-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 20 Jun 2003, Blaes, Rainer wrote:
in order to simulate a drive failure in our environment running SLES 8 with Raid1 for the system disk (set-up via Yast2) we - plugged out one drive
While the system was running? Ouch. What kind of disk drive is this? Are you sure that it's capable of such "hot-plugging"? From the Software-RAID-HOWTO: 5.1 Simulating a drive failure If you want to simulate a drive failure, then plug out the drive. You should do this with the power off. If you are interested in testing whether your data can survive with a disk less than the usual number, there is no point in being a hot-plug cowboy here. Take the system down, unplug the disk, and boot it up again.
- shutdown the system ====> system is hanging
This is most likely the kernel's IDE layer (if it was an IDE disk), getting confused about the missing disk.
- power off + booting ====> system is hanging during fsck cycle
This should usually not happen, but might be due to the way you shut down the drive before. What file system are you using? What does /proc/mdstat tell you?
- power off + reconnection of the former disk =====> system came up successfully Any help, tips for a proper handling of this scenario are welcome especially if we have to insert a "new" disk!
Take a look at the excellent Software-RAID-HOWTO - it gives many detailed
and helpful hints.
/usr/share/doc/howto/en/html/Software-RAID-HOWTO.html
Bye,
LenZ
- --
- ------------------------------------------------------------------
Lenz Grimmer