1

I've just updated a CentOS 5.10 box to 5.11 with all the latest patches and rebooted.

Following the reboot I'm now getting the below error message every few minutes:

init: Id "2" respawning too fast: disabled for 5 minutes

Line 2 of /etc/inittab reads:

2:2345:respawn:/sbin/mingetty tty1

Which looks consistent with other CentOS 5.11 servers.

I can see a running /sbin/mingetty process in ps aux with a recently started time and cannot see any other signs of the process crashing.

How can I trace the cause and/or rectify this.

moztech
  • 188
  • 6
  • Any errors in syslog? Is there a /dev/tty1? – Dan Feb 04 '15 at 16:49
  • Only entries in /var/log/messages (which is syslog on Centos I believe) is the init.. one shown above ps aux shows: >root xxxx 0.0 0.0 1756 460 tty1 Ss+ 16:49 0:00 /sbin/mingetty tty1 – moztech Feb 04 '15 at 16:54
  • No error messages on the actual tty1 (the first text console)? Same permissions and on tty1 as say tty2? – Dan Feb 04 '15 at 18:26
  • When I look on the console there is just the same message repeated: `INIT: Id "2" respawning too fast: disabled for 5 minutes` – moztech Feb 04 '15 at 18:39
  • Maybe you could `strace /sbin/mingetty tty1`? – Dan Feb 04 '15 at 18:47
  • strace outputs a whole load of stuff but finishes with 'nanosleep({5, 0}, {5, 0}) = 0 exit_group(1) = ?' Is there something specific I should be looking for? – moztech Feb 04 '15 at 19:03
  • Was hoping for an error around the end of the trace. Blind debugging is no fun. For example on ubuntu I get `tty1: no controlling tty: Operation not permitted` a few lines before the exit. But default on ubuntu is getty which works. – Dan Feb 04 '15 at 19:10
  • Which begs a question: is mingetty default on centos? If not, switch to getty or whatever is default – Dan Feb 04 '15 at 19:17
  • I've just tried running strace on the "live" mingetty process and just got the below as it died and was respawned. I believe mingetty is the default on CentOS, /sbin does have agetty but no sign of getty it'self `strace -p 30819 Process 30819 attached - interrupt to quit read(0, "", 1) = 0 exit_group(0) = ? Process 30819 detached` – moztech Feb 04 '15 at 19:30
  • Unfortunately I don't have a centos server around and I'm pretty much out of guesses. Especially if only tty1 is not working. – Dan Feb 04 '15 at 19:44

1 Answers1

4

For the benefit of anyone reading this with the same issue. My resolution after much research and experimentation was that this is a virtual machine which doesn't have multiple windows.

As a result I have remmed out the below line in /etc/inittab

#2:2345:respawn:/sbin/mingetty tty1

Doing so has resolved the issue for me on this system and another one in each case a CentOS 5 system which started reporting this following a recent patch (inc Kernel) cycle.

moztech
  • 188
  • 6