that they are not alphabeticall correct ? | 09:49 |
hrw | I would not complain | 09:50 |
kevko | or leave it as it is and just add my change ? | 09:50 |
hrw | sort them | 09:50 |
kevko | ok | 09:50 |
openstackgerrit | Michal Arbet proposed openstack/kolla master: Add proxysql image https://review.opendev.org/c/openstack/kolla/+/769385 | 09:52 |
kevko | btw, I am getting this in one of my review -> "This change depends on a change that failed to merge." ... how can i know what happened ? which dependent change failed to merge ? | 09:54 |
kevko | can't see any log | 09:57 |
hrw | rebase? | 09:58 |
kevko | hmm, going to try | 10:07 |
kevko | is Depends-on used only when cross dependency ? or i should add depends-on on previous commit in chain ? | 10:08 |
kevko | and could you help me with test job to zuul ? i would like to test switch to proxysql | 10:09 |
*** wuchunyang has quit IRC | 10:17 | |
*** Fl1nt has joined #openstack-kolla | 10:20 | |
Fl1nt | Hi everyone! | 10:20 |
Fl1nt | I hope you all had wonderful holydays! | 10:20 |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Add proxysql support for database https://review.opendev.org/c/openstack/kolla-ansible/+/770215 | 10:21 |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Edit services roles to support database sharding https://review.opendev.org/c/openstack/kolla-ansible/+/770216 | 10:21 |
*** iniazi_ has joined #openstack-kolla | 10:29 | |
*** iniazi has quit IRC | 10:31 | |
hrw | kevko: only on cross. | 10:39 |
hrw | kevko: if you need depend in same project then chain | 10:39 |
kevko | hmm, so, i don't really understand why it fails to merge :( | 10:39 |
kevko | if I rebase to upstream master it is up to date :/ | 10:40 |
*** hjensas is now known as hjensas|afk | 10:51 | |
openstackgerrit | Michal Arbet proposed openstack/kolla master: Add support for checking more daemons https://review.opendev.org/c/openstack/kolla/+/765781 | 10:55 |
openstackgerrit | Michal Arbet proposed openstack/kolla master: Add proxysql image https://review.opendev.org/c/openstack/kolla/+/769385 | 10:56 |
*** emccormick has quit IRC | 11:06 | |
*** emccormick has joined #openstack-kolla | 11:06 | |
*** wuchunyang has joined #openstack-kolla | 11:06 | |
*** rpittau has quit IRC | 11:06 | |
*** rpittau has joined #openstack-kolla | 11:07 | |
openstackgerrit | Mark Goddard proposed openstack/kolla stable/ussuri: Remove footer block from intermediate images https://review.opendev.org/c/openstack/kolla/+/770086 | 11:10 |
openstackgerrit | Mark Goddard proposed openstack/kolla stable/ussuri: CI: revert to public package mirrors after build https://review.opendev.org/c/openstack/kolla/+/770089 | 11:11 |
openstackgerrit | Mark Goddard proposed openstack/kolla stable/ussuri: CI: revert to public package mirrors after build https://review.opendev.org/c/openstack/kolla/+/770089 | 11:12 |
openstackgerrit | Mark Goddard proposed openstack/kolla stable/ussuri: DNM: testing mirror revert https://review.opendev.org/c/openstack/kolla/+/770092 | 11:13 |
openstackgerrit | Mark Goddard proposed openstack/kolla stable/train: CI: revert to public package mirrors after build https://review.opendev.org/c/openstack/kolla/+/770090 | 11:15 |
openstackgerrit | Mark Goddard proposed openstack/kolla stable/train: DNM: testing mirror revert https://review.opendev.org/c/openstack/kolla/+/770093 | 11:15 |
