TFTP Open Timeout
- 
 Forgive me if this has been answered before but I couldn’t find the solution. I used to have v.32 working just fine. I keep getting a TFTP Open Timeout error when a client tries to boot. I can’t get this thing to boot from the server. CentOS 6 Server running FOG 1.1.1 (TFTP without DHCP) 
 PFsense router with DHCP correctly configured. The clients are given DHCP addresses properly.
 firewall is disabled
 Option 67 is set as /undionly.kpxe.Am I doing this right? I can’t tell if TFTP is running because it’s under xinet. That part is new to me. 
- 
 Is it set as: /undionly.kpxe or undionly.kpxe? 
- 
 Please read wiki! 
 [url]http://www.fogproject.org/wiki/index.php/Tftp_timeout[/url]…
- 
 option 67 was set to undionly.kpxe. [CODE] tftp x.x.x.x get undionly.kpxe[/CODE] gives me a server error: tftpd: read(ack): Connection refused selinux is set to permissive and iptables is off. Permissions on /tftpboot are 777. 
- 
 Connection refused means that there is a firewall(iptables/ufw) issue or the service is no longer operating correctly. On some versions of Ubuntu tftp will fail for unknown reasons and will need to be restarted. 
- 
 I’ve solved this. I hadn’t configured dhcpd.conf correctly in my pfSense router. Apparently, [CODE]option tftp-server-name “ipaddressofserver”;[/CODE] doesn’t do anything, which I find weird. Only the [CODE]next-server[/CODE] seems to work. Thanks for your help everybody. 
