rcmd socket: All ports in use - ERROR

[Danny Fang]

Hi,

This error is seen when performing multiple rcp operations for files from a LINUX machine to another HPUX11 machine.

Note that this is a Red Hat Enterprise Linux AS release 4
Kernel 2.6.9-22.EL on an i686

May I know what is the cause of this error?

Is there a number of files which is limited to RCP loops?

What configuration should be adjusted to avoid this error?

Thanks
Danny

[Steven E. Protter]

Shalom,

It may be a problem with network configuration or the xinetd daemon. Try restarting the daemon.

Causes:
1) Bugs, network saturation.


Limits:
Not that I know of but you might want to provide a script or numbers so we can put it in perspective.

Configuration:
I can't see any changes so far. You may wish to make sure both systems are properly patched and restart the inetd daemon on HP-UX.

Note this problem could be originating in the /var/adm/syslog/syslog.log file of the HP-UX system.

openssh is slower, but more secure and since it has its down daemon may have more capacity then rcp which runs out of xinetd.

SEP

[Danny Fang]

HI Steven,

When you mention that the problem could originate from the file /var/adm/syslog/syslog.log in the HPUX machine, may I know what are the types of problems from this file that could relate the " rcmd socket: All ports in use" error?

Could you explain on the capacity issues with rcp running with the xinetd?

Thanks
Danny