You are not logged in.
Hello,
I have just added a SSD to my HTPC, changing drives order (not eactly on purpose).
Currently, the configuration is SSD (/, /swap and /home with LVM) + 3x1To in RAID5 (/var and /mnt/data with LVM).
Since, the LVM partitions over the raid array cannot be mounted on start up. I fall in rescue mode (type root password and correct blabla or Ctrl+D blabla).
But if I disabled and enabled the vg, everything goes fine.
I named ssd-mediacenter the VG on the SSD and raid-mediacenter the VG on the raid array. I labeled lv-[mount point] LVs and [ssd|raid]-[mount opint] the filesystem
ssd-mediacenter/lv-root is the ext4 partition labeled ssd-root
/dev/sdd2 isn't used anymore.
raid-mediacenter/lv-rootfs is my old root partition (before the SSD) still kept for config files.
[root@mediacenter ~]# pvs
PV VG Fmt Attr PSize PFree
/dev/md127 raid-mediacenter lvm2 a-- 1,82t 20,00g
/dev/sda2 ssd-mediacenter lvm2 a-- 476,74g 424,74g
[root@mediacenter ~]# vgs
VG #PV #LV #SN Attr VSize VFree
raid-mediacenter 1 3 0 wz--n- 1,82t 20,00g
ssd-mediacenter 1 3 0 wz--n- 476,74g 424,74g
[root@mediacenter ~]# lvs
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert
lv-data raid-mediacenter -wi-a----- 1,77t
lv-var raid-mediacenter -wi-a----- 15,00g
rootfs raid-mediacenter -wi-a----- 10,00g
lv-home ssd-mediacenter -wi-ao---- 30,00g
lv-root ssd-mediacenter -wi-ao---- 20,00g
lv-swap ssd-mediacenter -wc-ao---- 2,00g
[root@mediacenter ~]# LANG=C lsblk
lsblk: dm-3: failed to get device path
lsblk: dm-4: failed to get device path
lsblk: dm-5: failed to get device path
lsblk: dm-3: failed to get device path
lsblk: dm-4: failed to get device path
lsblk: dm-5: failed to get device path
lsblk: dm-3: failed to get device path
lsblk: dm-4: failed to get device path
lsblk: dm-5: failed to get device path
lsblk: dm-3: failed to get device path
lsblk: dm-4: failed to get device path
lsblk: dm-5: failed to get device path
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 477G 0 disk
|-sda1 8:1 0 200M 0 part /boot
|-sda2 8:2 0 476.8G 0 part
| |-ssd--mediacenter-lv--root 253:0 0 20G 0 lvm /
| |-ssd--mediacenter-lv--swap 253:1 0 2G 0 lvm [SWAP]
| `-ssd--mediacenter-lv--home 253:2 0 30G 0 lvm /home
`-sda3 8:3 0 1007K 0 part
sdb 8:16 0 931.5G 0 disk
`-sdb1 8:17 0 930G 0 part
`-md127 9:127 0 1.8T 0 raid5
sdc 8:32 0 931.5G 0 disk
`-sdc1 8:33 0 930G 0 part
`-md127 9:127 0 1.8T 0 raid5
sdd 8:48 0 931.5G 0 disk
|-sdd1 8:49 0 930G 0 part
| `-md127 9:127 0 1.8T 0 raid5
`-sdd2 8:50 1007K 0 part
[root@mediacenter ~]# vgchange -a n raid-mediacenter
0 logical volume(s) in volume group "raid-mediacenter" now active
[root@mediacenter ~]# vgchange -a y raid-mediacenter
3 logical volume(s) in volume group "raid-mediacenter" now active
[root@mediacenter ~]# LANG=C lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 477G 0 disk
|-sda1 8:1 0 200M 0 part /boot
|-sda2 8:2 0 476.8G 0 part
| |-ssd--mediacenter-lv--root 253:0 0 20G 0 lvm /
| |-ssd--mediacenter-lv--swap 253:1 0 2G 0 lvm [SWAP]
| `-ssd--mediacenter-lv--home 253:2 0 30G 0 lvm /home
`-sda3 8:3 0 1007K 0 part
sdb 8:16 0 931.5G 0 disk
`-sdb1 8:17 0 930G 0 part
`-md127 9:127 0 1.8T 0 raid5
|-raid--mediacenter-rootfs 253:3 0 10G 0 lvm
|-raid--mediacenter-lv--var 253:4 0 15G 0 lvm
`-raid--mediacenter-lv--data 253:5 0 1.8T 0 lvm
sdc 8:32 0 931.5G 0 disk
`-sdc1 8:33 0 930G 0 part
`-md127 9:127 0 1.8T 0 raid5
|-raid--mediacenter-rootfs 253:3 0 10G 0 lvm
|-raid--mediacenter-lv--var 253:4 0 15G 0 lvm
`-raid--mediacenter-lv--data 253:5 0 1.8T 0 lvm
sdd 8:48 0 931.5G 0 disk
|-sdd1 8:49 0 930G 0 part
| `-md127 9:127 0 1.8T 0 raid5
| |-raid--mediacenter-rootfs 253:3 0 10G 0 lvm
| |-raid--mediacenter-lv--var 253:4 0 15G 0 lvm
| `-raid--mediacenter-lv--data 253:5 0 1.8T 0 lvm
`-sdd2 8:50 0 1007K 0 part
Last edited by elukas (2015-02-25 08:46:35)
Offline
The issue seems to be due to an error with the raid array itself or rebuild the raid array forces rescan LVM configuration, I don't know.