*** e0ne has quit IRC | 11:16 | |
*** structz has joined #openstack-kolla | 11:18 | |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Add proxysql support for database https://review.opendev.org/c/openstack/kolla-ansible/+/770215 | 11:21 |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Edit services roles to support database sharding https://review.opendev.org/c/openstack/kolla-ansible/+/770216 | 11:21 |
*** LinPeiWen has quit IRC | 11:29 | |
*** ivan_lin has quit IRC | 11:30 | |
*** e0ne has joined #openstack-kolla | 11:39 | |
*** e0ne has quit IRC | 11:44 | |
*** Luzi has joined #openstack-kolla | 11:47 | |
*** e0ne has joined #openstack-kolla | 11:47 | |
*** e0ne has quit IRC | 11:54 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible stable/train: Remove the cpuset_cpus parameter for nova_libvirt container https://review.opendev.org/c/openstack/kolla-ansible/+/770275 | 12:00 |
*** ricolin_ has joined #openstack-kolla | 12:03 | |
openstackgerrit | wu.chunyang proposed openstack/kolla-ansible master: octavia: support tenant management network https://review.opendev.org/c/openstack/kolla-ansible/+/755589 | 12:03 |
*** wuchunyang has quit IRC | 12:04 | |
*** structz has quit IRC | 12:07 | |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Add proxysql support for database https://review.opendev.org/c/openstack/kolla-ansible/+/770215 | 12:08 |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Edit services roles to support database sharding https://review.opendev.org/c/openstack/kolla-ansible/+/770216 | 12:08 |
*** cah_link1 has joined #openstack-kolla | 12:32 | |
*** cah_link has quit IRC | 12:36 | |
*** cah_link1 is now known as cah_link | 12:36 | |
*** ricolin_ has quit IRC | 12:47 | |
kevko | hrw: succeded ..is it now OK ? | 12:48 |
openstackgerrit | Pedro Henrique Pereira Martins proposed openstack/kolla-ansible master: Add support to OpenID Connect Authentication flow https://review.opendev.org/c/openstack/kolla-ansible/+/695432 | 12:51 |
openstackgerrit | Piotr Parczewski proposed openstack/kolla-ansible master: Add Prometheus 2.x deployment https://review.opendev.org/c/openstack/kolla-ansible/+/765973 | 13:19 |
Fl1nt | team, what's the reasoning behind adding proxysql to the roles? I mean, to my knowledge none of the openstack services support sharding. It seems like we won't really benefit any big improvement over using haproxy. | 13:24 |
Fl1nt | MariaDB being natively Active/Active multi master and because you can set kolla-ansible to deploy your HA to be Active/Active (Keepalived + HAProxy) what kind of feature improvement do we gain from it? | 13:25 |
Fl1nt | additionally, that is going against CELL layout as you're supposed to get one DB per cell which is kind of a sharding already. | 13:28 |
openstackgerrit | Merged openstack/kolla-ansible master: Fix remove libvirt apparmor disabled profile https://review.opendev.org/c/openstack/kolla-ansible/+/768885 | 13:35 |
Fl1nt | at least at nova level. | 13:36 |
hrw | I have nothing against proxysql image in kolla (have to check license etc first). | 13:43 |
hrw | will it replace haproxy in k-a is a thing someone else has to decide | 13:43 |
*** wuchunyang has joined #openstack-kolla | 13:45 | |
openstackgerrit | Merged openstack/kolla-ansible master: Register resources info for octavia upgrading https://review.opendev.org/c/openstack/kolla-ansible/+/746005 | 13:46 |
openstackgerrit | Merged openstack/kolla-ansible master: Use Docker healthchecks for cloudkitty services https://review.opendev.org/c/openstack/kolla-ansible/+/768039 | 13:46 |
openstackgerrit | Merged openstack/kolla-ansible master: Fix dpdk deploy failed https://review.opendev.org/c/openstack/kolla-ansible/+/768510 | 13:46 |
kevko | guys, let's say i introduced change (proxysql) which is changing the way how kolla-ansible deployment works by default depends on some values (enable_proxysql) .. how I can test in zuul that if original and new solution works both ? | 13:48 |
kevko | simply said , i need to test scenario with enable_proxysql: false (haproxy enabled) working , and reconfigure with enable_proxysql: true - working | 13:49 |
openstackgerrit | wu.chunyang proposed openstack/kolla-ansible master: octavia: support tenant management network https://review.opendev.org/c/openstack/kolla-ansible/+/755589 | 13:55 |
*** ricolin_ has joined #openstack-kolla | 13:58 | |
*** ricolin_ has quit IRC | 13:58 | |
*** ricolin has joined #openstack-kolla | 13:59 | |
yoctozepto | kevko: well, the easiest would be to base on one (or more) upgrade jobs | 13:59 |
