80 to 90 TIME_WAIT records, is it reasonable?

[baijianpeng]

Sorry I am very newbie on server management and VestaCP.

Today I followed a command to check the TIME_WAIT records on my server, the result is:

Code: Select all

[root@ ~]# netstat -n | awk '/^tcp/ {++S[$NF]} END {for(a in S) print a, S[a]}'
CLOSE_WAIT 1
ESTABLISHED 32
FIN_WAIT2 1
SYN_SENT 1
CLOSING 1
TIME_WAIT 103

Why I run such a command? Because I feel my website pages open very slow. Then someone suggested me to check this.

It seems that the TIME_WAIT value is very high (103) on my server.

Is this value reasonable? If not, what should I do to improve the performance?

Thank you.

[BBuchanan1013]

Sorry I am very newbie on server management and VestaCP.

Today I followed a command to check the TIME_WAIT records on my server, the result is:

Code: Select all

[root@ ~]# netstat -n | awk '/^tcp/ {++S[$NF]} END {for(a in S) print a, S[a]}'
CLOSE_WAIT 1
ESTABLISHED 32
FIN_WAIT2 1
SYN_SENT 1
CLOSING 1
TIME_WAIT 103

Why I run such a command? Because I feel my website pages open very slow. Then someone suggested me to check this.

It seems that the TIME_WAIT value is very high (103) on my server.

Is this value reasonable? If not, what should I do to improve the performance?

Thank you.

To help make everyones life easier, could you post your OS and version. Also, are you running Apache or Apache+Nginx?

[baijianpeng]

OS： CentOS 7 64-bit
Web: Apache 2.4.6 + NginX 1.9.13

Thank you.