Enabling the rexec service under Redhat 9

I saw Hanny Tidore's effort to make this work last month, but did not see any
indication of success.  This is what I have been up to this weekend to get
rexec working from a z/OS client (IBM mainframe):

Response from REXEC after working over RH 9 iptables and xinetd.conf:

EZA4801I  MVS TCP/IP REXEC CS V1R5
EZA4810E  The call to rexec_af() function failed:
EZA4744I  Foreign host aborted the connection.
rexecd-connect    :rexecd-connect-success:

The bottom line is the contents of splash files I set up in the xinetd.conf
member:

[flexes@tserver01 flexes]$ cat /etc/xinetd.d/rexec
# default: off
# description: Rexecd is the server for the rexec(3) routine.  The server \
# provides remote execution facilities with authentication based \
# on user names and passwords.
service exec
{
 disable = no
 socket_type  = stream
 wait   = no
 user   = root
 log_on_success  += USERID
 log_on_failure   += USERID
 server   = /usr/sbin/in.rexecd
 only_from = 192.168.1.7
 log_type  = FILE /usr/flexes/rexecd.log
 log_on_success += PID
 log_on_success += HOST
 log_on_success += EXIT
 log_on_success += DURATION

 log_on_failure += HOST

 banner         = /usr/flexes/rexecd-connect.txt
 banner_success = /usr/flexes/rexecd-connect-success.txt
 banner_fail    = /usr/flexes/rexecd-connect-fail.txt
}
[flexes@tserver01 flexes]$


The log shows:

START: exec pid=27821 from=192.168.1.7
EXIT: exec status=1 pid=27821 duration=31(sec)
START: exec pid=27868 from=192.168.1.7
EXIT: exec status=1 pid=27868 duration=30(sec)
START: exec pid=27886 from=127.0.0.1             <=== Loopback inside Linux
FAIL: exec address from=127.0.0.1
EXIT: exec status=0 pid=27886 duration=0(sec)
START: exec pid=5885 from=192.168.1.7             <=== tried touch
EXIT: exec signal=13 pid=5885 duration=30(sec)    <=== Broken pipe
START: exec pid=6101 from=192.168.1.7
EXIT: exec status=1 pid=6101 duration=30(sec)
START: exec pid=6179 from=192.168.1.7
EXIT: exec status=1 pid=6179 duration=30(sec)

Note that the rexec on the loopback (i.e. from inside Linux) had to be aborted
manually as it sat and did nothing for several minutes.

Here's selected netstat and iptables output:

[root@tserver01 root]# netstat -ap|grep exec
tcp        0      0 *:exec                  *:*                     LISTEN
6174/xinetd

Chain RH-Lokkit-0-50-INPUT (2 references)
target     prot opt source               destination
ACCEPT     tcp  --  192.168.1.7          anywhere           tcp dpt:exec
ACCEPT     tcp  --  192.168.1.7          anywhere           tcp dpt:shell
ACCEPT     tcp  --  anywhere             anywhere           tcp dpt:telnet flags
:SYN,RST,ACK/SYN

I tried with and without --syn

I also enabled and tried rsh and got less response.


By the way, I tried rexec from the same client to a Unixware system -- it works
just fine -- excepting that I should tighten up that firewall.

--
George Haeh

-- 
Shrike-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/shrike-list