mount problem at startup after a crash

[vincentbach]

after a crash, there is a problem during the file system mount. only the filesystems from vg00 get mounted. i need to activate the other vg before beeing able to mount.

the message is vxfs mount: Cannot open /dev/vg01/lvol1: No such device or address.
the lvmtab file has been recreated, but the problem still exists.

as a temporary solution, i modified the S100localmount file but i'd like to find the cause of this problem.

[melvyn burnard]

what does /etc/lmvrc say for the AUTO_ACTIVATE parameter?
if it is a 0, then it is set NOT to activate any VG other than vg00, usually because it is part of a Serviceguard cluster

[Ganesan R]

Hi,

>>the message is vxfs mount: Cannot open /dev/vg01/lvol1: No such device or address <<

By looking the above message it doesn't seems to be /etc/lmvrc issue.

It is trying to activate but could not find the device files.

Are you doing any kind of vgexport/vgimport activity on this server?

All device files are exist?

[vincentbach]

sorry, i forgot to write that AUTO_VG_ACTIVATE=1

[vincentbach]

The devices exists, and mount correctly after activating the volume groups with vgchange -a y

[Simeon Harwood]

I have the same problem on my system. I have a server that is part of a service guard cluster. All of my service guard volumes activate fine with in SG, but I have 2 that are only on this node, and they do not automatically activate after a reboot.
I came to conclusion that this is how it is supposed to be because of the flag AUTO_VG_ACTIVATE=0. I assumed that the install of SG did this.
So I added a little script into the startup of the machine that activated these volumes, and now it's all fine.

Sime.