On 30-Aug-05, at 12:15, Tom Lane wrote:
=?ISO-8859-1?Q?R=E9my_Beaumont?= <[EMAIL PROTECTED]> writes:
The stats of the NetApp do confirm that it is sitting idle.
Really?
CPU NFS CIFS HTTP Total Net kB/s Disk kB/s Tape kB/s
Cache Cache CP CP Disk DAFS FCP iSCSI FCP kB/s
in out read write read write
age hit time ty util in out
2% 0 0 0 139 0 0 2788 0 0 0
3 96% 0% - 15% 0 139 0 3 2277
2% 0 0 0 144 0 0 2504 0 0 0
3 96% 0% - 18% 0 144 0 3 2150
2% 0 0 0 130 0 0 2212 0 0 0
3 96% 0% - 13% 0 130 0 3 1879
3% 0 0 0 169 0 0 2937 80 0 0
3 96% 0% - 13% 0 169 0 4 2718
2% 0 0 0 139 0 0 2448 0 0 0
3 96% 0% - 12% 0 139 0 3 2096
I know zip about NetApps, but doesn't the 8th column indicate pretty
steady disk reads?
Yes, but they are very low.
At 15% usage, it's pretty much sitting idle if you consider that the OS
reports that one of the processor is spending more then 80% of it's
time in IOwait.
Rémy
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
