On Tue, Dec 19, 2000 at 10:10:46AM -0600, Dave Dykstra wrote:
> I noticed this same problem on my redhat 6.2 machine last week. Check
> /var/adm/messages. Mine reported
>
> inetd[415]: rsync/tcp server failing (looping or being flooded), service
>terminated for 10 min
>
> I could see no way to configure Linux inetd to avoid that, so I ended up
> starting rsync as an independent daemon out of /etc/rc.d/rc3.d.
Someone forwarded to me a message from a mailing list of another project
discussing a similar thing right now which gave the answer: if you change
the "nowait" in the inetd.conf entry to "nowait.NNNN" then NNNN specifies
the maximum number of spawns allowed for that entry per 60 second period.
The man page section on that is misleading because it seems to indicate
that whole column is for datagrams only, but I tested it and it does work
for tcp too. I set mine to 600 for 10 per second; the default is only 40
for one every 1.5 seconds.
- Dave Dykstra
