On Thu, Feb 25, 2010 at 5:15 PM, Christine Tran
<christine.tran at gmail.com> wrote:
> On Thu, Feb 25, 2010 at 6:00 PM, Dan Price <dp at eng.sun.com> wrote:
>> On Thu 25 Feb 2010 at 05:42PM, Christine Tran wrote:
>>> Hi,
>>>
>>> We have on our server a little bot widget which allows a user to
>>> create, modify, boot and pretty much manage his own zones. ?What we're
>>> finding out is that on the rare occasions when the bot is down
>>> (administrative or fault), SMF helpfully restarts all our zones
>>> because all the zoneadm live in the same CTID as the bot. ?To fix this
>>
>> Restarting zoneadmd should not cause the zone to restart, so I'm
>> perplexed by this statement...
>
> Yeah, actually, you are right. ?I can't remember exactly our encounter
> with this specific issue except at the time I noted the fix to put the
> whole thing in ctrun. ?Perhaps it was during the zone creation that
> when our bot went down it caused things to break.
>
>>
>> What are you running? ?I ask because, at least in Nevada, zoneadmd
>> puts itself into its own contract when it starts up, automatically.
>
> Unfortunately, not Nevada, U8.
Perhaps you were running something other than the stock zones start
method. From /lib/svc/method/svc-zones on U8:
51 # We don't (yet) support restart for zones, so
52 # we need to get all of the zones stuff off into
53 # a different contract.
54 #
55 ctrun -l none -i none zoneadm -z $zone boot &
--
Mike Gerdts
http://mgerdts.blogspot.com/
