nebula 版本：3.0.2
部署方式：分布式
安装方式：RPM
是否为线上版本：N
硬件信息
磁盘 2T x 3
CPU 12核 x 3
内存 96G x 3
在华为云购买3台弹性云服务器EC2，搭建Nebula 3.0.2集群，nebula-graph.conf 配置文件中设置了内存水位线阈值，但是仍然发生Graph进程OOM。
system_memory_high_watermark_ratio 这个配置并没有杀掉session，无法保证 graphd进程不发生OOM吗？
########## memory ##########
# System memory high watermark ratio, cancel the memory checking when the ratio greater than 1.0
--system_memory_high_watermark_ratio=0.9
查看系统日志
[root@0 etc]# dmesg -T | grep nebula
[Wed Aug 10 20:40:09 2022] [1432744] 0 1432744 520384 87982 2957312 0 0 nebula-metad
[Wed Aug 10 20:40:09 2022] [2198418] 0 2198418 22180855 5306114 153513984 0 0 nebula-storaged
[Wed Aug 10 20:40:09 2022] [951088] 0 951088 183813 5831 221184 0 0 nebula-stats-ex
[Wed Aug 10 20:40:09 2022] [952680] 0 952680 590543 4460 364544 0 0 nebula-httpd
[Wed Aug 10 20:40:09 2022] [2226288] 0 2226288 21067337 10451643 136519680 0 0 nebula-graphd
[Wed Aug 10 20:40:09 2022] oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0-1,global_oom,task_memcg=/user.slice/user-0.slice/session-434.scope,task=nebula-graphd,pid=2226288,uid=0
[Wed Aug 10 20:40:09 2022] Out of memory: Killed process 2226288 (nebula-graphd) total-vm:84269348kB, anon-rss:41806572kB, file-rss:0kB, shmem-rss:0kB, UID:0
[Wed Aug 10 20:40:11 2022] oom_reaper: reaped process 2226288 (nebula-graphd), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB
