Percona Cluster Down

hi all, i am using Percona cluster with three node(two database nodes, one arbiter node). After running smoothly one month, cluster down service down. now only one node up. when we try to add new node like arbiter, it can add to cluster but other db node can join to cluster. how can add this down node to cluster ?

Dear Concern,

Have a Good Day, We are using Percona Cluster with three nodes(two db server + one arbiter). One of the node was goes down one moth ago. Now only single node is running as primary. Too much inconsistency made between two databases. It was failed for out of memory(OOM).
How to rejoin this fault database in the cluster with full sst(rsync)?
How to prevent OOM in this cluster?

OS: ubuntu 16
RAM:24GB
DB Size: 18GB
CPU:24 core

Hello Isahak
To properly help you with this it is likely that the team will ask for these details:
[LIST]
[]precise version of Percona XtraDB Cluster
[
]precise details of your operating environment
[]the my.cnf for each of the three nodes
[
]any error logs or messages that you are seeing
[/LIST] Meanwhile, I found this post in case it is relevant in your case: https://www.percona.com/forums/questions-discussions/percona-xtradb-cluster/50752-cluster-crashing-daily-oom-killer

And also some blog posts, again in case relevant, the second is about postgres but looks at linux kernel settings too:
https://www.percona.com/blog/2018/06/28/what-to-do-when-mysql-runs-out-of-memory-troubleshooting-guide/
https://www.percona.com/blog/2019/08/02/out-of-memory-killer-or-savior/

If these don’t help, do feel free to post back with the details as requested and I will see if the team have any suggestions for you.