hrw | the more I look at oslo_config the less I understand it | 14:00 |
kevko | yoctozepto: could you please refer me to some example upgrade job which i can retrace ? | 14:01 |
yoctozepto | hrw: in what sense? | 14:03 |
hrw | yoctozepto: it is easier to add cli option than getting it read entry from config file | 14:03 |
Fl1nt | hrw, I'm not against proxysql, tbh, it's just a question about does it worth it to be integrated if it doesn't bring any value? | 14:04 |
hrw | oslo_config.cfg.NoSuchOptError: no such option allowed_to_fail2 in group [DEFAULT] | 14:04 |
hrw | [DEFAULT] | 14:04 |
hrw | allowed_to_fail2 = test | 14:04 |
yoctozepto | kevko: wdym? logs? or path to scripts? | 14:04 |
yoctozepto | hrw: odd, it usually just works with both sources :S | 14:05 |
yoctozepto | Fl1nt: proxysql is much smarter about down sql backends | 14:05 |
yoctozepto | we can get rid of clumsy haproxy checks | 14:06 |
yoctozepto | that is one benefit | 14:06 |
hrw | yoctozepto: "--allowed-to-fail 'test test2'" works ;P | 14:07 |
*** kevko has quit IRC | 14:08 | |
yoctozepto | hrw: well, there is the "2" in one of them | 14:09 |
hrw | ok. got it working | 14:10 |
yoctozepto | :-) | 14:14 |
Fl1nt | yoctozepto, define much... we have few of them, they're just natively support sharding and L7 ruling, which isn't supported on HAProxy but isn't honestly required anywhere else than on Application traffic. | 14:17 |
Fl1nt | Even with a multi region, multi continent, multi cell structure our study on proxysql vs haproxy didn't clearly set a performance or operational advantage ^^ hence why I talked about value added | 14:17 |
yoctozepto | Fl1nt: that's interesting; performance should not really be the main concern in here but I thought our haproxy check for sql backend being up leaves much to desire compared to it being native in proxysql | 14:18 |
Fl1nt | on this specific topic, of course we could improve the MariaDB Cluster check a bit, but all in all, replacing a simple working and efficient process by adding yet another layer is just freaking me out as I know from experience what staking layers is producing ^^ | 14:20 |
Fl1nt | ask Ubisoft about that for instance :p | 14:20 |
Fl1nt | all in all, adding a role that is optional isn't bad in it ^^ | 14:21 |
Fl1nt | and can even be something useful for one specific kolla ops use case, but it need to be maintained, not to be the defacto default option and doesn't really (to my opinion, but I'll be glad to be proved wrong) a lot of value added. | 14:22 |
yoctozepto | Fl1nt: it is not a layer; just a smarter proxy to sql | 14:23 |
yoctozepto | haproxy is well-suited for proxying http | 14:23 |
yoctozepto | and just fine for general tcp | 14:24 |
Fl1nt | It is adding a layer as you put it behind the HAProxy whatever happens :D | 14:25 |
Fl1nt | Exemple of additional complexity for proxySQL: TLS management. | 14:25 |
yoctozepto | Fl1nt: it was supposed to be *instead* of haproxy there (kevko please correct me) | 14:25 |
Fl1nt | it can't | 14:25 |
yoctozepto | Fl1nt: haproxy can't tls for sql ;-) | 14:25 |
yoctozepto | Fl1nt: what can't what? :D | 14:26 |
Fl1nt | Good point for HAProxy TLS on MariaDB as it's TCP level. | 14:27 |
Fl1nt | if you get ride of HAProxy, how will you balance HTTP APIs services ? | 14:28 |
Fl1nt | how will you balance SQL requests on ProxySQL node 01 and 02/0x on a HA/HA install? | 14:30 |
Fl1nt | you could use Keepalived and DNS Round robin but... meh, it's not really efficient. | 14:30 |
hrw | 5 files changed, 53 insertions(+), 21 deletions(-) | 14:30 |
hrw | allowed to fail images are coming ;D | 14:30 |
Fl1nt | ?? | 14:30 |
Fl1nt | woot? | 14:31 |
hrw | Fl1nt: one of features for Wallaby release | 14:31 |
*** Luzi has quit IRC | 14:31 | |
Fl1nt | oh ok ^^ cool :D | 14:31 |
hrw | Fl1nt: we do not care if images x, y, z fail on CI | 14:32 |
yoctozepto | hrw: gooood | 14:32 |
