I am running a server with Durability Level set to “Majority and persist to active”.
After about a month, replication is delayed and response is slow.
I have tried changing various kernel parameters, but there seems to be no change.
If there are any points that should be reviewed, please let us know.
The configuration of the target server is as follows.
CPU：AMD EPYC 7J13 64-Core Processor 4Core8Thread
Memory：32GB
OS：Oracle Linux 7.8（4.14.35-1902.303.4.1.el7uek.x86_64）
$ cat /etc/sysctl.conf
fs.aio-max-nr = 1048576
kernel.sem = 250 32000 100 128
net.core.rmem_default = 262144
net.core.rmem_max = 16777216
net.core.netdev_max_backlog = 30000
net.core.rps_sock_flow_entries = 32768
net.core.somaxconn = 32768
net.core.wmem_default = 262144
net.core.wmem_max = 16777216
net.ipv4.ip_local_port_range = 1024 65500
net.ipv4.tcp_rmem = 4096 87380 16777216
net.ipv4.tcp_wmem = 4096 65536 16777216
vm.swappiness = 0
vm.dirty_background_ratio = 1
vm.dirty_ratio = 3
vm.vfs_cache_pressure = 50
vm.min_free_kbytes = 524288
Version：Community Edition 6.6.0 build 7909
Service：Only “data” is valid
Memory Quotas：21504MB
Reader Thread Settings：Disk i/o optimized
Writer Thread Settings：Disk i/o optimized
number of buckets：1
Type：Couchbase
Replicas：1
Server Nodes：4
Item count：about 20,000,000
Since it is possible that the machine translation may not have conveyed the message well, the original text (in Japanese) is also included below.
運用しているサーバーでDurability Levelを「Majority and persist to active」にして動かしているのですが、
1ヶ月程度経った際にレプリケーションが遅延して応答が遅くなってしまいます。
色々とカーネルパラメータを変更する等行ってみましたが、変化が無いようです。
見直すべきポイント等ございましたら、ご教授いただけないでしょうか。