Hi Dave,

tx for your response.  I will read it in a few mins, but wanted to respond.  I
knew ftp would be part of the inetd but wasn't 100% sure.  The odd thing is, it
has been working fine until a few days ago, and now it is denying access.  But, it
appears to have worked from another site only a few days ago, after it started
acting up otherwise.

Will restart the system and see if this perhaps clears things up, and will also
read your missive carefully.  As always, really appreciate your help. . . and will
letcha know what I find out.

david



Dave Wreski wrote:

> On 17-Mar-98 David Hughes wrote:
> > Am running a RH 4.2 box.  Have created a hosts.allow scheme to allow
> > access from specific domains only.  Everything has worked fine until a
> > few days ago when all of a sudden FTP access went south.  I can telnet
> > into it, web works, everything works fine, except ftp.  Dunno what
> > happened.
>
> What do you have listed in your hosts.allow?  How about hosts.deny?  Since I
> see that your concerned about security, you might as well be thorough.  This
> should be your /etc/hosts.deny:
>
> ALL: ALL
>
> And list individually the services you want to allow in /etc/hosts.allow:
>
> ALL: 127.
> in.ftpd: .my.domain
> in.telnetd: .my.domain
> in.popd: .my.domain
>
> This describes the `Mostly Closed' senerio.  You can also just use 'ALL: ALL'
> in hosts.allow during the time you are debugging..
>
> > Anyone suggest where to start looking??  Is there an FTP daemon that I
> > might need to stop and restart (will look into this of course).  Any
> > advice would be most appreciated. .
>
> No, you do not need to restart an ftpd -- it should be running from inetd,
> which spawns the ftpd upon each connection.  Your /etc/inetd.conf should look
> as follows:
>
> ftp     stream  tcp     nowait  root    /usr/sbin/tcpd  in.ftpd -l -a
>
> You can add the `-dv' argument to the end of that, which will produce more
> error logging.  If you are building a new machine, you might consider having
> more logging information for all your servers running on it.  You can increase
> the syslog error reporting by adding this to your /etc/syslog.conf:
>
> *.debug                         /var/log/debug.log
>
> Then signal syslogd:
>
> # kill -HUP `cat /var/run/syslog.pid`
>
> Dave




-- 
  PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
http://www.redhat.com/RedHat-FAQ /RedHat-Errata /RedHat-Tips /mailing-lists
         To unsubscribe: mail [EMAIL PROTECTED] with 
                       "unsubscribe" as the Subject.

Reply via email to