Failure to Log Off Users on OSR5.0.7
Lucky Leavell
I have a customer who reports that OSR5.0.7 fails to log off users whose
telnet session is terminated by a power failure on the client side. (I.e.,
who still shows them logged on.) This problem did not occur on earlier
versions of OSR5.
Has anyone else experienced this?
Solutions? (Other than a script to manually kill the telnet sessions on the
OSR5.0.7 side)
Thank you,
Lucky
Lucky Leavell Phone: (800) 481-2393 (US/Canada)
UniXpress - Your Source for SCO OR: (812) 366-4066
1560 Zoar Church Road NE FAX: (812) 366-3618
Corydon, IN 47112-7374 Email: lu...@UniXpress.com
WWW Home Page: http://www.UniXpress.com
Scott McMillan
wrote:
>OS: OSR5.0.7 with MP1/UP1 installed
>
>I have a customer who reports that OSR5.0.7 fails to log off users whose
>telnet session is terminated by a power failure on the client side. (I.e.,
>who still shows them logged on.) This problem did not occur on earlier
>versions of OSR5.
>
>Has anyone else experienced this?
>
>Solutions? (Other than a script to manually kill the telnet sessions on the
>OSR5.0.7 side)
>
>Thank you,
>Lucky
>
As in the user's sessions are still active, or the 'who' command shows
those users are still logged in? If the latter, see TA# 105610 at
http://support.sco.com, and http://tinyurl.com/399wj for Bela's
additional comments.
Scott McMillan
Bela Lubkin
> OS: OSR5.0.7 with MP1/UP1 installed
>
> I have a customer who reports that OSR5.0.7 fails to log off users whose
> telnet session is terminated by a power failure on the client side. (I.e.,
> who still shows them logged on.) This problem did not occur on earlier
> versions of OSR5.
A machine which is going down in a power failure does not send out "I'm
dying now" messages. It is normal for sessions from a rebooted machine
to stay alive for a while. Once the client is finished rebooting, next
time the server sends it anything on the telnet session, it will get a
TCP "reset" packet (meaning "What are you talking about? I don't have a
session open with you on that socket number!")
If the users are running an application that chatters (e.g. updates an
on-screen time of day once a minute), the sessions will disappear pretty
quickly. If not, they'll stay around until a TCP "keepalive" packet is
sent. The default values of the keepalive kernel parameters will keep
such sessions alive for over 2 hours. See the sections on keepalives in
telnetd(ADMN), tcp(ADMP).
> Has anyone else experienced this?
>
> Solutions? (Other than a script to manually kill the telnet sessions on the
> OSR5.0.7 side)
Tuning the keepalive parameters sharply down may help. You can do it
systemwide with the parameters described in tcp(ADMP); or for telnet
sessions only, by adding command line flags to the telnetd invocation in
/etc/inetd.conf. If you see a big difference from "earlier versions of
OSR5", it must be that they had done such tuning on the earlier
releases.
The down side of tightening the keepalive parameters is that you also
sharply limit TCP's resilience in the face of cable outages. A session
whose both ends are alive and ready to run normally stays alive even if
an intermediate cable or router is disconnected for a few minutes. If
you've told the server to frantically send keepalives, it will notice a
cable break and disconnect sessions that would otherwise have survived
nicely.
>Bela<