Weirdness. When I IPL and the LVM fails to mount - If I chccwdev --offline 0.0.0203 then chccwdev --online 0.0.0203 the logical volume appears.
No idea why it does not work during IPL. On Tue, Apr 7, 2009 at 1:34 PM, Mark Pace <[email protected]> wrote: > That's what I was thought about LVM also. The device is coming online now > (repair filesystem) # > lsdasd > lsdasd > > Bus-ID Status Name Device Type BlkSz Size > Blocks > ============================================================================== > > 0.0.0202 active dasda 94:0 ECKD 4096 2347MB > 601020 > 0.0.0203 active dasdb 94:4 ECKD 4096 2347MB > 601020 > 0.0.0200 active dasdc 94:8 FBA 512 48MB > 98304 > 0.0.0201 active dasdd 94:12 FBA 512 128MB > 262144 > > But LVM is not working with 0203. > > I guess I'll try some LVM commands and see what it thinks about that > device. I'll let you know if I figure it out. > > Thanks for all your input. > > On Tue, Apr 7, 2009 at 1:26 PM, Bruce Hayden <[email protected]> wrote: > >> I thought that they would be mounted in the order that you ran >> dasd_configure. That's how SLES 10 worked, so I presumed that SLES 11 >> works this way, but some testing shows that it doesn't. On my system, >> the root volume is always dasda, then it seems to mount them in the >> order the files appear in /etc/udev/rules.d - but one boot I just did >> wasn't quite in this order. On SLES 10, you ran mkinitrd which saved >> the activation order in the initrd. SLES 11 doesn't store this in the >> initrd. The initrd activates the root device then reads the rest of >> the udev rules from it. (I don't know how it behaves if the root >> device is in an LVM. I don't set up my systems that way.) >> >> I think the real answer is to always use /dev/disk/by-path (best for >> virtual device numbers under VM) or /dev/disk/by-id (to use the volume >> labels) to refer to disks. LVM should scan all disks and look for an >> LVM signature, so it shouldn't care what dasd letter they are mounted >> with. >> >> On Tue, Apr 7, 2009 at 12:57 PM, Mark Pace <[email protected]> wrote: >> > Well that fixed the problem of the device not coming online at IPL. But >> now >> > all the dasd labels have been changed. >> > 0200 was dasda now dasdc >> > 0201 was dasdb now dasdd >> > 0202 was dasdc now dasda >> > 0203 was dasdd now dasdb >> > >> > root dasd mounts fine because it is mounted "by-path", but the swap disk >> > were by /dev/dasdx1 and now obviously that is incorrect. Also the LVM >> > doesn't mount as 0203 is no longer /dev/dasdd1. >> > >> > What, now, determines the order? >> > >> > >> > >> > >> > -- >> > Mark Pace >> > Mainline Information Systems >> > 1700 Summit Lake Drive >> > Tallahassee, FL. 32317 >> > >> > ---------------------------------------------------------------------- >> > 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 >> > >> >> >> >> -- >> Bruce Hayden >> Linux on System z Advanced Technical Support >> IBM, Endicott, NY >> >> ---------------------------------------------------------------------- >> 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 >> > > > > -- > Mark Pace > Mainline Information Systems > 1700 Summit Lake Drive > Tallahassee, FL. 32317 > -- Mark Pace Mainline Information Systems 1700 Summit Lake Drive Tallahassee, FL. 32317 ---------------------------------------------------------------------- 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
