1

I "inherited" RH 6.4 on a computer with an nVidia video card. I took the image and dd'd it to another computer that has the same hardware.

By design, the second computer will be headless. This second computer boots and runs just fine. However, when I remove the video card from the second computer so that it's headless, it hangs and does not boot. There is little debugging I can do to determine why. The BIOS seems to post just fine.

I really don't want to reinstall the system from scratch. My assumption is that the kernel module for the video card is trying to do something to the hardware but it fails to find it and hangs.

So short of a reinstall, what can I do to get this to boot? I can do whatever I need to do with the video card in, so I can get it to boot up to make the changes.

Thanks!

Edit: This is RHEL 6.5, not 6.4.

The content of /var/log/messages is:

  • Feb 11 10:34:06 MDCS-Server kernel: device virbr0-nic entered promiscuous mode
  • Feb 11 10:34:06 MDCS-Server kernel: virbr0: starting userspace STP failed, starting kernel STP
  • Feb 11 10:34:07 MDCS-Server kernel: ip_tables: (C) 2000-2006 Netfilter Core Team
  • Feb 11 10:34:07 MDCS-Server kernel: nf_conntrack version 0.5.0 (16384 buckets, 65536 max)
  • Feb 11 10:34:08 MDCS-Server dnsmasq[1972]: started, version 2.48 cachesize 150
  • Feb 11 10:34:08 MDCS-Server dnsmasq[1972]: compile time options: IPv6 GNU-getopt DBus no-I18N DHCP TFTP "--bind-interfaces with SO_BINDTODEVICE"
  • Feb 11 10:34:08 MDCS-Server dnsmasq-dhcp[1972]: DHCP, IP range 192.168.122.2 -- 192.168.122.254, lease time 1h
  • Feb 11 10:34:08 MDCS-Server dnsmasq[1972]: no servers found in /etc/resolv.conf, will retry
  • Feb 11 10:34:08 MDCS-Server dnsmasq[1972]: read /etc/hosts - 2 addresses
  • Feb 11 10:34:08 MDCS-Server dnsmasq[1972]: read /var/lib/libvirt/dnsmasq/default.addnhosts - 0 addresses
  • Feb 11 10:34:08 MDCS-Server dnsmasq[1972]: read /var/lib/libvirt/dnsmasq/default.hostsfile
  • Feb 11 10:34:08 MDCS-Server kernel: Ebtables v2.0 registered
  • Feb 11 10:34:08 MDCS-Server kernel: ip6_tables: (C) 2000-2006 Netfilter Core Team
  • Feb 11 10:34:10 MDCS-Server kernel: lo: Disabled Privacy Extensions
  • Feb 11 10:34:22 MDCS-Server init: tty (/dev/tty2) main process (1941) killed by TERM signal
  • Feb 11 10:34:22 MDCS-Server init: tty (/dev/tty3) main process (1943) killed by TERM signal
  • Feb 11 10:34:22 MDCS-Server init: tty (/dev/tty4) main process (1945) killed by TERM signal
  • Feb 11 10:34:22 MDCS-Server init: tty (/dev/tty5) main process (1949) killed by TERM signal
  • Feb 11 10:34:22 MDCS-Server init: tty (/dev/tty6) main process (1951) killed by TERM signal
  • Feb 11 10:34:23 MDCS-Server abrtd: Got signal 15, exiting
  • Feb 11 10:34:27 MDCS-Server init: Disconnected from system bus
  • Feb 11 10:34:27 MDCS-Server console-kit-daemon[2031]: WARNING: no sender#012
  • Feb 11 10:34:27 MDCS-Server rpcbind: rpcbind terminating on signal. Restart with "rpcbind -w"
  • Feb 11 10:34:27 MDCS-Server auditd[1279]: The audit daemon is exiting.
  • Feb 11 10:34:27 MDCS-Server kernel: type=1305 audit(1423672467.509:38): audit_pid=0 old=1279 auid=4294967295 ses=4294967295 subj=system_u:system_r:auditd_t:s0 res=1
  • Feb 11 10:34:27 MDCS-Server kernel: type=1305 audit(1423672467.606:39): audit_enabled=0 old=1 auid=4294967295 ses=4294967295 subj=system_u:system_r:auditctl_t:s0 res=1
  • Feb 11 10:34:27 MDCS-Server kernel: Kernel logging (proc) stopped.
  • Feb 11 10:34:27 MDCS-Server rsyslogd: [origin software="rsyslogd" swVersion="5.8.10" x-pid="1304" x-info="http://www.rsyslog.com"] exiting on signal 15

I've tried various combinations of all the options below. I can't get this thing to boot. NUM-LOCK also does not work, so the system really is out to lunch.

SOLVED: Finally, it is solved. The log above is at shutdown, not boot. It never got to the kernel, apparently. (No one ever accused me of being smart, though in my defense it's hard to debug it when you can't see anything useful like, you know, the clock.)

In any event, it was a GRUB problem. I removed 'rhbg' from the kernel parameters and commented out the splashimage line, which is spelled out nicely here, and everything started working splendidly.

Many thanks to those who tried to help.

  • Does /var/log/messages or /var/log/dmesg have anything useful? Is there on-board VGA you can connect to so you can watch it boot up? – Bratchley Feb 6 '15 at 19:06
  • Boot up without the nvidia card in it I mean. – Bratchley Feb 6 '15 at 19:07
  • This link talks about headless/vnc setup on CentOS. It may be of help. – Vic Feb 6 '15 at 19:29
  • Also investigate the novga, nofb and console options in grub as they do show up in headless and xen installs. – Vic Feb 6 '15 at 19:36
  • Vic's link is also probably a pretty good one since you will probably want to make sure the serial console is working even after solving this issue. – Bratchley Feb 6 '15 at 19:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.