系统平台为
Oracle Solaris on SPARC (64-bit)
数据库版本
11.2.0.3
遭遇这个错误主要是因为备份恢复出问题了,执行介质恢复一下吃光了而且系统告警日志报错如下:
/opt/app/oracle/diag/rdbms/standbydbrtps/RTPS/trace/RTPS_psp0_6332.trc: ORA-27300: OS system dependent operation:fork failed with status: 12 ORA-27301: OS failure message: Not enough space ORA-27302: failure occurred at: skgpspawn3
在系统的/var/adm/messages中同时也收到信息如下:
shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry,
经过诊断与Oracle bug 18387547吻合.该bug信息如下:
Hdr: 18387547 11.2.0.3 RDBMS 11.2.0.3 SECURITY PRODID-5 PORTID-23 Abstract: MEMORY LEAK OF RECOVERY PROCESS IF TDE IS ENABLED *** 03/12/14 01:45 am *** PROBLEM: -------- Customer uses TDE and physical standby. The memory will be eaten by the recovery process (manual or managed). DIAGNOSTIC ANALYSIS: -------------------- When media recovery starts, process that do media recovery (ora_mrp if recovery is run without parallelism or ora_pr if it is parallel recovery), is allocating 96 bytes buffers in order to do ioctl() for /dev/crypto, but the memory is not beeing freed (dtrace shows only malloc() calls, and no free() calls). Each ioctl() call will be issued with new 96 buffer, and that will deplete memory on server, causing server to stop once all virtual memory is used. In /var/adm/messages we can see the errors: Feb 9 18:19:58 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16090 (oracle_server_mo) Feb 9 18:21:43 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16097 (oracle_server_mo) Feb 9 18:23:33 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16103 (oracle_server_mo) Feb 9 18:25:19 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16106 (oracle_server_mo) Feb 9 18:27:05 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16110 (oracle_server_mo) Feb 9 18:28:54 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16117 (oracle_server_mo) Feb 9 18:30:42 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16129 (oracle_server_mo) Feb 9 18:32:31 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16138 (oracle_server_mo) Feb 9 18:34:17 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16148 (oracle_server_mo) Feb 9 18:36:03 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16151 (oracle_server_mo) Feb 9 18:37:50 shadowdog-FO genunix: [ID 470503 kern.warning] WARNING: Sorry, no swap space to grow stack for pid 16156 (oracle_server_mo) In alert log: Process startup failed, error stack: Errors in file /u01/app/oracle/diag/rdbms/standbydbrtps/RTPS/trace/RTPS_psp0_6631.trc: ORA-27300: OS system dependent operation:fork failed with status: 12 ORA-27301: OS failure message: Not enough space ORA-27302: failure occurred at: skgpspawn3 WORKAROUND: ----------- Restart MRP RELATED BUGS: ------------- REPRODUCIBILITY: ---------------- TEST CASE: ---------- STACK TRACE: ------------ SUPPORTING INFORMATION: ----------------------- 24 HOUR CONTACT INFORMATION FOR P1 BUGS: ---------------------------------------- DIAL-IN INFORMATION: -------------------- IMPACT DATE: ------------ *** 03/12/14 01:46 am *** *** 03/12/14 01:47 am *** *** 03/12/14 01:48 am *** *** 03/27/14 04:40 am *** *** 03/27/14 05:30 am *** (CHG: Sta->10) *** 03/27/14 05:30 am *** *** 04/01/14 01:27 am *** (CHG: Sta->16) *** 04/01/14 01:27 am *** *** 04/01/14 03:28 am *** *** 04/01/14 04:54 am *** (CHG: Sta->10) *** 04/01/14 04:54 am *** *** 04/01/14 05:02 am *** *** 04/15/14 11:25 pm *** (CHG: Sta->16) *** 04/15/14 11:25 pm *** *** 04/20/14 12:59 am *** (CHG: Sta->91) *** 04/20/14 01:00 am *** *** 04/20/14 01:00 am *** (ADD: Impact/Symptom->LEAK - DB MEMORY/OTH...)