Hi, We have some SLES 10 SP1 and SP2 systems which were installed using the default of identifying disks by device ID.
This weekend our raised floor took a power hit so all LPARs and disk arrays crashed. Most Linux systems came back fine, but a few SLES 10 SP1 and SP2 systems that identified disks by device ID failed. Here are the important console messages: ... Loading jbd Loading ext3 Waiting for device /dev/disk/by-id/ccw-IBM.75000000030375.010b.22-part1 to appear: ..............................not found -- exiting to /bin/sh The systems had to be repaired manually. Once /etc/fstab and /etc/zipl.conf (followed by mkinitrd and zipl) were modified to identify disk by file name (e.g. /dev/dasda1), the systems came up fine. What strikes me as surprising is that it seems the disks "by-id" could not be found after a power hit (I have no idea where to find the value ccw-IBM.75000000030375.010b.22 in the above example). Why could the system find the correct disk by name but not by ID? Has anyone experienced this? (rebooting such systems has worked fine normally, so the power hit seems to be related). Also, just a "heads-up" for anyone with SLES 10 SP1 and SP2. If you are installing, you might want to click "Fstab Options" on the "Edit Partition" panel and set the "Mount in /etc/fstab" radio button group to "device name". If you have systems that identify disk by ID, you might want to test scenarios such as this. "Mike MacIsaac" <[EMAIL PROTECTED]> (845) 433-7061 ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
