getpid : peer died: No such file or directory

[steve_586]

I have this message appearing in the syslog.log every 5 mins. I have checked nstrpty is 200, npty is 400 and i have 400 pty's in /dev. User count averages around 50.
I have been getting reports of users sessions being kicked out but only at certain times in the day. I also seem to have multiple telnetd processes running. System is running 10.20.
Has anyone got the time to post a suggestion?
Much Appreciated.

[Muthukumar_5 · Accepted solution]

May be some telnet client or network connectivity made client is making problem here. Try to start inetd in logging mode as,

# inetd -l

what is the debugging messages you are getting?

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=501458
http://unix.derkeiler.com/Newsgroups/comp.sys.hp.hpux/2004-07/0053.html

Read the above as well.

--
Muthu

[Arunvijai_4]

Hi Steve,

You can check these threads for more information,

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=964544
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=100539

It usually happens when communication terminates abrubtly between telnet server and client. You can analyse more with # inetd -l
(Inetd with logging enabled)

-Arun

[Muthukumar_5]

Problem is because of

a) virus attack from client side. Use scanners
b) dummy requests from port scanners from remote machine
c) Improper termination from login connections like telnet or remsh

Try to tight your machine access with,

a) TCP Wrappers
b) /var/adm/inetd.sec
c) /etc/profile scripting

--
Muthu

[Arunvijai_4]

Hi Steve,

You can search ITRC threads to find more information in lieu with this problem.

http://www1.itrc.hp.com/service/james/search.do?todo=search&admit=-682735245+1144316653467+28353475&searchtext=getpid+%3A+peer+died&from=other&origin=0&submit.x=0&wpa=www1.itrc.hp.com%3A80&submit.y=0&searchcategory=ALL&hpl=1&searchcriteria=allwords&rn=25&source=7000&presort=rank&chkServStor=on&esc=us.support.itrc.hp.com

There are lots of Knowledge trees available. You can check them as well.

-Arun

[steve_586]

Thank you all ever so much, your pointers have helped me track down at leat what is causing the error message, whether it is the cause of the strange port diconnects i will have to wait and see. After enabling inetd logging i could see that these failures were sourced from one IP. Tracked this down to a Linux box that was checking/monitoring ports, thus the abrupt getpid closures.

Thank you all for your time.