Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.misc -> Re: Oracle 9i on Solaris 8
David Kurz wrote:
> Hi,
>
> here is a typical frame from 'top'. What's strange are the high values
> for 'SIZE' and 'RES' for the oracle processes. Is this normal?
>
> I mean, the system has 4Gigs memory - and each oracle process consumes
> 1,3Gigs!
>
>
> load averages: 1.69, 1.68, 1.73 12:54:08
> 513 processes: 511 sleeping, 1 zombie, 1 on cpu
>
> Memory: 4096M real, 67M free, 4845M swap in use, 2201M swap free
>
>
> PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU COMMAND
> 20331 johndoe 1 31 0 28M 11M run 294:15 21.70% myapp1
> 20156 johndoe 1 22 0 13M 7280K run 168:25 12.94% myapp1
> 17877 johndoe 1 59 0 12M 7184K sleep 204:40 12.10% myapp2
> 20340 johndoe 14 59 0 1367M 1345M sleep 135:31 10.20% oracle
> 15796 johndoe 1 59 0 12M 8072K sleep 52:11 5.48% myapp3
> 20162 johndoe 4 59 0 1363M 1341M sleep 49:38 3.74% oracle
> 16625 johndoe 1 59 0 12M 7712K sleep 11:31 2.46% myapp3
> 8899 oracle 14 59 0 1365M 1331M sleep 172:21 2.35% oracle
> 8891 oracle 198 59 0 1368M 1333M sleep 69:46 1.27% oracle
> 10977 oracle 33 59 0 1365M 1332M sleep 41:16 1.11% oracle
> 8895 oracle 124 59 0 1367M 1332M sleep 44:09 1.02% oracle
> 18437 johndoe 1 59 0 11M 6680K sleep 15:14 0.54% myapp2
> 15800 johndoe 4 59 0 1369M 1347M sleep 2:23 0.34% oracle
> 17887 johndoe 4 59 0 1363M 1341M sleep 4:17 0.32% oracle
> 29430 johndoe 1 59 0 3056K 1672K sleep 2:07 0.27% top
> 8903 oracle 15 59 0 1365M 1332M sleep 33:17 0.23% oracle
> 561 root 8 59 0 13M 9624K sleep 141:29 0.12% jre
> 16658 johndoe 4 59 0 1364M 1341M sleep 0:21 0.11% oracle
> 21805 johndoe 1 37 4 2216K 1288K cpu2 0:00 0.09% top
> 15616 johndoe 4 59 0 1363M 1341M sleep 1:02 0.09% oracle
> 17398 johndoe 4 59 0 1367M 1344M sleep 0:31 0.09% oracle
> 279 root 11 59 0 5360K 2056K sleep 19:12 0.05% picld
> 20793 oracle 4 59 0 1363M 1342M sleep 0:00 0.05% oracle
> 18979 johndoe 22 29 10 143M 26M sleep 1:15 0.04% java
> 664 root 11 59 0 14M 7504K sleep 36:11 0.03% jre
>
>
> kind regards,
> Dave
Look at your "swap in use" and "swap free" values. Physical memory and virtual memory are two different things.
Rather than asking, "is this normal" (which you can determine for yourself by simply taking repeated samples over some period of time), you should be asking, "is this a problem".
I'd be more concerned about process 20331 if I were you...
-- Mark Bole http://www.bincomputing.comReceived on Wed Feb 09 2005 - 20:36:08 CST