One thing I've noticed since day one and never really bothered with it, although now I wonder.... If I ssh to any of my servers, execute an init script restart (or stop, start), then exit, the remote ssh process never quits. I have to kill the actual process from where I ssh-ed from (or presumably the remote process would produce the same effect):
[ server-1 ] --> ssh into --> [ server-2 ]
In this scenario, say I execute any restart command, rndc, sendmail, nfs, anything, then type exit, it just hangs. server-2 never issues a connection close or anything. I have a log in from a second terminal to server-1 and kill the ssh process that goes to server-2 (or log onto server-2 from a different terminal and kill the process.)
It doesn't matter what server I'm ssh FROM nor TO (I have about 10 of them,) it happens consistently on all of them.
All machines are running RH7.3, with the following openssh packages installed:
# rpm -qa | grep openssh openssh-3.1p1-6 openssh-clients-3.1p1-6 openssh-server-3.1p1-6
Has anyone noticed this?
I've noticed this on machines where the services attaches itself to your /dev/pts/## session. Until the process is restarted or the SSH session is remotely killed, the session will remain open. VSFTPd (standalone from RawHide) and webmin are two services off the top of my head that will do this guarenteed.
-Rick -- Rick Johnson, RHCE #807302311706007 - [EMAIL PROTECTED] Linux/Network Administrator - Medata, Inc. (from home) PGP Public Key: https://mail.medata.com/pgp/rjohnson.asc
-- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
