Weblog entry #4 for gpall

can't deactivate volume group (LVM2 shutdown..
Posted by gpall on Tue 4 Mar 2008 at 19:38
Tags: ,
I am really curious! Am I the only one suffering from this?

My root fs is on LVM2 and each time I shut down the computer I get:

Can't deactivate volume group "vg-main" with 1 open logical volume(s)

It seems it is annoyed by the fact that lvm must be deactivated, but root fs has not been unmounted...

Any ideas?

 

Comments on this Entry

Posted by dkg (216.254.xx.xx) on Tue 4 Mar 2008 at 21:22
[ Send Message | View dkg's Scratchpad | View Weblogs ]
I get this error message as well on my LVM-based systems. I don't think it's a problem, actually, just an expected complaint as /etc/rc0.d/lvm2 stop tries to deactivate all groups and one of them can't be deactivated.

I haven't seen any detrimental effects from it, though i'd be curious to learn if anyone else has a way around this.

If deeper minds than mine think that it's OK to leave the last volume group active (providing that it backs the root filesystem), it might be worthwhile to consider patching the lvm2 shutdown scripts to not try to deactivate the volume used by the root filesystem. This would help avoid the spurious error message while not hiding real error messages that might come up if some other volume group can't be deactivated.

[ Parent | Reply to this comment ]

Posted by ajt (81.6.xx.xx) on Wed 5 Mar 2008 at 09:11
[ Send Message | View Weblogs ]
I've seen this too. On a VM system I have both LVM2 and RAID complaining when it gets to the end of the shutdown sequence as both are obviously still in use before the system gets shutdown. It's probably harmless but it looks worrying, more so to a newbie, plus I don't like false positives.

--
"It's Not Magic, It's Work"
Adam

[ Parent | Reply to this comment ]

Posted by dkg (216.254.xx.xx) on Wed 5 Mar 2008 at 16:45
[ Send Message | View dkg's Scratchpad | View Weblogs ]
Yeah, false positives are really bad. They're particularly bad because they train administrators and users to ignore errors.

[ Parent | Reply to this comment ]

Posted by ajt (81.6.xx.xx) on Wed 5 Mar 2008 at 17:29
[ Send Message | View Weblogs ]
I know what you mean. I believe in getting a message only if it's important, then I take note. False positives or needless prompting just makes you click happy and you miss important events.

I'm told you can break into a ICBM silo compound using rabbits to generate false positives in the middle of the night. Eventually the guards get so annoyed with the alarms going off that they disable them...

--
"It's Not Magic, It's Work"
Adam

[ Parent | Reply to this comment ]

Posted by Anonymous (86.47.xx.xx) on Wed 14 May 2008 at 12:56
I, for the record, have reported this to my local ICBM silo compound...

[ Parent | Reply to this comment ]

Posted by Anonymous (139.18.xx.xx) on Sat 19 Apr 2008 at 07:52
i got this error message too, well not exactly the same, i'm using two volume groups, one of them holds the root fs, but the one fires the error is the other one :(

[ Parent | Reply to this comment ]

Posted by eric (194.2.xx.xx) on Thu 6 Mar 2008 at 15:52
[ Send Message | View Weblogs ]
Yes, in fact, I have the same error at home with root fs on LVM2. For my laptop and work station, with root fs on LVM2 on LUKS... hum, I don't know but I don't remember this error message (a quick grep on work station doesn't return anything).

:eric:
http://blog.sietch-tabr.com

[ Parent | Reply to this comment ]

Posted by gpall (79.103.xx.xx) on Thu 6 Mar 2008 at 16:14
[ Send Message | View Weblogs ]
I, for the record, have reported this as a bug to debian...

[ Parent | Reply to this comment ]

Posted by ajt (81.6.xx.xx) on Fri 7 Mar 2008 at 09:55
[ Send Message | View Weblogs ]
Thanks, do you think it will help if others add to your bug report?

--
"It's Not Magic, It's Work"
Adam

[ Parent | Reply to this comment ]

Posted by Anonymous (155.207.xx.xx) on Fri 7 Mar 2008 at 09:57
Well, yes, I think it would help...

[ Parent | Reply to this comment ]