| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
Home -> Community -> Usenet -> c.d.o.server -> Oracle checkpoint process suddenly using lots of CPU
Hi All,
Just a quick question! We've been running Oracle 8.0.5 on Solaris for some time without any major problems. Yesterday the Oracle Checkpoint daemon (ora_ckpt_<instance name>) seems to have changed its behaviour dramatically. Currently its using up to 90% of the CPU (per top) on our Enterprise 450. e.g.
PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU COMMAND
372 oracle 14 -25 0 39M 20M run 22.4H 85.90% oracle
20251 robb 1 33 0 2168K 1392K cpu 2:22 0.20% top
483 oracle 11 29 0 39M 24M sleep 163:27 0.17% oracle
Or, from ps -ef:
UID PID PPID C STIME TTY TIME CMD
oracle 374 1 0 May 10 ? 0:04 ora_smon_RS2
oracle 432 1 0 May 10 ? 0:18
/opt/oracle/product/8.0.5/bin/tnslsnr LISTENER -inherit
oracle 362 1 0 May 10 ? 0:00 ora_pmon_RS2
oracle 365 1 0 May 10 ? 7:36 ora_dbw0_RS2
oracle 367 1 0 May 10 ? 0:50 ora_arch_RS2
oracle 369 1 0 May 10 ? 7:04 ora_lgwr_RS2
oracle 372 1 85 May 10 ? 1346:44 ora_ckpt_RS2
oracle 376 1 0 May 10 ? 0:03 ora_reco_RS2
oracle 483 1 0 May 10 ? 163:28 oracleRS2 (LOCAL=NO)
The Oracle server doesn't seem to be any busier than usual i.e. the client workload hasn't changed as far as I know. If I truss the process I don't see any obvious errors, mostly it seems to be calling pwrite() and some LWP related stuff. The alert log contains no unusual messages.
Any ideas why this would happen? Or better still, how to fix it?
--
Yours,
Robb.
+-----------------------------------------------------------+
| "I've seen things you people wouldn't believe. |
| Attack ships on fire off the shoulder of Orion. I watched |
| C-Beams glitter in the dark near the Tannhauser Gate. All |
| those moments will be lost in time, like tears in rain. |
| Time to die." -- Roy Batty, Nexus6, N6MAA10816, Combat |
+-----------------------------------------------------------+
Received on Thu May 27 1999 - 04:45:21 CDT
![]() |
![]() |