Hi all. I've got a small cluster of machines, still running etch, using the redhat-cluster tools and cluster-LVM to share access to a fibre-channel disk array. I'm preparing to replace the disk array with a new/better/shinier one, but am running in to trouble with the migration of data from the old array to the new one. The plan is to use vgextent to add the capacity of the new array to the existing volume group, then use pvmove to migrate the physical extents associated with the existing logical volumes to the new physical device. Up through vgextend, this has all gone well, but pvmove is failing like this: pvmove -n /dev/xenvg/imprimatur_cache /dev/sdb /dev/sdc Error locking on node tig-hyperc-1: Internal lvm error, check syslog Failed to activate imprimatur_cache syslog contains: Jan 7 09:31:57 tig-hyperc-1 lvm[3028]: Couldn't find device with uuid 'NeyE2O-CPSF-tP9v-kK5G-8g02-nTsw-M8Ls5U'. Jan 7 09:31:57 tig-hyperc-1 lvm[3028]: Couldn't find all physical volumes for volume group xenvg. Jan 7 09:31:57 tig-hyperc-1 lvm[3028]: Couldn't find device with uuid 'NeyE2O-CPSF-tP9v-kK5G-8g02-nTsw-M8Ls5U'. Jan 7 09:31:57 tig-hyperc-1 lvm[3028]: Couldn't find all physical volumes for volume group xenvg. Jan 7 09:31:57 tig-hyperc-1 lvm[3028]: Couldn't find device with uuid 'NeyE2O-CPSF-tP9v-kK5G-8g02-nTsw-M8Ls5U'. Jan 7 09:31:57 tig-hyperc-1 lvm[3028]: Couldn't find all physical volumes for volume group xenvg. Jan 7 09:31:57 tig-hyperc-1 lvm[3028]: Couldn't find device with uuid 'NeyE2O-CPSF-tP9v-kK5G-8g02-nTsw-M8Ls5U'. Jan 7 09:31:57 tig-hyperc-1 lvm[3028]: Couldn't find all physical volumes for volume group xenvg. Jan 7 09:31:57 tig-hyperc-1 lvm[3028]: Volume group for uuid not found: wKj56UijD2ZIn9d15fxqvWUwcD47TCdqsKXbiumi66SWoXQGA9Jy0sCoJo01EHpM The UUID that it claims it can't find is the UUID of the new PV: pvdisplay /dev/sdc --- Physical volume --- PV Name /dev/sdc VG Name xenvg PV Size 8.17 TB / not usable 0 Allocatable yes PE Size (KByte) 4096 Total PE 2141695 Free PE 2141695 Allocated PE 0 PV UUID NeyE2O-CPSF-tP9v-kK5G-8g02-nTsw-M8Ls5U vgdisplay seems to indicate that the associated volume group is OK and has the full capacity of sdc available to it: vgdisplay xenvg --- Volume group --- VG Name xenvg System ID Format lvm2 Metadata Areas 2 Metadata Sequence No 315 VG Access read/write VG Status resizable MAX LV 0 Cur LV 108 Open LV 19 Max PV 0 Cur PV 2 Act PV 2 VG Size 13.25 TB PE Size 4.00 MB Total PE 3473500 Alloc PE / Size 122112 / 477.00 GB Free PE / Size 3351388 / 12.78 TB VG UUID wKj56U-ijD2-ZIn9-d15f-xqvW-UwcD-47TCdq Has anybody seen anything like this? Any clues you might be able to share? Thanks. noah
Attachment:
signature.asc
Description: Digital signature