Hallo... Memory dan CPU sudah balance, itu berarti "load balance" sudah tercapai. CMIIW, term "load balance" itu menitik beratkan pada load system secara keseluruhan, bukan semata-mata jumlah session di masing-masing instance.
Kadang (bahkan sering) jumlah session tidak berbanding lurus dengan load (pemakaian resource). Load untuk 1 session transaksi insert 1 row BERBEDA dengan load untuk 1 session yang melakukan deleting 1m rows. Salam, Rohmad http://rohmad.net/category/database-oracle/ --- In [email protected], "Ujang Jaenudin" <[EMAIL PROTECTED]> wrote: > > lists, > > i have rac environment: > hp-ux itanium > 2 node rac > db 10.2.0.2 > hp-serviceguard > > now we got problem may be bug... > from the OS side it seem balance, that the currently processes running > is same (throughtop/glance) > the memory utilization is almost balance as well as cpu > > but from inside oracle node2 has 70% from total sessions. > i have configured server-side and client side load balancing altogether. > clb_goal set to short, throughput in the service > > from SR support told me that querying V$osstat on this platform return > no rows, that may be the root cause. > > does anyone have this kind of case? > thanks for sharing > > > -- > thanks and regards > ujang | oracle dba > jakarta - indonesia >
