[Linux-cluster] RHEL5 LVM Cluster locking problem

[Thomas Reiter]

Hi everyone in this community.

I have installed a working two node cluster using the new RHEL5 
clustersuite and RHEL5 x64.
kernel:  ......  2.6.18-8.el5
cman: cman-2.0.64-1.0.1.el5
lvm2-cluster version: lvm2-cluster-2.02.16-3.el5
iscsi: iscsi-initiator-utils-6.2.0.742-0.6.el5
multipath: device-mapper-multipath-0.4.7-8.el5

What i am trying to do now is to connect my shared storage (iscsi) via 
lvm2 clustered VG and mirrored LV.

I can successful create the PVs and VG.... but when activating the LV i 
get following errors in /var/log/messages:

/kernel: device-mapper: table: *253:6: mirror: Error creating mirror 
dirty log*
kernel: device-mapper: ioctl: error adding target to table
dmeventd[4626]: dmeventd ready for processing.
dmeventd[4626]: Monitoring mirror device, VG_mail-LV_mail for events
lvm[4590]: Error locking on node mail1: device-mapper: reload ioctl 
failed: Invalid argument
lvm[4590]: Error locking on node mail1: device-mapper: reload ioctl 
failed: Invalid argument
lvm[4590]: Failed to activate new LV./

What does this mean? Where to search for the mistake?

Tried already to change locking mode in /etc/lvm/lvm.conf to 2 and use 
locking_library liblvm2clusterlock.so from x32 RHEL4.5 without success 
also. (i know x32 and x64 os, but i thought it's worth a try..... 
couldn't find an external locking lib for RHEL5 x64)

Just for info, please find below my system config:

*lvm.conf:*
/library_dir = "/usr/lib64"
locking_type = 3
fallback_to_clustered_locking = 1
fallback_to_local_locking = 1
locking_dir = "/var/lock/lvm"
mirror_log_fault_policy = "allocate"
mirror_device_fault_policy = "remove"/

*iscsi devices:*
/2 infortrend iscsi storage boxes with each raid 6 + Hotspare
LV size should be 6 TeraByte -> mirrored to 2nd box/

*multipath -l shows: (right now there is only one path configured)*
/mail_m0 (3600d0230006fb0c300065568b855f301) dm-2 IFT,A16E-G2130-4
[size=5.9T][features=0][hwhandler=0]
\_ round-robin 0 [prio=0][active]
 \_ 3:0:0:0 sde 8:64  [active][undef]
mail_mlog (3600d0230006fb0c300065568b855f300) dm-3 IFT,A16E-G2130-4
[size=200M][features=0][hwhandler=0]
\_ round-robin 0 [prio=0][active]
 \_ 3:0:0:1 sdf 8:80  [active][undef]
mail_m1 (3600d0230006fbccc0006624409e63600) dm-4 IFT,A16E-G2130-4
[size=5.9T][features=0][hwhandler=0]
\_ round-robin 0 [prio=0][active]
 \_ 4:0:0:0 sdb 8:16  [active][undef]/

*vgdisplay shows:
*/--- Volume group ---
  VG Name               VG_mail
  System ID
  Format                lvm2
  Metadata Areas        3
  Metadata Sequence No  7
  VG Access             read/write
  VG Status             resizable
  Clustered             yes
  Shared                no
  MAX LV                0
  Cur LV                4
  Open LV               0
  Max PV                0
  Cur PV                3
  Act PV                3
  VG Size               11.82 TB
  *PE Size               128.00 MB*
  Total PE              96824
  Alloc PE / Size       96667 / 11.80 TB
  Free  PE / Size       157 / 19.62 GB
  VG UUID               cdjlgF-n0dh-mlDo-wPco-RvFl-C3fO-PUQSuh/*

*PV Size 128MB for getting 6 TB volume.

*command issued to create LV:*
/lvcreate -m1 -L 5.90T -n LV_mail VG_mail /dev/mapper/mail_m0 
/dev/mapper/mail_m1 /dev/mapper/mail_mlog
/
*lvs shows:*
/LV       VG         Attr   LSize   Origin Snap%  Move Log          Copy%
  LV_mail  VG_mail    mwi-d-   5.90T                    LV_mail_mlog   0.00
  LogVol00 VolGroup00 -wi-ao   3.41G
  LogVol01 VolGroup00 -wi-ao 480.00M
/
but LV is NOT active because i get the error mentioned above!

Any ideas?
Every hint would be highly appriaciated.

Thanks
Thomas

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/linux-cluster/attachments/20071129/35f0ed1c/attachment.htm>