problem about add default route

[zhang yijian_1]

hi all:
when I add a route use command : route add default gateway_ip_address 1 ,the system report a error message : route: fork function is failed, too many process exist and memory is 8GB ,nproc is 4092. it report when the system is booting and not heavy load application running
who can help me ? thank you

[LiPEnS]

Hi
Can you tell which os is on the machine and result of the command "swapinfo -tam"?

[zhang yijian_1]

hi :
the error message is : route :The fork function failed. Too many processes already exist
swapinfo -atm
Mb Mb Mb PCT START/ Mb
TYPE AVAIL USED FREE USED LIMIT RESERVE PRI NAME
dev 12288 0 12288 0% 0 - 1 /dev/vg00/lvol2
reserve - 256 -256
memory 6305 180 6125 3%
total 18593 436 18157 2% - 0 -
os is hp-ux 11.11
thank you

[zhang yijian_1]

and somee info I collect is as attach....

[Devender Khatana]

Hi,

nproc kernel parameters is fully utilized. Find out the utilization of nproc by

#sar -v 1 11

See the column proc-sz

xxx/yyy

Here xxx is the current valu and yyy is the maximum value.

If fully utilized you need to increase it.

HTH,
Devender

[Steven E. Protter]

1) Just for grins check /etc/rc.config.d/netconf and make sure there is no conflict between current values and what you tried to add.

2) netstat -in


3) I agree with the assessment that this box has a lot of processes on it. It does however not seem to be doing an awful lot of paging. Memory use itself is not high.

ps -ef | wc -l

How many proceses actually running.

kmtune | grep maxuprc

Is the maximum number of processes for a given user, the root user being exceeded.

vmstat 10

or

http://www.hpux.ws/system.perf.sh
A collection of performance scripts that do so in the background.

SEP

[zhang yijian_1]

thank your for your suggestion , I find the source of problem and already slove it...
someone edit the /etc/route as ascii file.
in the file , only one line : route add default ip_address
so ,when use command route to add route ,it run endless......

[zhang yijian_1]

thank your very much....