On Thu, Mar 8, 2018 at 2:44 AM, qutic development <[email protected]>
wrote:

> Hi,
>
> I did a reboot inside of a zone, but the reboot process seems to be stale.
>
> A vmadm list the zone as down. vmadm stop cannot stop it.
> In /var/adm/messages I found:
>
> 2018-03-08T07:26:20.782281+00:00 qsm02 genunix: [ID 621342 kern.notice]
> Zone '9919e66b-1678-cd15-c3b1-b37e0c1e4f7c' (ID: 81) is shutting down,
> but 2 zone references and 6 credential references are still extant [VFS: 1]
>
> How do I find out which references block the zone from shutting down? A
>

I'm pretty sure the best hint is the information that you have from the
message above, as cryptic as that is.


>
> ps -z 9919e66b-1678-cd15-c3b1-b37e0c1e4f7c
>

That will only list processes that are in the zone.  There may be global
zone processes that have a hold on something in there too.  This should
tell you about them:

# grep $uuid/ /etc/mnttab | nawk '{print "fuser -c", $2}' | sh -x

The resulting "fuser /zones/$uuid" could turn up some false positives of
processes that have holds outside of /zones/$uuid/root.



>
> is not listing any process. strange. OS is: SmartOS 20171123T033756Z x86_64
>
> So the question is: how can I stop the zone - without rebooting the
> machine.
>

After killing any processes that the command above gave, you can also try
unmounting any file systems that are still in the zone.

# grep $uuid/root/ /etc/mnttab | nawk '{print "umount", $2}' | sort -r | sh
-x

Follow that up with a 'zoneadm -z $uuid halt'.

That may or may not work for you.  If it doesn't, I think you will need to
take a reboot.  I'm not aware of the history in this area since the Oracle
closed down OpenSolaris.  Those that have been on the outside all these
years may know of other developments in this area that can be helpful.

Mike



-------------------------------------------
smartos-discuss
Archives: https://www.listbox.com/member/archive/184463/=now
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb
Powered by Listbox: http://www.listbox.com

Reply via email to