In case of something like that append to someone :
[root@mediacenter ~]# history | tail -n 20
...
486 vgchange -a n raid-mediacenter
487 pvchange --help
488 mdadm --stop /dev/md50
489 mdadm --assemble --help
490 mdadm --assemble --update=?
491 mdadm --assemble /dev/md50 --name=mediacenter01 --update=name
492 pvs
493 vgs
494 lvs
495 vgchange -a y raid-mediacenter
496 lvs
497 mdadm -Es
...
after a reboot lsblck displays the right devices configuration.
Thank you for viewers who search to help.
EDIT January 30 2015:
The problem persists.
The lvm2-pvscan@9:127.service fails to load.
[root@mediacenter ~]# systemctl status lvm2-pvscan@9\:127.service -l
? lvm2-pvscan@9:127.service - LVM2 PV scan on device 9:127
Loaded: loaded (/usr/lib/systemd/system/lvm2-pvscan@.service; static; vendorpreset: disabled)
Active: failed (Result: signal) since ven. 2015-01-30 11:16:56 CET; 16min ago
Docs: man:pvscan(8)
Main PID: 167 (code=killed, signal=KILL)
janv. 30 11:16:56 mediacenter systemd[1]: lvm2-pvscan@9:127.service stop-sigterm timed out. Killing.
janv. 30 11:16:56 mediacenter systemd[1]: lvm2-pvscan@9:127.service: main process exited, code=killed, status=9/KILL
janv. 30 11:16:56 mediacenter systemd[1]: Unit lvm2-pvscan@9:127.service entered failed state.
janv. 30 11:16:56 mediacenter systemd[1]: lvm2-pvscan@9:127.service failed.
[root@mediacenter ~]# journalctl -b0
...
janv. 30 11:15:24 localhost kernel: md127: detected capacity change from 0 to 1996891357184
janv. 30 11:15:25 localhost kernel: md127: unknown partition table
janv. 30 11:15:25 localhost kernel: EXT4-fs (dm-0): mounted filesystem with ordered data mode. Opts: (null)
janv. 30 11:15:25 localhost systemd-udevd[59]: '/usr/bin/mdadm --incremental --export /dev/sdb1 --offroot ${DEVLINKS}' [110] terminated by signal 15 (Terminated)
janv. 30 11:15:25 localhost lvm[162]: Daemon lvmetad returned error 104
janv. 30 11:15:25 localhost lvm[162]: WARNING: Failed to connect to lvmetad. Falling back to internal scanning.
janv. 30 11:15:25 localhost lvm[162]: Cannot proceed since lvmetad is not active.
janv. 30 11:15:25 localhost systemd[1]: lvm2-pvscan@8:2.service: control process exited, code=exited status=5
janv. 30 11:15:25 localhost systemd[1]: Unit lvm2-pvscan@8:2.service entered failed state.
janv. 30 11:15:25 localhost systemd[1]: lvm2-pvscan@8:2.service failed.
...
Last edited by elukas (2015-01-30 13:45:23)
Offline
I have a question,
If a make a backup config for the VG raid-mediacenter, and then remove everything, reboot and restore, will I recover my partitions ?
Last edited by elukas (2015-01-30 13:47:39)
Offline
Hi,
Is my problem could be due to the hardware (chipset) ?
My HTPC is based on LGA775 socket on top of the Asus P5N7A-VM motherboard. There is 3 red SATA slots, and 2 black.
Is it possible that the 2 black ports (handled, I think, by the additional JMicron JMB368 controller) can't be used before the system has been fully started ?
When I press F8 to get the bios selector boot device menu, only SATA devices plugged into the red slots are enabled.
If is that so, how could I embed JMicron driver into the initramfs image ?
Thanks.
Offline
I finally solved my issue.
It seems that the problem was due to mdadm_udev hook, which causes this trouble in combination with systemd (I didn't try adding udev).
this configuration works for me : HOOKS="base systemd autodetect modconf block mdadm mdadm_udev sd-lvm2 filesystems keyboard sd-vconsole fsck"
Offline