Topic: INIT: id "1" respawning too fast: disabled for 5 m

[Gordon]

Hi
We have just started getting this message on 5.5. The logs appear to stop during the backup process, so suspect a dump of MySql but not sure.
Does anyone have any suggestions on how to trap this error. At the moment the system requires a cold reboot (power off) to recover. All shutdown commands are ignored! Still looking for more clues.
Gordon

[Lloyd Keen]

Have a look in /etc/inittab for a line matching the above. The id is the first column and should be similar to:
1:23457:respawn:/sbin/mingetty tty1 (where 1 is the id)
Just comment out the line then run
#init q
This will stop the offending program, but I don't know why it's doing it. Why you can't shutdown is because it appears to be the getty program running on tty1 that's playing up. Try changing to terminal2 ctrl+alt+f2 I think or shift+f2 and you should be able to reboot.

[Gordon]

Thanks Lloyd
The problem has not occurred for the last 2 nights as we have not inserted the tape. So the call appears to be coming from the backup routine. Does anyone know why inittab would be called by backup?
Gordon

[Gordon]

Ok
We have a bit more to go on.
The problem was being caused by a corrupted 4Mb client data file. It would hang the server even issuing a copy command in MC on the console. To get rid of the problem needed to drop back to init 1 and rm -f. As this was only a small file in 6Gb data, does anyone know how to find corrupted files without having to try and move all the data to see if it crashes? fsck reported the partition as clean so it did not help.
Regards
Gordon