| RAC: "global cache cr request" events take 1 second ocasionally [message #139359] |
Tue, 27 September 2005 11:29  |
ollibrandt
Messages: 1
Registered: September 2005
|
Junior Member |
|
|
Hi All,
we have a problem with RAC and the cluster interconnect on Sun hardware (Ora9.2.0.4 + Solaris8). Once an hour, a SQL statement which usually is processed within 100 milliseonds takes 5-10 seconds to finish.
In the SQL trace we see the following wait events (see below).
The sum up to 8 seconds for the single FETCH. We tried
to change several Solaris udp parameters for optimitzation.
but nothing helps. With netstat,ping,.. no latency problem can be seen on the interconnect. Has anyone seen this before
and can help ??
WAIT #1: nam='global cache cr request' ela= 242 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='global cache cr request' ela= 1222975 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='global cache cr request' ela= 1230396 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='global cache cr request' ela= 1230344 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='global cache cr request' ela= 1230297 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='global cache cr request' ela= 1220738 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='global cache cr request' ela= 1230326 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='global cache cr request' ela= 882920 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='cr request retry' ela= 4 p1=10 p2=3697 p3=0
WAIT #1: nam='global cache cr request' ela= 190 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='global cache cr request' ela= 316 p1=10 p2=3697 p3=4402878193440
WAIT #1: nam='SQL*Net message to client' ela= 4 p1=1229996800 p2=1 p3=0
FETCH #1:c=0,e=8252388,p=0,cr=11,cu=0,mis=0,r=2,dep=0,og=4,tim=186006013477
|
|
|
|
|
|
] 
Blog
Search
Help
Members
Register
Login
Home
