database 频繁hang
本帖最后由 renjixinchina 于 2014-9-17 16:45 编辑系统:AIX 6.1
数据库:10.2.0.4 RAC
故障描述:
数据库在数据库频繁出现hang,杀掉进程后正常
第一次出怀疑是序列SEQ_SYS没有cache的问题,调整cache后依然出现
Top 5 Timed Events
Event Waits Time(s) Avg Wait(ms) % Total Call Time Wait Class
enq: SQ - contention 18,877 55,186 2,923 23.4 Configuration
cursor: pin S wait on X 5,414,192 52,972 10 22.4 Concurrency
latch: row cache objects 470,631 48,409 103 20.5 Concurrency
row cache lock 13,778 34,722 2,520 14.7 Concurrency
latch: library cache 136,207 32,988 242 14.0 Concurrency 10.2.0.4.0+RAC
Hard parses: 6.72/s
Parses: 924.54/s
sql execute elapsed time 97,800.89 41.40
sequence load elapsed time 86,099.64 36.45
hard parse elapsed time 84,005.89 35.56
failed parse elapsed time 10,212.20 4.32
部分出现 failed parse 解析失败
Pool Name Begin MB End MB % Diff
java free memory 106.16 106.16 0.00
java joxlod exec hp 5.65 5.65 0.00
large PX msg pool 1.03 1.03 0.00
large free memory 14.97 14.97 0.00
shared CCursor 1,394.47 1,337.92 -4.06
shared KQR L PO 200.66 -100.00
shared PCursor 945.13 927.50 -1.87
shared free memory 2,229.16 2,576.62 15.59
shared library cache 502.79 487.84 -2.97
shared sql area 10,318.58 10,260.79 -0.56
buffer_cache 13,856.00 13,856.00 0.00
fixed_sga 2.04 2.04 0.00
log_buffer 13.95 13.95 0.00
shared free memory 2,229.16 2,576.62 15.59
shared pool的free memory大幅增长过，shared pool的其他组件都有增长过， 是否可能有人 手动执行 alter system flush shared_pool?
由于该命令不会记录在alert.log 里 所以较难确定。
没有执行过alter system flush shared_pool alter system flush shared_pool
页:
[1]