[PVE-User] LVM autoactivation failed with multipath over iSCSI
nada
nada at verdnatura.es
Tue Jan 14 13:47:11 CET 2020
good day
a week ago I upgraded our proxmox 'cluster' to buster
(it has just 2 nodes and both nodes are upgraded)
MANY thanks for your guidelines at proxmox wiki !!! it works great :-)
now i am testing multipath over iSCSI to SAN storage
it works good, snapshots, volume resizing and migrations are functional
but after reboot
the LVMThin VG and relevant LVs are NOT autoactivated
what am i missing ?
temporally i created simple rc-local service/script to activate VG
PLS can anybody write me how it should be done ?
following are some details
thank you
Nada
root at mox:~# pveversion
pve-manager/6.1-5/9bf06119 (running kernel: 5.3.13-1-pve)
root at mox:~# grep pve /etc/lvm/lvm.conf
# Also do not scan LVM disks from guests on both VGs named & not named
'pve'
global_filter = [ "r|/dev/zd.*|", "r|/dev/mapper/pve-.*|",
"r|/dev/mapper/pve-(vm|base)--[0-9]+--disk--[0-9]+|",
"a|/dev/mapper/3600.*|", "a|/dev/mapper/san.*|" ]
root at mox:~# multipath -ll
3600c0ff000195f8e2172de5d01000000 dm-8 HP,P2000 G3 iSCSI
size=23G features='1 queue_if_no_path' hwhandler='1 alua' wp=rw
|-+- policy='round-robin 0' prio=50 status=active
| |- 9:0:0:5 sdn 8:208 active ready running
| |- 8:0:0:5 sdh 8:112 active ready running
| |- 11:0:0:5 sdo 8:224 active ready running
| `- 7:0:0:5 sdm 8:192 active ready running
`-+- policy='round-robin 0' prio=10 status=enabled
|- 10:0:0:5 sde 8:64 active ready running
|- 5:0:0:5 sdj 8:144 active ready running
`- 6:0:0:5 sdk 8:160 active ready running
root at mox:~# la /dev/mapper/
total 0
drwxr-xr-x 2 root root 340 Jan 13 19:09 .
drwxr-xr-x 20 root root 5820 Jan 13 19:31 ..
lrwxrwxrwx 1 root root 7 Jan 13 19:09
3600c0ff000195f8e2172de5d01000000 -> ../dm-8
crw------- 1 root root 10, 236 Jan 13 19:08 control
lrwxrwxrwx 1 root root 7 Jan 13 19:08 pve-data -> ../dm-5
lrwxrwxrwx 1 root root 7 Jan 13 19:08 pve-data_tdata -> ../dm-3
lrwxrwxrwx 1 root root 7 Jan 13 19:08 pve-data_tmeta -> ../dm-2
lrwxrwxrwx 1 root root 7 Jan 13 19:08 pve-data-tpool -> ../dm-4
lrwxrwxrwx 1 root root 7 Jan 13 19:08 pve-root -> ../dm-1
lrwxrwxrwx 1 root root 7 Jan 13 19:08 pve-swap -> ../dm-0
lrwxrwxrwx 1 root root 7 Jan 13 19:08 pve-vm--104--disk--1 ->
../dm-7
lrwxrwxrwx 1 root root 7 Jan 13 19:08 pve-zfs -> ../dm-6
lrwxrwxrwx 1 root root 8 Jan 13 19:09 santest-santestpool ->
../dm-12
lrwxrwxrwx 1 root root 8 Jan 13 19:09 santest-santestpool_tdata
-> ../dm-10
lrwxrwxrwx 1 root root 7 Jan 13 19:09 santest-santestpool_tmeta
-> ../dm-9
lrwxrwxrwx 1 root root 8 Jan 13 19:09 santest-santestpool-tpool
-> ../dm-11
lrwxrwxrwx 1 root root 8 Jan 13 19:09 santest-vm--901--disk--0 ->
../dm-13
root at mox:~# pvs -a
PV VG Fmt Attr PSize
PFree
/dev/mapper/3600c0ff000195f8e2172de5d01000000 santest lvm2 a--
