Yes, it was tcpdump output. This is the command string I used tcpdump -n port 123 &
Then I did the ntpdate -q commands on the same system. Phil ----- Original Message ----- From: Brian Reichert [EMAIL PROTECTED] Sent: 08/27/2008 09:43 PM AST To: General Linux-HA mailing list <[email protected]> Subject: Re: [Linux-HA] ha cluster - responding from wrong IP address? On Wed, Aug 27, 2008 at 08:03:46PM -0400, Phil N wrote: > Thank you for the response, > > >this is a bug in you NTP server, however it's a common one. > > do you mean the OS, or the NTP client? I've googled until my head hurts..... > > >my solution to this is to have the clients point at the individual boxes. > >the NTP protocol is designed to deal with multiple sources so it doesn't > >need to work through the VIP. > > I have one IP resolving for ntp.mydomain.com: the virtual IP of the HA > cluster. I could change this of course, but would prefer to fix the real > problem - the response from the wrong IP This really isn't a heartbeat issue, per se. All heartbeat does for managing a virtual IP is set up an alias for eth0 (or whatever), called eth0:0. It has the same MAC address as the original NIC. As to what the Linux network stack does is anyone's guess. I have access to RHEL5 and RHEL5 boxes; Ill try to reproduce this next week. Your original post included what seemed to be a tcpdump output; what host did you genernate that on, and what arguments did you use? > > Thanks! Phil > -- Brian Reichert <[EMAIL PROTECTED]> 55 Crystal Ave. #286 Daytime number: (603) 434-6842 Derry NH 03038-1725 USA BSD admin/developer at large _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
