Environment:
- Oracle Linux
- Red Hat Linux
- Oracle 11G
- Oracle 12C
Symptoms:
High cluster wait and performance loss due to cluster wait.Diagnose:
[root@node2 ~]# netstat -s |grep assem
1169750855 reassemblies required
203909271 packets reassembled ok
3180797 packet reassembles failed
and countinuesly increases. Monitor reassembly failure with script: [root@node2 ~]# cat packages.sh while true; do x=`netstat -s | grep "packet reassembles failed"|awk '{print $1}'` echo `date` ": " `expr $x - $t`; t=$x sleep 5; done; [root@node2 ~]# ./packages.sh Pzt May 6 14:57:08 +03 2019 : 126 Pzt May 6 14:57:13 +03 2019 : 132 Pzt May 6 14:57:18 +03 2019 : 186 Pzt May 6 14:57:23 +03 2019 : 152 ^C[root@node2 ~]# [root@trakyatipdbnode2 ~]# sysctl -a|grep ipfrag net.ipv4.ipfrag_high_thresh = 4194304 net.ipv4.ipfrag_low_thresh = 3145728 net.ipv4.ipfrag_max_dist = 64 net.ipv4.ipfrag_secret_interval = 0 net.ipv4.ipfrag_time = 30
Solution:
Default value for buffer usage during IP packet reassembly 4M for high 3M for low threashold. Problem is solved after incrementing to 15-16M values.[root@node2 ~]# echo "16777216" > /proc/sys/net/ipv4/ipfrag_high_thresh
[root@node2 ~]# echo "15728640" > /proc/sys/net/ipv4/ipfrag_low_thresh
To make this change permanent across reboots, edit the /etc/sysctl.conf file and add the line:
net.ipv4.ipfrag_high_threshold = 16777216
net.ipv4.ipfrag_low_thresh = 15728640
Immediately after setting new kernel values, cluster wait value drops to 0 and "netstat -s |grep assem" shows there is no change in packet reassembles failed.
标签：Poor,Network,lost,2019,ipv4,net,node2,root,ipfrag
From： https://www.cnblogs.com/lkj371/p/17025721.html