I'm following the Red Hat Documentation (https://access.redhat.com/documentation ... -HAAA.html) to enable Exclusive Activation of a Volume Group in a Cluster (by editing the volume_list parameter in /etc/lvm/lvm.conf to exclude the clustered volume groups). However, when I do this and reboot the node, the device nodes are missing under /dev/<vg_name>/<lv_name>. Obviously, this makes creating a filesystem on the logical volumes impossible, and I can't create a clustered LVM resource as the device is not available. If I comment out the volume_list entry in /etc/lvm/lvm.conf and re-run dracut and reboot, the device nodes are visible - but I don't want to risk corrupting my metadata this way (which the Red Hat documentation warns against).
My cluster is quorate and DLM and CLVMD are running:
Code: Select all
# pcs status
Cluster name: webcluster
Last updated: Mon Aug 11 21:19:01 2014
Last change: Mon Aug 11 21:11:18 2014 via cibadmin on centos05
Stack: corosync
Current DC: centos07 (2) - partition with quorum
Version: 1.1.10-32.el7_0-368c726
2 Nodes configured
5 Resources configured
Online: [ centos05 centos07 ]
Full list of resources:
iscsi-stonith-device (stonith:fence_scsi): Started centos05
Clone Set: dlm-clone [dlm]
Started: [ centos05 centos07 ]
Clone Set: clvmd-clone [clvmd]
Started: [ centos05 centos07 ]
PCSD Status:
centos05: Online
centos07: Online
Daemon Status:
corosync: active/enabled
pacemaker: active/enabled
pcsd: active/enabled
Code: Select all
[root@centos05 ~]# lvs
LV VG Attr LSize Pool Origin Data% Move Log Cpy%Sync Convert
root centos -wi-ao---- 17.51g
swap centos -wi-ao---- 2.00g
lv_apache vg_data -wi------- 1.00g
[root@centos05 ~]# vgs
VG #PV #LV #SN Attr VSize VFree
centos 1 2 0 wz--n- 19.51g 0
vg_data 1 1 0 wz--nc 7.99g 6.99g
[root@centos05 ~]# ls -ld /dev/vg_data
ls: cannot access /dev/vg_data: No such file or directory
[root@centos05 ~]# lvscan
inactive '/dev/vg_data/lv_apache' [1.00 GiB] inherit
ACTIVE '/dev/centos/swap' [2.00 GiB] inherit
ACTIVE '/dev/centos/root' [17.51 GiB] inherit
Code: Select all
[root@centos05 ~]# pcs resource create test_vg LVM volgrpname=vg_data exclusive=true --group=apachegroup
[root@centos05 ~]# pcs resource show
Clone Set: dlm-clone [dlm]
Started: [ centos05 centos07 ]
Clone Set: clvmd-clone [clvmd]
Started: [ centos05 centos07 ]
Resource Group: apachegroup
test_vg (ocf::heartbeat:LVM): Stopped
Code: Select all
[root@centos05 ~]# pcs resource debug-start test_vg
Operation start for test_vg (ocf:heartbeat:LVM) returned 7
> stderr: INFO: Activating volume group vg_data
> stderr: INFO: Reading all physical volumes. This may take a while... Found volume group "vg_data" using metadata type lvm2 Found volume group "centos" using metadata type lvm2
> stderr: INFO: 0 logical volume(s) in volume group "vg_data" now active
> stderr: ERROR: LVM Volume vg_data is not available (stopped)
> stderr: ERROR: LVM: vg_data did not activate correctly
I can supply further information/configuration if required.
Cheers,
Toki