Red Hat Bugzilla – Bug 72268
hang during NTP configuration
Last modified: 2008-05-01 11:38:03 EDT
After installing from CD, I rebooted. During firstboot's NTP configuration I
entered an IP address. After waiting on it to do something for over a minute, I
realized that the ethernet cable was unplugged. I plugged it in, and waited a
bit longer. During this time I had switched to a different virtual console, and
upon returning to VC7, firstboot did not repaint the screen.... ever.
Upon going to another computer, I got in via ssh. I found that the cause of the
hang was ping -c 2 <ipaddress> had been running for quite some time. Upon using
kill to get rid of the ping, firstboot repainted the screen and popped up an
error dialog. A retry made everything go OK.
There are a couple problems here...
1) If you are trying to see if a machine is a suitable NTP server, use an NTP
command rather than ping to see if it is alive. Ping tests ICMP ECHO REQUEST
and ECHO REPLY packets. A different command such as ntpdate will test NTP
packets. Since ICMP ECHO REQUEST or ECHO REPLY packets may be firewalled, this
is a bad test to do.
2) If you really insist on using ping, consider the -w option. I have not
tested to see if it would help in this situation or not, but the info in the man
page suggests that it should.
*** This bug has been marked as a duplicate of 72012 ***