Recommended Network Settings if Application Job Performance Degradation on the BDA is Correlated with High "socket buffer overrun" Statistics in the netstat Output
(Doc ID 2740140.1)
Last updated on APRIL 17, 2023
Applies to:
Big Data Appliance Integrated Software - Version 4.13.0 and laterLinux x86-64
Purpose
If high values of "socket buffer overrun" are reported in netstat output, change the default values in /etc/sysctl.conf on all cluster hosts, to these values to increase the maximum socket buffer size:
- net.core.rmem_max=134217728
- net.core.wmem_max=134217728
- net.ipv4.tcp_rmem=4096 87380 134217728
- net.ipv4.tcp_wmem=4096 65536 134217728
- net.core.netdev_max_backlog=300000
- net.ipv4.tcp_moderate_rcvbuf=1
Note that these values are based on the recommendations in "Oracle VM3:10GbE Network Performance Tuning Technical Brief".
Scope
Details
To view full details, sign in with your My Oracle Support account. |
|
Don't have a My Oracle Support account? Click to get started! |
In this Document
| Purpose |
| Scope |
| Details |
| Check for high levels of "socket buffer overrun" netstat statistcs |
| Cause of high "socket buffer overrun" netstat statistcs values |
| Solution |