这是下午正式环境不可用,数据库报告如下,等待事件是row cache lock.
Snap Id | Snap Time | Sessions | Cursors/Session | |
---|---|---|---|---|
Begin Snap: | 27865 | 05-11月-12 13:59:28 | 79 | 15.9 |
End Snap: | 27869 | 05-11月-12 18:00:00 | 85 | 12.3 |
Elapsed: | 240.54 (mins) | |||
DB Time: | 4,459.16 (mins) |
Event |
Waits | Time(s) | Avg Wait(ms) | % Total Call Time | Wait Class |
---|---|---|---|---|---|
row cache lock | 39,921 | 109,476 | 2,742 | 40.9 | Concurrency |
latch: row cache objects | 516,535 | 103,579 | 201 | 38.7 | Concurrency |
cursor: pin S wait on X | 3,081,048 | 30,252 | 10 | 11.3 | Concurrency |
enq: TX - row lock contention | 20,048 | 9,766 | 487 | 3.7 | Application |
rdbms ipc reply | 4,102 | 6,169 | 1,504 | 2.3 | Other |
第一步,如果序列是no cache,会发生这种问题。检查user_sequences表的cache_size字段,都是20,那排除这种情况。
第二步,检查数据库回收站recyclebin,发现有接近8000张表,清除,后续继续跟踪是否会有问题。