Jason Bradfield wrote:
> Jerry Jelinek wrote:
>> Jason Bradfield wrote:
>>> Hi,
>>> I am still having trouble with this if anyone can help out..
>>> Can someone explain..
>>>  From prstat -Z
>>> SWAP 941M
>>> RSS 844M
>>>  From df -h
>>> /tmp 1.1G
>>>  From Suns doc 
>>> http://docs.sun.com/app/docs/doc/819-2450/6n4o5mdis?a=view
>>> "The output of prstat *-Z* displays a SWAP column. The swap reported is
>>> the total swap consumed by the zone's processes and tmpfs mounts"
>>> Then how come my tmpfs is larger then the prstat -Z SWAP total for the
>>> zone.
>> The SWAP column tells you how much space the zone is actually
>> using 
> Then can anyone explain why I am getting out of swap space when my 
> zone.max-swap is 2GB and prstat is reporting less then 1GB.

I am not sure what you are hitting but since swap is a shared
resource it is possible that you are running out because of
other activity.  The rctl is a limit, not a reservation, so
if something else in another zone or the global zone uses up
swap, even for a short time, while the one specific zone needs
it, then you could get an error.

>> whereas looking at the size of tmpfs via 'df' will tell
>> you how much total VM the machine has.  The df output within
>> the zone is not adjusted downward to reflect any zone.max-swap
> True, but when I do a df -h, the usage column for /tmp is vastly 
> different between my zones.

Yes, that sounds normal since each zone is using something
different.  I was referring to the 'size' column, not the 'used'

zones-discuss mailing list

Reply via email to