Timeline for Oracle 11g delete by unique key slow
Current License: CC BY-SA 3.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 3, 2013 at 0:32 | vote | accept | Philip Tenn | ||
| May 2, 2013 at 18:30 | answer | added | Njal | timeline score: 2 | |
| Apr 30, 2013 at 4:03 | answer | added | Gary | timeline score: 2 | |
| Apr 29, 2013 at 8:30 | comment | added | Colin 't Hart | While Oracle does row-level locking, there is a limit to how many rows can be locked per block. If you have a lot of concurrent updates resulting in contention on a particular block, this could be causing your problem. | |
| Apr 28, 2013 at 15:30 | comment | added | Chris Saxon | Are there any updates taking place on this table? If so, is it possible that the updates are blocking the delete? Also, how large is the table? | |
| Apr 28, 2013 at 14:06 | history | tweeted | twitter.com/#!/StackDBAs/status/328510469826560001 | ||
| Apr 28, 2013 at 13:04 | history | migrated | from stackoverflow.com (revisions) | ||
| Apr 28, 2013 at 9:30 | comment | added | user1822 |
Is it possible, that the insert was trying to insert the same (unique) ID as the delete was trying to delete? Did you check e.g. dba_waiters during that time? Or v$lock$ or v$session_wait or V$SESSION_BLOCKERS
|
|
| Apr 28, 2013 at 8:30 | answer | added | haki | timeline score: 0 | |
| Apr 28, 2013 at 0:58 | answer | added | Lokesh | timeline score: 2 | |
| Apr 28, 2013 at 0:32 | history | asked | Philip Tenn | CC BY-SA 3.0 |