Fl1nt | aaah nice ^^ | 14:32 |
yoctozepto | Fl1nt: not get rid of haproxy, just not use it for sql | 14:32 |
yoctozepto | Fl1nt: it is still there, just not as a layer for proxysql | 14:32 |
yoctozepto | that was my point :-) | 14:33 |
Fl1nt | then you're trading a fully redundant and HA solution to a HA using only active/passive. | 14:33 |
Fl1nt | except if someone managed to introduce active/active architecture on ProxySQL tbh I didn't checked lately ^^ | 14:34 |
Fl1nt | Here is another argument, most of time, what's overloaded isn't your API database but rather your messaging queue cluster or your CELL database. | 14:34 |
yoctozepto | Fl1nt: how so? proxysql is active/active as long as you have multiple galera replicas (as we do now) | 14:35 |
Fl1nt | kevko, I'm curious about your use case with proxySQL. | 14:35 |
yoctozepto | yeah, the performance argument is not that strong | 14:35 |
yoctozepto | I am buying it for saner backend failure discovery | 14:35 |
mnasiadka | Fl1nt: you don't have to use it, but it's nice that K-A users will have a choice after this functionality is merged :) | 14:35 |
yoctozepto | rather than anything else | 14:36 |
mnasiadka | yoctozepto: anything better than current xinetd + clustercheck script - I'm in :) | 14:36 |
Fl1nt | yoctozepto, nooo I talk about frontend side of things, yes your proxySQL service is able to balance/shard agains SQL instances, but if you put multiple proxySQL instance in front of your cluster then you get either a active/passive using keepalived or a active/active using keepalived with multiple IPs round robin with DNS and the last option is to get only one ProxySQL instance so SPOF. | 14:37 |
yoctozepto | mnasiadka: indeed! | 14:37 |
openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: WIP: implement "allowed to fail" images https://review.opendev.org/c/openstack/kolla/+/765807 | 14:38 |
Fl1nt | mnasiadka, I'm absolutely 100% with you on having options, however, I already see how hard the team is working to maintain kolla/k-a/kayobe, adding roles like that without discussing it could just lead to more and more dead code on the repo. | 14:38 |
yoctozepto | Fl1nt: we already have haproxy and a single VIP; forget that sharding and focus on replicas; maybe I am missing something here but I feel like you went down wrong rabbit hole | 14:38 |
Fl1nt | We install our cluster using DUAL HA VIP, so I may be biased yes. | 14:39 |
mnasiadka | Fl1nt: if it will be tested in CI properly, then it's not really dead | 14:39 |
Fl1nt | what about replicas @yoctozepto ? | 14:39 |
hrw | please take a look at allowed-to-fail images as it may need some improvements | 14:40 |
Fl1nt | you already get replicas as we deploy a multinode multi master mariadb cluster. Or I don't get what you're pointing out? | 14:41 |
yoctozepto | Fl1nt: I think I know what you wanted to say - that we don't use all proxysql at once | 14:41 |
Fl1nt | yes | 14:41 |
yoctozepto | but that's not getting worse from what we have | 14:41 |
Fl1nt | In many enterprise installation case it is. | 14:41 |
Fl1nt | as we deployed everything using keepalived multi VIP and so address that passive nodes making nothing useful except waiting for someone else to die ^^ | 14:42 |
Fl1nt | once again, I'm not against using proxySQL, but I'd like to hear from kevko about it's use case as for now, proxySQL is only really usefull when you're able to use it at application level to spread your requests against sharded DBs. | 14:44 |
Fl1nt | which isn't going to happen soon on Openstack services (to my limited knowledge once again). | 14:44 |
*** TrevorV has joined #openstack-kolla | 14:52 | |
*** sorin-mihai__ has quit IRC | 14:59 | |
*** wuchunyang has quit IRC | 15:05 | |
*** kevko has joined #openstack-kolla | 15:10 | |
*** skramaja has quit IRC | 15:49 | |
*** sorin-mihai__ has joined #openstack-kolla | 15:50 | |
openstackgerrit | Mark Goddard proposed openstack/kolla stable/train: Remove footer block from intermediate images https://review.opendev.org/c/openstack/kolla/+/770087 | 15:53 |
