rcmd: Lost connection

[John Hegedus]

When I try to connect to other servers through remsh from Machine A to either B or C I have no problem. When I try to remsh from Machine B or C or even itself (A) I get the error rcmd: Lost connection???? I checked the .rhosts and /etc/hosts.equiv and everything seems fine??? Where do I go from here??

[A. Clay Stephenson]

You are confusing the ability to run the remsh client with the ability to act as a remsh server. I'll assume that firewalls are not in play as that could be another "gotcha". First, examine /etc/inetd.conf and make sure that the "shell" entry is not commented out and does contain remshd (the remsh deamon) as the executable.

[John Hegedus]

Inetd.conf checked out? I compared the files on the other servers that work and everything is the same?? The onll difference is that this particular machine is an NFS server

[A. Clay Stephenson]

Ok, next step. On 'A' execute "inetd -l" to start logging connection attempts. Now do something like "remsh serverA ls -l /etc" on serverB and then examine the log entries on A to see if the requests are making it to A.

[John Hegedus]

i ran remsh from itself and also the other boxes immediately followed by an ls command and it works just fine

[TheJuiceman]

I had a similar problem.  I could rlogin, rsh, etc from ServerA into ServerB without a problem.  However, going from ServerB to ServerA would not work (I got the familar "rcmd: Lost connection" error).  I checked all of the following on both servers to make sure the entries were correct:

/etc/hosts.equiv

/etc/hosts.allow

/etc/hosts.deny (just to be sure)

/root/.rhosts

/var/adm/inetd.sec

/etc/inetd.conf

I even checked to be sure that the server was resolving by the correct domain name.  No problem there.  Could not find the problem.

I  tried "inetd -c" multiple times along the way, but that did not help either.

Finally, ran "inetd -k" and then restarted inetd.  That worked!!!!

Don't know why inetd had to be killed and restarted, but that brought success.