Bug 4872999 - RAC instance cannot get global enqueue

..............

[@more@]

Symptoms:	Related To:
	RAC (Real Application Clusters) / OPS

Description

An instance might not be able to obtain a global enqueue after
hitting some earlier memory shortage issue at the master instance
on the same enqueue. This can show up as trying to get some
global enqueue but with no visible holder.



Workaround:
  Restart either the shadow instance experiencing 
  the problem or the master instance will solve the problem.

---

錯誤資訊:

>>> WAITED TOO LONG FOR A ROW CACHE ENQUEUE LOCK! <<<
row cache enqueue: session: 7000002482f7ab8, mode: N, request: S

PROCESS 30:
...
SO: 700000223372e00, type: 50, owner: 7000002493480a8, flag: INIT/-/-/0x00
row cache enqueue: count=1 session=7000002472a24a0 object=70000024c118568, request=S
savepoint=0x6
row cache parent object: address=70000024c118568 cid=16(dc_histogram_defs)
hash=3ddab3f3 typ=11 transaction=0 flags=00000000
own=70000024c118638[70000024c118638,70000024c118638] wat=70000024c118648[700000223372e30,700000223372e30] mode=N
status=-/-/-/-/-/-/-/-/-
request=N release=FALSE flags=4
instance lock id=QQ 8ede389f decf010e
set=0, complete=FALSE
set=1, complete=FALSE
...
==> This points in the direction of bug 4872999, this bug is fixed in 10.2.0.3.