)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":30449,"name":"jinyuanliu","email":"liujinyuan@inspur.com","username":"jinyuanliu"},"change_message_id":"1eb1b07ffec89c64d08ac7eb9e7d301685cdb0f6","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"c6752cdb_0c97f413","updated":"2022-03-15 09:28:45.000000000","message":"I think it\u0027s more appropriate to have these. A single node also runs wsrep processes.\nthis seems unnecessary.","commit_id":"6561d96d62b5390b05524b36499d04015ab1de94"},{"author":{"_account_id":34520,"name":"Sergiy Markin","email":"smarkin@mirantis.com","username":"sm515x"},"change_message_id":"1921185a6ca9dd414633a67f51f21dd9789f6fec","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":3,"id":"1c6ada13_f11a8242","updated":"2022-03-15 15:01:08.000000000","message":"Please review this patch.","commit_id":"6561d96d62b5390b05524b36499d04015ab1de94"},{"author":{"_account_id":8898,"name":"Chris Wedgwood","email":"cw@f00f.org","username":"anticw"},"change_message_id":"6e760168e3f5edb93c28f5605f5c3cb39a491524","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"cf310c36_3585ce58","in_reply_to":"a0de6a89_33231657","updated":"2022-03-17 02:31:16.000000000","message":"Done","commit_id":"6561d96d62b5390b05524b36499d04015ab1de94"},{"author":{"_account_id":34520,"name":"Sergiy Markin","email":"smarkin@mirantis.com","username":"sm515x"},"change_message_id":"6d7d1f935d5647ac8695fd844d96d6ab69d865da","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":3,"id":"a0de6a89_33231657","in_reply_to":"c6752cdb_0c97f413","updated":"2022-03-15 15:20:14.000000000","message":"We noticed that the single-node version of galera-enabled cluster takes longer time to start and initialize. Our approach to work it around is to disable galera functionality in our deployments. And this readiness.sh does not let us to leverage this approach. So the proposed change makes transparent adjustment to this check. The idea behind this is simple - in case of multiple instances it is vital to have the galera enabled. In case of single instance - galera functional is not a vital part, so the readiness check should succeed even if galera is not enabled.","commit_id":"6561d96d62b5390b05524b36499d04015ab1de94"},{"author":{"_account_id":8898,"name":"Chris Wedgwood","email":"cw@f00f.org","username":"anticw"},"change_message_id":"6e760168e3f5edb93c28f5605f5c3cb39a491524","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"c2cdb499_b5599556","updated":"2022-03-17 02:31:16.000000000","message":"I think removing wsrep and the other overhead from the replicas \u003d 1 is the right thing to do.\n\nFor complicated reasons this doesn\u0027t work right, and as it\u0027s not useful we don\u0027t need it.  If you were to scale (render again, not k8s scale verb) the wsrep status would return and behave as expected.\n\nFWIW, I have labs which are failing right now because of complexities from this very overhead, so I would like to see this merged (or something like it) to address that.\n\nBeyond that, we can talk about what should be done.  If there is contention here I\u0027m happy to setup a call (ping me on irc about this and I\u0027ll schedule something).","commit_id":"b569693c73ecd51c5dc13dc712e0ce9dcc0fc72c"},{"author":{"_account_id":18250,"name":"Roy Tang","email":"roy.s.tang@att.com","username":"rt7380"},"change_message_id":"86b591144f673562f345542ef76a8898038e57aa","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"d5915562_2dfa1086","updated":"2022-03-18 16:41:16.000000000","message":"I am ok to simply \"ignore\" wsrep status in a single node setup I suppose, it\u0027s really kind of useless to check for something that doesn\u0027t provide much.","commit_id":"848f392b3a56b02e135a5df4579b9599ddbbc5f9"},{"author":{"_account_id":21420,"name":"Gage Hugo","email":"gagehugo@gmail.com","username":"ghugo"},"change_message_id":"c3ea4b7fab07ebdfae54f2f98462d97976d1b684","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"b8ba66be_fd2da415","updated":"2022-03-18 16:38:02.000000000","message":"I can confirm we saw issues when scaling down to 1 replica and the clustering overhead would still cause issues.","commit_id":"848f392b3a56b02e135a5df4579b9599ddbbc5f9"},{"author":{"_account_id":18250,"name":"Roy Tang","email":"roy.s.tang@att.com","username":"rt7380"},"change_message_id":"a9050cbecbc938934b9c28c7b6b9319b5f7bfd1d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"8fdacbb1_7ab44e97","updated":"2022-03-18 16:43:25.000000000","message":"downgrading my vote a bit... I just worried in case someone manually scale up the deployment, then we lose the checking on wsrep, this may or may not break anything, but I would have like to hear if this is ok for everyone.","commit_id":"848f392b3a56b02e135a5df4579b9599ddbbc5f9"},{"author":{"_account_id":21420,"name":"Gage Hugo","email":"gagehugo@gmail.com","username":"ghugo"},"change_message_id":"5d0884195f3e23905206baf943b3b868fbdcc380","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"848cd265_8c03dc16","updated":"2022-03-18 17:40:25.000000000","message":"recheck","commit_id":"848f392b3a56b02e135a5df4579b9599ddbbc5f9"}]}