Oh, and I forgot the details.  The mdb command used:

 mdb -k <<EOF
::walk thread myvar|::print kthread_t t_schedflag|::grep .==0|::eval <myvar=K|::
print kthread_t t_procp->p_user.u_comm
EOF

What I found was that the threads where swapped out due to a memory shortfall 
and they where sleeping. Once swapped out they will not "un-swap" unless they 
go to running. The number displayed by the mdb output matched the 'w' column 
from vmstat.


--
This message posted from opensolaris.org
_______________________________________________
dtrace-discuss mailing list
[email protected]

Reply via email to