*** Fl1nt has quit IRC | 15:57 | |
openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: WIP: implement "allowed to fail" images https://review.opendev.org/c/openstack/kolla/+/765807 | 16:11 |
*** e0ne has joined #openstack-kolla | 16:18 | |
openstackgerrit | Mark Goddard proposed openstack/kolla master: Add a storm-base image https://review.opendev.org/c/openstack/kolla/+/770445 | 16:21 |
*** openstackgerrit has quit IRC | 16:55 | |
*** e0ne has quit IRC | 17:12 | |
*** sorin-mihai_ has joined #openstack-kolla | 17:14 | |
*** sorin-mihai__ has quit IRC | 17:16 | |
*** maharg101 has quit IRC | 17:27 | |
*** dougsz has quit IRC | 17:31 | |
*** rpittau is now known as rpittau|afk | 17:32 | |
*** e0ne has joined #openstack-kolla | 17:33 | |
*** e0ne has quit IRC | 17:35 | |
hrw | meh. | 17:42 |
hrw | monasca-thresh failed to fail when I wanted it to fail | 17:42 |
*** sorin-mihai has joined #openstack-kolla | 17:52 | |
*** sorin-mihai_ has quit IRC | 17:53 | |
*** sorin-mihai_ has joined #openstack-kolla | 17:57 | |
*** sorin-mihai has quit IRC | 18:01 | |
yoctozepto | hrw: sounds like an unfulfilled masochistic desire the way you put it | 18:10 |
*** nikparasyr has left #openstack-kolla | 18:13 | |
*** cah_link has quit IRC | 18:19 | |
*** gfidente is now known as gfidente|afk | 18:19 | |
*** bengates has quit IRC | 18:48 | |
*** ianw_pto is now known as ianw | 18:58 | |
*** andrewbonney has quit IRC | 19:07 | |
*** maharg101 has joined #openstack-kolla | 19:23 | |
*** maharg101 has quit IRC | 19:28 | |
*** e0ne has joined #openstack-kolla | 19:35 | |
*** e0ne has quit IRC | 19:39 | |
kevko | someone still here ? :) | 19:43 |
kevko | i have weird behaviour on zuul :/ | 19:44 |
*** sorin-mihai_ has quit IRC | 19:50 | |
*** sorin-mihai_ has joined #openstack-kolla | 19:50 | |
*** e0ne has joined #openstack-kolla | 20:05 | |
*** e0ne has quit IRC | 20:18 | |
*** e0ne has joined #openstack-kolla | 20:19 | |
*** e0ne has quit IRC | 20:34 | |
*** e0ne_ has joined #openstack-kolla | 20:34 | |
*** e0ne_ has quit IRC | 20:40 | |
*** vishalmanchanda has quit IRC | 20:49 | |
*** sorin-mihai_ has quit IRC | 20:50 | |
*** sorin-mihai has joined #openstack-kolla | 20:50 | |
*** TrevorV has quit IRC | 20:56 | |
*** openstackgerrit has joined #openstack-kolla | 21:00 | |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Refactor mariadb to support multiple separate shards https://review.opendev.org/c/openstack/kolla-ansible/+/766952 | 21:00 |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Add proxysql support for database https://review.opendev.org/c/openstack/kolla-ansible/+/770215 | 21:00 |
openstackgerrit | Michal Arbet proposed openstack/kolla-ansible master: Edit services roles to support database sharding https://review.opendev.org/c/openstack/kolla-ansible/+/770216 | 21:00 |
*** sorin-mihai_ has joined #openstack-kolla | 21:14 | |
*** hjensas|afk is now known as hjensas | 21:15 | |
*** sorin-mihai has quit IRC | 21:16 | |
*** sorin-mihai has joined #openstack-kolla | 21:17 | |
*** sorin-mihai_ has quit IRC | 21:19 | |
adeberg | I don't see how I could re-use that proxysql container for a custom service. A bit like the custom services patch that is available for HAproxy.... looks like it's pretty much thought with sharding only in mind. And yes Fl1nt, that's for my wan keystone-dedicated cluster before you ask | 21:22 |
*** maharg101 has joined #openstack-kolla | 21:24 | |
*** maharg101 has quit IRC | 21:29 | |
*** sorin-mihai_ has joined #openstack-kolla | 21:51 | |
*** sorin-mihai has quit IRC | 21:54 | |
*** sorin-mihai_ has quit IRC | 22:48 | |
*** sorin-mihai_ has joined #openstack-kolla | 22:49 | |
*** kevko has quit IRC | 23:22 | |
*** kevko has joined #openstack-kolla | 23:25 | |
*** sorin-mihai_ has quit IRC | 23:37 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!