23.28g 3.24g
/dev/mapper/santest-vm--901--disk--0 ---
0 0
/dev/sda2 ---
0 0
/dev/sda3 pve lvm2 a--
<232.57g <1.86g
/dev/sdb1 ---
0 0
/dev/sdb9 ---
0 0
/dev/sdc1 ---
0 0
/dev/sdc9 ---
0 0
/dev/sdd1 ---
0 0
/dev/sdd9 ---
0 0
/dev/sdf1 ---
0 0
/dev/sdf9 ---
0 0
/dev/sdg1 ---
0 0
/dev/sdg9 ---
0 0
/dev/sdi1 ---
0 0
/dev/sdi9 ---
0 0
/dev/sdl1 ---
0 0
/dev/sdl9 ---
0 0
root at mox:~# lvs -a
LV VG Attr LSize Pool Origin
Data% Meta% Move Log Cpy%Sync Convert
data pve twi-aotz-- 138.57g 0.59
10.69
[data_tdata] pve Twi-ao---- 138.57g
[data_tmeta] pve ewi-ao---- 72.00m
[lvol0_pmspare] pve ewi------- 72.00m
root pve -wi-ao---- 58.00g
swap pve -wi-ao---- 4.00g
vm-104-disk-1 pve Vwi-a-tz-- 50.00g data 1.62
zfs pve -wi-ao---- 30.00g
[lvol0_pmspare] santest ewi------- 20.00m
santestpool santest twi-aotz-- 20.00g 3.40
12.21
[santestpool_tdata] santest Twi-ao---- 20.00g
[santestpool_tmeta] santest ewi-ao---- 20.00m
vm-901-disk-0 santest Vwi-aotz-- 2.50g santestpool
27.23
root at mox:~# grep santest /var/log/syslog.1 |tail
Jan 13 19:02:26 mox lvm[2005]: santest: autoactivation failed.
Jan 13 19:04:43 mox lvm[441]: Monitoring thin pool
santest-santestpool-tpool.
Jan 13 19:06:14 mox lvm[441]: No longer monitoring thin pool
santest-santestpool-tpool.
Jan 13 19:06:14 mox blkdeactivate[12609]: [LVM]: deactivating Volume
Group santest... done
Jan 13 19:09:02 mox lvm[2003]: Cannot activate LVs in VG santest while
PVs appear on duplicate devices.
Jan 13 19:09:02 mox lvm[2003]: Cannot activate LVs in VG santest while
PVs appear on duplicate devices.
Jan 13 19:09:02 mox lvm[2003]: 0 logical volume(s) in volume group
"santest" now active
Jan 13 19:09:02 mox lvm[2003]: santest: autoactivation failed.
Jan 13 19:09:11 mox lvm[442]: Monitoring thin pool
santest-santestpool-tpool.
Jan 13 19:09:12 mox rc.local[1767]: 2 logical volume(s) in volume
group "santest" now active
root at mox:~# pvesm status
Name Type Status Total Used
Available %
backup dir active 59600812 28874144
27669416 48.45%
local dir active 59600812 28874144
27669416 48.45%
local-lvm lvmthin active 145301504 857278
144444225 0.59%
santestpool lvmthin active 20971520 713031
20258488 3.40%
zfs zfspool active 30219964 13394036
16825928 44.32%
root at mox:~# cat /lib/systemd/system/rc-local.service
[Unit]
Description=/etc/rc.local Compatibility
Documentation=man:systemd-rc-local-generator(8)
ConditionFileIsExecutable=/etc/rc.local
After=network.target iscsid.service multipathd.service
open-iscsi.service
[Service]
Type=forking
ExecStart=/etc/rc.local
TimeoutSec=0
RemainAfterExit=yes
GuessMainPID=no
[Install]
WantedBy=multi-user.target
root at mox:~# cat /etc/rc.local
#!/bin/bash
# just to activate VGs from SAN
/bin/sleep 10
/sbin/vgchange -aly santest
More information about the pve-user
mailing list