Unfortunately, we are not using inetd or xinetd. I've read about some of the files in /proc/sys/net/ipv4 but I don't understand too much about these and don't know if they even have anything to do with the tcp backlog.
Thanks! Jeff -----Original Message----- From: Ed Wilts [mailto:[EMAIL PROTECTED]] Sent: Friday, April 12, 2002 1:34 PM To: [EMAIL PROTECTED] Subject: Re: TCP/IP backlog queue On Fri, Apr 12, 2002 at 01:22:20PM -0600, Williams, Jeff wrote: > What is the default for the tcp/ip backlog queue for redhat 6.2 and 7.2? > And how would you go about changing it. > > I'm seeing a odd problem with an application that is sending extremely fast > simultaneous connects get their connection refused and wanting to try and > see if this might fix it.... It looks like you're getting nailed by the inetd incoming connection count. After the wait in inetd.conf for the service, add a large number, like 120 to allow 120 connections in a minute - the default is 40. man inetd.conf and look for the wait/nowait[.max] info. 7.2 uses xinetd and the syntax is different - I can't remember where the same concept got moved to. You can verify if this is your problem by looking at your server messages file. If my memory serves me correctly, you'll see flooding messages in there. If you're not running your service through inetd, then ignore all my rambling above... -- Ed Wilts, Mounds View, MN, USA mailto:[EMAIL PROTECTED] _______________________________________________ Redhat-list mailing list [EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list _______________________________________________ Redhat-list mailing list [EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
