thanks Gordan, looks like we are in the same timezone, here is the command, same as previous one except for using mkfs.gfs2 instead of gfs_mkfs mkfs.gfs2 -t clearcase:gfs -p lock_dlm -j 6 /dev/vg100/lvol0 [root@wplccdlvm445 gfs]# clustat Cluster Status for clearcase @ Wed Dec 30 14:56:37 2009 Member Status: Quorate Member Name ID Status ------ ---- ---- ------ wplccdlvm445.cn.ibm.com 1 Online, Local wplccdlvm446.cn.ibm.com 2 Online On Wed, Dec 30, 2009 at 2:44 PM, Gordan Bobic <gordan@xxxxxxxxxx> wrote: > Diamond Li wrote: >> >> after I use mkfs.gfs2, it works. However, I did not see any document >> to mention this command, always gfs_mkfs. > > I'm not sure what you're doing differntly (you omitted the FS creation > command in your previous email), but this works just fine for me: > > gfs_mkfs -j 2 -p lock_dlm -t test:root /dev/hdb > mount /mnt/gfs > > The fstab line is: > /dev/hdb /mnt/gfs gfs defaults,noatime,nodiratime 0 0 > > Just tested it on a scratch VM. > > I'm assuming you have your cluster.conf configured right and the cman > service (which provides fenced, groupd, etc.) has started without any > errors? Again, you haven't posted your cluster.conf so it's impossible to > tell. > > You also haven't specified whether your intention is to use gfs or gfs2. > They are not the same. > >> in my humble opnion, redhat has a log way to provide real enterprise >> solution, both from software quality and documentation. > > There doesn't seem to be enough in this thread to persuade me that the cause > of problems isn't user error. :) > > Gordan > > -- > Linux-cluster mailing list > Linux-cluster@xxxxxxxxxx > https://www.redhat.com/mailman/listinfo/linux-cluster > -- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster
