lvmpud messages in syslog

[Tim Medford]

Upgraded our production HPUX rx3600 to 11iv3 over the weekend.  I create the volume groups at version 2.1, although they are not shared and it is not a serviceguard cluster.  Very vanilla configuration, standalone server with storage direct attached to an WVA via fibre channel.

I am getting a lot of these messages in the syslog.  They are not errors, but an annoyance.  In reading about this in the man pages, it is a daemon for handling online shared volume group reconfiguration.  If my volume groups are not shared, is there any harm in shutting down this daemon?

Thanks in advance, Tim

Apr 15 10:34:06 saifprod lvmpud[4072]: Read task ticket=93 type=1
Apr 15 10:34:06 saifprod lvmpud[4072]: Process task ticket=93 type=1
Apr 15 10:34:06 saifprod lvmpud[4072]: Complete task ticket=93 type=1 status=0
Apr 15 10:49:05 saifprod lvmpud[4072]: Read task ticket=94 type=1
Apr 15 10:49:05 saifprod lvmpud[4072]: Process task ticket=94 type=1
Apr 15 10:49:05 saifprod lvmpud[4072]: Complete task ticket=94 type=1 status=0

[Patrick Wallek · Accepted solution]

From what I have read, if you are not using shared volume groups (do not desire the MORE functionality) and are  not using LVM 2.2 snapshots, then you should be able to stop this daemon.

From the lvmpud man page:

lvmpud is the daemon that handles LVM online shared volume group reconfiguration and automatic increase of pre-allocated extents for space efficient snapshots when threshold value is reached.

In order to prevent the daemon from starting when the system reboots next you need to modify the /etc/rc.config.d/lvmconf file and set:

START_LVMPUD=0

[Tim Medford]

Thanks Patrick.  I read the same thing in the man pages, and it seemed harmless enough, I just wanted to see if anyone else had experienced the same issue.  Not much on the forums when you search for lvmpud, must be a newer part of LVM.

Tim